Skip to content

Instantly share code, notes, and snippets.

@insop

insop/resnet_18_vta.ipynb

Created October 6, 2020 09:09
Show Gist options
  • Save insop/18aaab92d48061c6c0e440be0bd83733 to your computer and use it in GitHub Desktop.
Save insop/18aaab92d48061c6c0e440be0bd83733 to your computer and use it in GitHub Desktop.
Download ZIP
Deploy Pretrained Vision Model from Pytorch on VTA
Display the source blob
Display the rendered blob
Raw
resnet_18_vta.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Deploy Pretrained Vision Model from Pytorch on VTA"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Load pytorch model to tvm \n",
"- [tvm reference](https://tvm.apache.org/docs/tutorials/frontend/from_pytorch.html#sphx-glr-tutorials-frontend-from-pytorch-py)"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"\n",
"from tvm.contrib.download import download_testdata\n",
"\n",
"# PyTorch imports\n",
"import torch\n",
"import torchvision\n",
"\n",
"\n",
"import tvm\n",
"from tvm import te\n",
"from tvm import rpc, autotvm, relay\n",
"from tvm.contrib import graph_runtime, download\n",
"from tvm.contrib.debugger import debug_runtime\n",
"from tvm.relay import transform\n",
"from tvm import relay\n",
"\n",
"import vta\n",
"from vta.testing import simulator\n",
"from vta.top import graph_pack\n",
"\n",
"# Make sure that TVM was compiled with RPC=1\n",
"assert tvm.runtime.enabled(\"rpc\")"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"# transformer modules\n",
"\n",
"import transformer_simple\n",
"import classifier\n",
"import util"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"model_name = 'resnet18'\n",
"model = getattr(torchvision.models, model_name)(pretrained=True)\n",
"model = model.eval()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"input_shape = [1, 3, 224, 224]\n",
"input_data = torch.randn(input_shape)\n",
"scripted_model = torch.jit.trace(model, input_data).eval()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Import the graph to Relay"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"input_name = \"input0\"\n",
"shape_list = [(input_name, input_shape)]\n",
"mod, params = relay.frontend.from_pytorch(scripted_model, shape_list)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"type Option[A] {\n",
" Some(A),\n",
" None,\n",
"}\n",
"\n",
"type Tree[A] {\n",
" Rose(A, List[Tree[A]]),\n",
"}\n",
"\n",
"type tensor_float64_t {\n",
" tensor_nil_float64,\n",
" tensor0_float64(float64),\n",
" tensor1_float64(Tensor[(?), float64]),\n",
" tensor2_float64(Tensor[(?, ?), float64]),\n",
" tensor3_float64(Tensor[(?, ?, ?), float64]),\n",
" tensor4_float64(Tensor[(?, ?, ?, ?), float64]),\n",
" tensor5_float64(Tensor[(?, ?, ?, ?, ?), float64]),\n",
" tensor6_float64(Tensor[(?, ?, ?, ?, ?, ?), float64]),\n",
"}\n",
"\n",
"type tensor_uint16_t {\n",
" tensor_nil_uint16,\n",
" tensor0_uint16(uint16),\n",
" tensor1_uint16(Tensor[(?), uint16]),\n",
" tensor2_uint16(Tensor[(?, ?), uint16]),\n",
" tensor3_uint16(Tensor[(?, ?, ?), uint16]),\n",
" tensor4_uint16(Tensor[(?, ?, ?, ?), uint16]),\n",
" tensor5_uint16(Tensor[(?, ?, ?, ?, ?), uint16]),\n",
" tensor6_uint16(Tensor[(?, ?, ?, ?, ?, ?), uint16]),\n",
"}\n",
"\n",
"type tensor_int16_t {\n",
" tensor_nil_int16,\n",
" tensor0_int16(int16),\n",
" tensor1_int16(Tensor[(?), int16]),\n",
" tensor2_int16(Tensor[(?, ?), int16]),\n",
" tensor3_int16(Tensor[(?, ?, ?), int16]),\n",
" tensor4_int16(Tensor[(?, ?, ?, ?), int16]),\n",
" tensor5_int16(Tensor[(?, ?, ?, ?, ?), int16]),\n",
" tensor6_int16(Tensor[(?, ?, ?, ?, ?, ?), int16]),\n",
"}\n",
"\n",
"type List[A] {\n",
" Cons(A, List[A]),\n",
" Nil,\n",
"}\n",
"\n",
"type tensor_float16_t {\n",
" tensor_nil_float16,\n",
" tensor0_float16(float16),\n",
" tensor1_float16(Tensor[(?), float16]),\n",
" tensor2_float16(Tensor[(?, ?), float16]),\n",
" tensor3_float16(Tensor[(?, ?, ?), float16]),\n",
" tensor4_float16(Tensor[(?, ?, ?, ?), float16]),\n",
" tensor5_float16(Tensor[(?, ?, ?, ?, ?), float16]),\n",
" tensor6_float16(Tensor[(?, ?, ?, ?, ?, ?), float16]),\n",
"}\n",
"\n",
"type tensor_uint8_t {\n",
" tensor_nil_uint8,\n",
" tensor0_uint8(uint8),\n",
" tensor1_uint8(Tensor[(?), uint8]),\n",
" tensor2_uint8(Tensor[(?, ?), uint8]),\n",
" tensor3_uint8(Tensor[(?, ?, ?), uint8]),\n",
" tensor4_uint8(Tensor[(?, ?, ?, ?), uint8]),\n",
" tensor5_uint8(Tensor[(?, ?, ?, ?, ?), uint8]),\n",
" tensor6_uint8(Tensor[(?, ?, ?, ?, ?, ?), uint8]),\n",
"}\n",
"\n",
"type tensor_float32_t {\n",
" tensor_nil_float32,\n",
" tensor0_float32(float32),\n",
" tensor1_float32(Tensor[(?), float32]),\n",
" tensor2_float32(Tensor[(?, ?), float32]),\n",
" tensor3_float32(Tensor[(?, ?, ?), float32]),\n",
" tensor4_float32(Tensor[(?, ?, ?, ?), float32]),\n",
" tensor5_float32(Tensor[(?, ?, ?, ?, ?), float32]),\n",
" tensor6_float32(Tensor[(?, ?, ?, ?, ?, ?), float32]),\n",
"}\n",
"\n",
"type tensor_int8_t {\n",
" tensor_nil_int8,\n",
" tensor0_int8(int8),\n",
" tensor1_int8(Tensor[(?), int8]),\n",
" tensor2_int8(Tensor[(?, ?), int8]),\n",
" tensor3_int8(Tensor[(?, ?, ?), int8]),\n",
" tensor4_int8(Tensor[(?, ?, ?, ?), int8]),\n",
" tensor5_int8(Tensor[(?, ?, ?, ?, ?), int8]),\n",
" tensor6_int8(Tensor[(?, ?, ?, ?, ?, ?), int8]),\n",
"}\n",
"\n",
"type tensor_int32_t {\n",
" tensor_nil_int32,\n",
" tensor0_int32(int32),\n",
" tensor1_int32(Tensor[(?), int32]),\n",
" tensor2_int32(Tensor[(?, ?), int32]),\n",
" tensor3_int32(Tensor[(?, ?, ?), int32]),\n",
" tensor4_int32(Tensor[(?, ?, ?, ?), int32]),\n",
" tensor5_int32(Tensor[(?, ?, ?, ?, ?), int32]),\n",
" tensor6_int32(Tensor[(?, ?, ?, ?, ?, ?), int32]),\n",
"}\n",
"\n",
"type tensor_int64_t {\n",
" tensor_nil_int64,\n",
" tensor0_int64(int64),\n",
" tensor1_int64(Tensor[(?), int64]),\n",
" tensor2_int64(Tensor[(?, ?), int64]),\n",
" tensor3_int64(Tensor[(?, ?, ?), int64]),\n",
" tensor4_int64(Tensor[(?, ?, ?, ?), int64]),\n",
" tensor5_int64(Tensor[(?, ?, ?, ?, ?), int64]),\n",
" tensor6_int64(Tensor[(?, ?, ?, ?, ?, ?), int64]),\n",
"}\n",
"\n",
"def @main(%input0: Tensor[(1, 3, 224, 224), float32], %conv1.weight: Tensor[(64, 3, 7, 7), float32], %bn1.weight: Tensor[(64), float32], %bn1.bias: Tensor[(64), float32], %bn1.running_mean: Tensor[(64), float32], %bn1.running_var: Tensor[(64), float32], %layer1.0.conv1.weight: Tensor[(64, 64, 3, 3), float32], %layer1.0.bn1.weight: Tensor[(64), float32], %layer1.0.bn1.bias: Tensor[(64), float32], %layer1.0.bn1.running_mean: Tensor[(64), float32], %layer1.0.bn1.running_var: Tensor[(64), float32], %layer1.0.conv2.weight: Tensor[(64, 64, 3, 3), float32], %layer1.0.bn2.weight: Tensor[(64), float32], %layer1.0.bn2.bias: Tensor[(64), float32], %layer1.0.bn2.running_mean: Tensor[(64), float32], %layer1.0.bn2.running_var: Tensor[(64), float32], %layer1.1.conv1.weight: Tensor[(64, 64, 3, 3), float32], %layer1.1.bn1.weight: Tensor[(64), float32], %layer1.1.bn1.bias: Tensor[(64), float32], %layer1.1.bn1.running_mean: Tensor[(64), float32], %layer1.1.bn1.running_var: Tensor[(64), float32], %layer1.1.conv2.weight: Tensor[(64, 64, 3, 3), float32], %layer1.1.bn2.weight: Tensor[(64), float32], %layer1.1.bn2.bias: Tensor[(64), float32], %layer1.1.bn2.running_mean: Tensor[(64), float32], %layer1.1.bn2.running_var: Tensor[(64), float32], %layer2.0.conv1.weight: Tensor[(128, 64, 3, 3), float32], %layer2.0.bn1.weight: Tensor[(128), float32], %layer2.0.bn1.bias: Tensor[(128), float32], %layer2.0.bn1.running_mean: Tensor[(128), float32], %layer2.0.bn1.running_var: Tensor[(128), float32], %layer2.0.conv2.weight: Tensor[(128, 128, 3, 3), float32], %layer2.0.bn2.weight: Tensor[(128), float32], %layer2.0.bn2.bias: Tensor[(128), float32], %layer2.0.bn2.running_mean: Tensor[(128), float32], %layer2.0.bn2.running_var: Tensor[(128), float32], %layer2.0.downsample.0.weight: Tensor[(128, 64, 1, 1), float32], %layer2.0.downsample.1.weight: Tensor[(128), float32], %layer2.0.downsample.1.bias: Tensor[(128), float32], %layer2.0.downsample.1.running_mean: Tensor[(128), float32], %layer2.0.downsample.1.running_var: Tensor[(128), float32], %layer2.1.conv1.weight: Tensor[(128, 128, 3, 3), float32], %layer2.1.bn1.weight: Tensor[(128), float32], %layer2.1.bn1.bias: Tensor[(128), float32], %layer2.1.bn1.running_mean: Tensor[(128), float32], %layer2.1.bn1.running_var: Tensor[(128), float32], %layer2.1.conv2.weight: Tensor[(128, 128, 3, 3), float32], %layer2.1.bn2.weight: Tensor[(128), float32], %layer2.1.bn2.bias: Tensor[(128), float32], %layer2.1.bn2.running_mean: Tensor[(128), float32], %layer2.1.bn2.running_var: Tensor[(128), float32], %layer3.0.conv1.weight: Tensor[(256, 128, 3, 3), float32], %layer3.0.bn1.weight: Tensor[(256), float32], %layer3.0.bn1.bias: Tensor[(256), float32], %layer3.0.bn1.running_mean: Tensor[(256), float32], %layer3.0.bn1.running_var: Tensor[(256), float32], %layer3.0.conv2.weight: Tensor[(256, 256, 3, 3), float32], %layer3.0.bn2.weight: Tensor[(256), float32], %layer3.0.bn2.bias: Tensor[(256), float32], %layer3.0.bn2.running_mean: Tensor[(256), float32], %layer3.0.bn2.running_var: Tensor[(256), float32], %layer3.0.downsample.0.weight: Tensor[(256, 128, 1, 1), float32], %layer3.0.downsample.1.weight: Tensor[(256), float32], %layer3.0.downsample.1.bias: Tensor[(256), float32], %layer3.0.downsample.1.running_mean: Tensor[(256), float32], %layer3.0.downsample.1.running_var: Tensor[(256), float32], %layer3.1.conv1.weight: Tensor[(256, 256, 3, 3), float32], %layer3.1.bn1.weight: Tensor[(256), float32], %layer3.1.bn1.bias: Tensor[(256), float32], %layer3.1.bn1.running_mean: Tensor[(256), float32], %layer3.1.bn1.running_var: Tensor[(256), float32], %layer3.1.conv2.weight: Tensor[(256, 256, 3, 3), float32], %layer3.1.bn2.weight: Tensor[(256), float32], %layer3.1.bn2.bias: Tensor[(256), float32], %layer3.1.bn2.running_mean: Tensor[(256), float32], %layer3.1.bn2.running_var: Tensor[(256), float32], %layer4.0.conv1.weight: Tensor[(512, 256, 3, 3), float32], %layer4.0.bn1.weight: Tensor[(512), float32], %layer4.0.bn1.bias: Tensor[(512), float32], %layer4.0.bn1.running_mean: Tensor[(512), float32], %layer4.0.bn1.running_var: Tensor[(512), float32], %layer4.0.conv2.weight: Tensor[(512, 512, 3, 3), float32], %layer4.0.bn2.weight: Tensor[(512), float32], %layer4.0.bn2.bias: Tensor[(512), float32], %layer4.0.bn2.running_mean: Tensor[(512), float32], %layer4.0.bn2.running_var: Tensor[(512), float32], %layer4.0.downsample.0.weight: Tensor[(512, 256, 1, 1), float32], %layer4.0.downsample.1.weight: Tensor[(512), float32], %layer4.0.downsample.1.bias: Tensor[(512), float32], %layer4.0.downsample.1.running_mean: Tensor[(512), float32], %layer4.0.downsample.1.running_var: Tensor[(512), float32], %layer4.1.conv1.weight: Tensor[(512, 512, 3, 3), float32], %layer4.1.bn1.weight: Tensor[(512), float32], %layer4.1.bn1.bias: Tensor[(512), float32], %layer4.1.bn1.running_mean: Tensor[(512), float32], %layer4.1.bn1.running_var: Tensor[(512), float32], %layer4.1.conv2.weight: Tensor[(512, 512, 3, 3), float32], %layer4.1.bn2.weight: Tensor[(512), float32], %layer4.1.bn2.bias: Tensor[(512), float32], %layer4.1.bn2.running_mean: Tensor[(512), float32], %layer4.1.bn2.running_var: Tensor[(512), float32], %fc.weight: Tensor[(1000, 512), float32], %fc.bias: Tensor[(1000), float32]) -> Tensor[(1, 1000), float32] {\n",
" %0 = nn.conv2d(%input0, %conv1.weight, strides=[2, 2], padding=[3, 3, 3, 3], channels=64, kernel_size=[7, 7]) /* ty=Tensor[(1, 64, 112, 112), float32] */;\n",
" %1 = nn.batch_norm(%0, %bn1.weight, %bn1.bias, %bn1.running_mean, %bn1.running_var) /* ty=(Tensor[(1, 64, 112, 112), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %2 = %1.0;\n",
" %3 = nn.relu(%2) /* ty=Tensor[(1, 64, 112, 112), float32] */;\n",
" %4 = nn.max_pool2d(%3, pool_size=[3, 3], strides=[2, 2], padding=[1, 1, 1, 1]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %5 = nn.conv2d(%4, %layer1.0.conv1.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %6 = nn.batch_norm(%5, %layer1.0.bn1.weight, %layer1.0.bn1.bias, %layer1.0.bn1.running_mean, %layer1.0.bn1.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %7 = %6.0;\n",
" %8 = nn.relu(%7) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %9 = nn.conv2d(%8, %layer1.0.conv2.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %10 = nn.batch_norm(%9, %layer1.0.bn2.weight, %layer1.0.bn2.bias, %layer1.0.bn2.running_mean, %layer1.0.bn2.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %11 = %10.0;\n",
" %12 = add(%11, %4) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %13 = nn.relu(%12) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %14 = nn.conv2d(%13, %layer1.1.conv1.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %15 = nn.batch_norm(%14, %layer1.1.bn1.weight, %layer1.1.bn1.bias, %layer1.1.bn1.running_mean, %layer1.1.bn1.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %16 = %15.0;\n",
" %17 = nn.relu(%16) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %18 = nn.conv2d(%17, %layer1.1.conv2.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %19 = nn.batch_norm(%18, %layer1.1.bn2.weight, %layer1.1.bn2.bias, %layer1.1.bn2.running_mean, %layer1.1.bn2.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %20 = %19.0;\n",
" %21 = add(%20, %13) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %22 = nn.relu(%21) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %23 = nn.conv2d(%22, %layer2.0.conv1.weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %24 = nn.batch_norm(%23, %layer2.0.bn1.weight, %layer2.0.bn1.bias, %layer2.0.bn1.running_mean, %layer2.0.bn1.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %25 = %24.0;\n",
" %26 = nn.relu(%25) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %27 = nn.conv2d(%26, %layer2.0.conv2.weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %28 = nn.batch_norm(%27, %layer2.0.bn2.weight, %layer2.0.bn2.bias, %layer2.0.bn2.running_mean, %layer2.0.bn2.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %29 = %28.0;\n",
" %30 = nn.conv2d(%22, %layer2.0.downsample.0.weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=128, kernel_size=[1, 1]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %31 = nn.batch_norm(%30, %layer2.0.downsample.1.weight, %layer2.0.downsample.1.bias, %layer2.0.downsample.1.running_mean, %layer2.0.downsample.1.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %32 = %31.0;\n",
" %33 = add(%29, %32) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %34 = nn.relu(%33) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %35 = nn.conv2d(%34, %layer2.1.conv1.weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %36 = nn.batch_norm(%35, %layer2.1.bn1.weight, %layer2.1.bn1.bias, %layer2.1.bn1.running_mean, %layer2.1.bn1.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %37 = %36.0;\n",
" %38 = nn.relu(%37) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %39 = nn.conv2d(%38, %layer2.1.conv2.weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %40 = nn.batch_norm(%39, %layer2.1.bn2.weight, %layer2.1.bn2.bias, %layer2.1.bn2.running_mean, %layer2.1.bn2.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %41 = %40.0;\n",
" %42 = add(%41, %34) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %43 = nn.relu(%42) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %44 = nn.conv2d(%43, %layer3.0.conv1.weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %45 = nn.batch_norm(%44, %layer3.0.bn1.weight, %layer3.0.bn1.bias, %layer3.0.bn1.running_mean, %layer3.0.bn1.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %46 = %45.0;\n",
" %47 = nn.relu(%46) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %48 = nn.conv2d(%47, %layer3.0.conv2.weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %49 = nn.batch_norm(%48, %layer3.0.bn2.weight, %layer3.0.bn2.bias, %layer3.0.bn2.running_mean, %layer3.0.bn2.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %50 = %49.0;\n",
" %51 = nn.conv2d(%43, %layer3.0.downsample.0.weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=256, kernel_size=[1, 1]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %52 = nn.batch_norm(%51, %layer3.0.downsample.1.weight, %layer3.0.downsample.1.bias, %layer3.0.downsample.1.running_mean, %layer3.0.downsample.1.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %53 = %52.0;\n",
" %54 = add(%50, %53) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %55 = nn.relu(%54) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %56 = nn.conv2d(%55, %layer3.1.conv1.weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %57 = nn.batch_norm(%56, %layer3.1.bn1.weight, %layer3.1.bn1.bias, %layer3.1.bn1.running_mean, %layer3.1.bn1.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %58 = %57.0;\n",
" %59 = nn.relu(%58) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %60 = nn.conv2d(%59, %layer3.1.conv2.weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %61 = nn.batch_norm(%60, %layer3.1.bn2.weight, %layer3.1.bn2.bias, %layer3.1.bn2.running_mean, %layer3.1.bn2.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %62 = %61.0;\n",
" %63 = add(%62, %55) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %64 = nn.relu(%63) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %65 = nn.conv2d(%64, %layer4.0.conv1.weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %66 = nn.batch_norm(%65, %layer4.0.bn1.weight, %layer4.0.bn1.bias, %layer4.0.bn1.running_mean, %layer4.0.bn1.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %67 = %66.0;\n",
" %68 = nn.relu(%67) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %69 = nn.conv2d(%68, %layer4.0.conv2.weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %70 = nn.batch_norm(%69, %layer4.0.bn2.weight, %layer4.0.bn2.bias, %layer4.0.bn2.running_mean, %layer4.0.bn2.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %71 = %70.0;\n",
" %72 = nn.conv2d(%64, %layer4.0.downsample.0.weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=512, kernel_size=[1, 1]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %73 = nn.batch_norm(%72, %layer4.0.downsample.1.weight, %layer4.0.downsample.1.bias, %layer4.0.downsample.1.running_mean, %layer4.0.downsample.1.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %74 = %73.0;\n",
" %75 = add(%71, %74) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %76 = nn.relu(%75) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %77 = nn.conv2d(%76, %layer4.1.conv1.weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %78 = nn.batch_norm(%77, %layer4.1.bn1.weight, %layer4.1.bn1.bias, %layer4.1.bn1.running_mean, %layer4.1.bn1.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %79 = %78.0;\n",
" %80 = nn.relu(%79) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %81 = nn.conv2d(%80, %layer4.1.conv2.weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %82 = nn.batch_norm(%81, %layer4.1.bn2.weight, %layer4.1.bn2.bias, %layer4.1.bn2.running_mean, %layer4.1.bn2.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %83 = %82.0;\n",
" %84 = add(%83, %76) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %85 = nn.relu(%84) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %86 = nn.adaptive_avg_pool2d(%85, output_size=[1, 1]) /* ty=Tensor[(1, 512, 1, 1), float32] */;\n",
" %87 = reshape(%86, newshape=[0, -1, 1, 1]) /* ty=Tensor[(1, 512, 1, 1), float32] */;\n",
" %88 = squeeze(%87, axis=[2, 3]) /* ty=Tensor[(1, 512), float32] */;\n",
" %89 = transpose(%fc.weight, axes=[1, 0]) /* ty=Tensor[(512, 1000), float32] */;\n",
" %90 = transpose(%89, axes=[1, 0]) /* ty=Tensor[(1000, 512), float32] */;\n",
" %91 = nn.dense(%88, %90, units=1000) /* ty=Tensor[(1, 1000), float32] */;\n",
" add(%91, %fc.bias) /* ty=Tensor[(1, 1000), float32] */\n",
"}\n",
"\n"
]
}
],
"source": [
"print(mod)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## VTA testing\n",
"from https://tvm.apache.org/docs/vta/tutorials/frontend/deploy_classification.html#sphx-glr-vta-tutorials-frontend-deploy-classification-py\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Loading VTA parameters\n"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"env = vta.get_env()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## define the platform and model targets"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"# Load VTA parameters from the 3rdparty/vta-hw/config/vta_config.json file\n",
"env = vta.get_env()\n",
"\n",
"# Set ``device=arm_cpu`` to run inference on the CPU\n",
"# or ``device=vta`` to run inference on the FPGA.\n",
"device = \"vta\"\n",
"target = env.target if device == \"vta\" else env.target_vta_cpu"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## FPGA programming"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"if env.TARGET not in [\"sim\", \"tsim\"]:\n",
"\n",
" # Get remote from tracker node if environment variable is set.\n",
" # To set up the tracker, you'll need to follow the \"Auto-tuning\n",
" # a convolutional network for VTA\" tutorial.\n",
" tracker_host = os.environ.get(\"TVM_TRACKER_HOST\", None)\n",
" tracker_port = os.environ.get(\"TVM_TRACKER_PORT\", None)\n",
" # Otherwise if you have a device you want to program directly from\n",
" # the host, make sure you've set the variables below to the IP of\n",
" # your board.\n",
" device_host = os.environ.get(\"VTA_RPC_HOST\", \"192.168.2.99\")\n",
" device_port = os.environ.get(\"VTA_RPC_PORT\", \"9091\")\n",
" if not tracker_host or not tracker_port:\n",
" remote = rpc.connect(device_host, int(device_port))\n",
" else:\n",
" remote = autotvm.measure.request_remote(\n",
" env.TARGET, tracker_host, int(tracker_port), timeout=10000\n",
" )\n",
"\n",
" # Reconfigure the JIT runtime and FPGA.\n",
" # You can program the FPGA with your own custom bitstream\n",
" # by passing the path to the bitstream file instead of None.\n",
" reconfig_start = time.time()\n",
" vta.reconfig_runtime(remote)\n",
" vta.program_fpga(remote, bitstream=None)\n",
" reconfig_time = time.time() - reconfig_start\n",
" print(\"Reconfigured FPGA and RPC runtime in {0:.2f}s!\".format(reconfig_time))\n",
"\n",
"# In simulation mode, host the RPC server locally.\n",
"else:\n",
" remote = rpc.LocalSession()\n",
"\n",
"# Get execution context from remote\n",
"ctx = remote.ext_dev(0) if device == \"vta\" else remote.cpu(0)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Input placeholders"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"\n",
"input_vta = tvm.te.placeholder(input_shape, name=\"input\", dtype=env.acc_dtype)\n"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"target"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## build the inference graph runtime"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"input_name = 'input0'\n",
"img_shape = (1, 3, 224, 224)\n",
"shape_list = [(input_name, img_shape)]"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/plain": [
"[('input0', (1, 3, 224, 224))]"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"shape_list"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/plain": [
"FunctionNode([Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32)), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32)), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32)), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32)), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32)), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32)), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32)), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32)), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32)), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32)), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32)), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32)), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32)), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32)), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32)), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32)), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32)), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32)), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32)), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32)), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32)), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32)), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32)), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32)), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32)), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32)), Var(layer3.1.conv1.weight, ty=TensorType([256, 256, 3, 3], float32)), Var(layer3.1.bn1.weight, ty=TensorType([256], float32)), Var(layer3.1.bn1.bias, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_var, ty=TensorType([256], float32)), Var(layer3.1.conv2.weight, ty=TensorType([256, 256, 3, 3], float32)), Var(layer3.1.bn2.weight, ty=TensorType([256], float32)), Var(layer3.1.bn2.bias, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_var, ty=TensorType([256], float32)), Var(layer4.0.conv1.weight, ty=TensorType([512, 256, 3, 3], float32)), Var(layer4.0.bn1.weight, ty=TensorType([512], float32)), Var(layer4.0.bn1.bias, ty=TensorType([512], float32)), Var(layer4.0.bn1.running_mean, ty=TensorType([512], float32)), Var(layer4.0.bn1.running_var, ty=TensorType([512], float32)), Var(layer4.0.conv2.weight, ty=TensorType([512, 512, 3, 3], float32)), Var(layer4.0.bn2.weight, ty=TensorType([512], float32)), Var(layer4.0.bn2.bias, ty=TensorType([512], float32)), Var(layer4.0.bn2.running_mean, ty=TensorType([512], float32)), Var(layer4.0.bn2.running_var, ty=TensorType([512], float32)), Var(layer4.0.downsample.0.weight, ty=TensorType([512, 256, 1, 1], float32)), Var(layer4.0.downsample.1.weight, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.bias, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.running_mean, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.running_var, ty=TensorType([512], float32)), Var(layer4.1.conv1.weight, ty=TensorType([512, 512, 3, 3], float32)), Var(layer4.1.bn1.weight, ty=TensorType([512], float32)), Var(layer4.1.bn1.bias, ty=TensorType([512], float32)), Var(layer4.1.bn1.running_mean, ty=TensorType([512], float32)), Var(layer4.1.bn1.running_var, ty=TensorType([512], float32)), Var(layer4.1.conv2.weight, ty=TensorType([512, 512, 3, 3], float32)), Var(layer4.1.bn2.weight, ty=TensorType([512], float32)), Var(layer4.1.bn2.bias, ty=TensorType([512], float32)), Var(layer4.1.bn2.running_mean, ty=TensorType([512], float32)), Var(layer4.1.bn2.running_var, ty=TensorType([512], float32)), Var(fc.weight, ty=TensorType([1000, 512], float32)), Var(fc.bias, ty=TensorType([1000], float32))], TensorType([1, 1000], float32), CallNode(Op(add), [CallNode(Op(nn.dense), [CallNode(Op(squeeze), [CallNode(Op(reshape), [CallNode(Op(nn.adaptive_avg_pool2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv1.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111245d28), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn1.weight, ty=TensorType([256], float32)), Var(layer3.1.bn1.bias, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff11127fbc8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff102f2b2b8), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn2.weight, ty=TensorType([256], float32)), Var(layer3.1.bn2.bias, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff102f7ba18), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer4.0.conv1.weight, ty=TensorType([512, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10378b1b8), [TensorType([1, 256, 14, 14], float32), TensorType([512, 256, 3, 3], float32)]), Var(layer4.0.bn1.weight, ty=TensorType([512], float32)), Var(layer4.0.bn1.bias, ty=TensorType([512], float32)), Var(layer4.0.bn1.running_mean, ty=TensorType([512], float32)), Var(layer4.0.bn1.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff102f49db8), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0)], (nullptr), [TensorType([1, 512, 7, 7], float32)]), Var(layer4.0.conv2.weight, ty=TensorType([512, 512, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b66568), [TensorType([1, 512, 7, 7], float32), TensorType([512, 512, 3, 3], float32)]), Var(layer4.0.bn2.weight, ty=TensorType([512], float32)), Var(layer4.0.bn2.bias, ty=TensorType([512], float32)), Var(layer4.0.bn2.running_mean, ty=TensorType([512], float32)), Var(layer4.0.bn2.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff104b6f308), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv1.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111245d28), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn1.weight, ty=TensorType([256], float32)), Var(layer3.1.bn1.bias, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff11127fbc8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff102f2b2b8), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn2.weight, ty=TensorType([256], float32)), Var(layer3.1.bn2.bias, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff102f7ba18), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer4.0.downsample.0.weight, ty=TensorType([512, 256, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b72c88), [TensorType([1, 256, 14, 14], float32), TensorType([512, 256, 1, 1], float32)]), Var(layer4.0.downsample.1.weight, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.bias, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.running_mean, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff104b73238), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0)], (nullptr), [TensorType([1, 512, 7, 7], float32), TensorType([1, 512, 7, 7], float32)])], (nullptr), [TensorType([1, 512, 7, 7], float32)]), Var(layer4.1.conv1.weight, ty=TensorType([512, 512, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3248), [TensorType([1, 512, 7, 7], float32), TensorType([512, 512, 3, 3], float32)]), Var(layer4.1.bn1.weight, ty=TensorType([512], float32)), Var(layer4.1.bn1.bias, ty=TensorType([512], float32)), Var(layer4.1.bn1.running_mean, ty=TensorType([512], float32)), Var(layer4.1.bn1.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb9aa8), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0)], (nullptr), [TensorType([1, 512, 7, 7], float32)]), Var(layer4.1.conv2.weight, ty=TensorType([512, 512, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1037525d8), [TensorType([1, 512, 7, 7], float32), TensorType([512, 512, 3, 3], float32)]), Var(layer4.1.bn2.weight, ty=TensorType([512], float32)), Var(layer4.1.bn2.bias, ty=TensorType([512], float32)), Var(layer4.1.bn2.running_mean, ty=TensorType([512], float32)), Var(layer4.1.bn2.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff103797628), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv1.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111245d28), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn1.weight, ty=TensorType([256], float32)), Var(layer3.1.bn1.bias, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff11127fbc8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff102f2b2b8), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn2.weight, ty=TensorType([256], float32)), Var(layer3.1.bn2.bias, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff102f7ba18), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer4.0.conv1.weight, ty=TensorType([512, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10378b1b8), [TensorType([1, 256, 14, 14], float32), TensorType([512, 256, 3, 3], float32)]), Var(layer4.0.bn1.weight, ty=TensorType([512], float32)), Var(layer4.0.bn1.bias, ty=TensorType([512], float32)), Var(layer4.0.bn1.running_mean, ty=TensorType([512], float32)), Var(layer4.0.bn1.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff102f49db8), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0)], (nullptr), [TensorType([1, 512, 7, 7], float32)]), Var(layer4.0.conv2.weight, ty=TensorType([512, 512, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b66568), [TensorType([1, 512, 7, 7], float32), TensorType([512, 512, 3, 3], float32)]), Var(layer4.0.bn2.weight, ty=TensorType([512], float32)), Var(layer4.0.bn2.bias, ty=TensorType([512], float32)), Var(layer4.0.bn2.running_mean, ty=TensorType([512], float32)), Var(layer4.0.bn2.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff104b6f308), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv1.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111245d28), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn1.weight, ty=TensorType([256], float32)), Var(layer3.1.bn1.bias, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff11127fbc8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.1.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff102f2b2b8), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.1.bn2.weight, ty=TensorType([256], float32)), Var(layer3.1.bn2.bias, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.1.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff102f7ba18), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.conv1.weight, ty=TensorType([256, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104aa59e8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 3, 3], float32)]), Var(layer3.0.bn1.weight, ty=TensorType([256], float32)), Var(layer3.0.bn1.bias, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff104ab7848), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer3.0.conv2.weight, ty=TensorType([256, 256, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b99c48), [TensorType([1, 256, 14, 14], float32), TensorType([256, 256, 3, 3], float32)]), Var(layer3.0.bn2.weight, ty=TensorType([256], float32)), Var(layer3.0.bn2.bias, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_mean, ty=TensorType([256], float32)), Var(layer3.0.bn2.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff103771c38), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv1.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104bac5d8), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn1.weight, ty=TensorType([128], float32)), Var(layer2.1.bn1.bias, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b99308), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.1.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff111291018), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.1.bn2.weight, ty=TensorType([128], float32)), Var(layer2.1.bn2.bias, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.1.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff111296768), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.conv1.weight, ty=TensorType([128, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff103748688), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 3, 3], float32)]), Var(layer2.0.bn1.weight, ty=TensorType([128], float32)), Var(layer2.0.bn1.bias, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104afe728), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer2.0.conv2.weight, ty=TensorType([128, 128, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54b48), [TensorType([1, 128, 28, 28], float32), TensorType([128, 128, 3, 3], float32)]), Var(layer2.0.bn2.weight, ty=TensorType([128], float32)), Var(layer2.0.bn2.bias, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_mean, ty=TensorType([128], float32)), Var(layer2.0.bn2.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5f2c8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0), TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107f7b118), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn1.weight, ty=TensorType([64], float32)), Var(layer1.1.bn1.bias, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f7a2e8), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.1.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff1113e8ae8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.1.bn2.weight, ty=TensorType([64], float32)), Var(layer1.1.bn2.bias, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.1.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff104b52098), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.relu), [CallNode(Op(add), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)]), Var(layer1.0.conv1.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff10795fca8), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn1.weight, ty=TensorType([64], float32)), Var(layer1.0.bn1.bias, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107f27388), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer1.0.conv2.weight, ty=TensorType([64, 64, 3, 3], float32))], relay.attrs.Conv2DAttrs(0x7ff107fb3848), [TensorType([1, 64, 56, 56], float32), TensorType([64, 64, 3, 3], float32)]), Var(layer1.0.bn2.weight, ty=TensorType([64], float32)), Var(layer1.0.bn2.bias, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_mean, ty=TensorType([64], float32)), Var(layer1.0.bn2.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff107fb4f28), [TensorType([1, 64, 56, 56], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0), CallNode(Op(nn.max_pool2d), [CallNode(Op(nn.relu), [TupleGetItemNode(CallNode(Op(nn.batch_norm), [CallNode(Op(nn.conv2d), [Var(input0, ty=TensorType([1, 3, 224, 224], float32)), Var(conv1.weight, ty=TensorType([64, 3, 7, 7], float32))], relay.attrs.Conv2DAttrs(0x7ff111207fe8), [TensorType([1, 3, 224, 224], float32), TensorType([64, 3, 7, 7], float32)]), Var(bn1.weight, ty=TensorType([64], float32)), Var(bn1.bias, ty=TensorType([64], float32)), Var(bn1.running_mean, ty=TensorType([64], float32)), Var(bn1.running_var, ty=TensorType([64], float32))], relay.attrs.BatchNormAttrs(0x7ff11120b3c8), [TensorType([1, 64, 112, 112], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32), TensorType([64], float32)]), 0)], (nullptr), [TensorType([1, 64, 112, 112], float32)])], relay.attrs.MaxPool2DAttrs(0x7ff10795e458), [TensorType([1, 64, 112, 112], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32), TensorType([1, 64, 56, 56], float32)])], (nullptr), [TensorType([1, 64, 56, 56], float32)]), Var(layer2.0.downsample.0.weight, ty=TensorType([128, 64, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b54d38), [TensorType([1, 64, 56, 56], float32), TensorType([128, 64, 1, 1], float32)]), Var(layer2.0.downsample.1.weight, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.bias, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_mean, ty=TensorType([128], float32)), Var(layer2.0.downsample.1.running_var, ty=TensorType([128], float32))], relay.attrs.BatchNormAttrs(0x7ff104b5e3f8), [TensorType([1, 128, 28, 28], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32), TensorType([128], float32)]), 0)], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32), TensorType([1, 128, 28, 28], float32)])], (nullptr), [TensorType([1, 128, 28, 28], float32)]), Var(layer3.0.downsample.0.weight, ty=TensorType([256, 128, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff10379b9c8), [TensorType([1, 128, 28, 28], float32), TensorType([256, 128, 1, 1], float32)]), Var(layer3.0.downsample.1.weight, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.bias, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_mean, ty=TensorType([256], float32)), Var(layer3.0.downsample.1.running_var, ty=TensorType([256], float32))], relay.attrs.BatchNormAttrs(0x7ff10374d6e8), [TensorType([1, 256, 14, 14], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32), TensorType([256], float32)]), 0)], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32), TensorType([1, 256, 14, 14], float32)])], (nullptr), [TensorType([1, 256, 14, 14], float32)]), Var(layer4.0.downsample.0.weight, ty=TensorType([512, 256, 1, 1], float32))], relay.attrs.Conv2DAttrs(0x7ff104b72c88), [TensorType([1, 256, 14, 14], float32), TensorType([512, 256, 1, 1], float32)]), Var(layer4.0.downsample.1.weight, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.bias, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.running_mean, ty=TensorType([512], float32)), Var(layer4.0.downsample.1.running_var, ty=TensorType([512], float32))], relay.attrs.BatchNormAttrs(0x7ff104b73238), [TensorType([1, 512, 7, 7], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32), TensorType([512], float32)]), 0)], (nullptr), [TensorType([1, 512, 7, 7], float32), TensorType([1, 512, 7, 7], float32)])], (nullptr), [TensorType([1, 512, 7, 7], float32)])], (nullptr), [TensorType([1, 512, 7, 7], float32), TensorType([1, 512, 7, 7], float32)])], (nullptr), [TensorType([1, 512, 7, 7], float32)])], relay.attrs.AdaptivePool2DAttrs(0x7ff102f932d8), [TensorType([1, 512, 7, 7], float32)])], relay.attrs.ReshapeAttrs(0x7ff102f95cb8), [TensorType([1, 512, 1, 1], float32)])], relay.attrs.SqueezeAttrs(0x7ff102f35158), [TensorType([1, 512, 1, 1], float32)]), CallNode(Op(transpose), [CallNode(Op(transpose), [Var(fc.weight, ty=TensorType([1000, 512], float32))], relay.attrs.TransposeAttrs(0x7ff1048c86e8), [TensorType([1000, 512], float32)])], relay.attrs.TransposeAttrs(0x7ff1048ba068), [TensorType([512, 1000], float32)])], relay.attrs.DenseAttrs(0x7ff1048e1268), [TensorType([1, 512], float32), TensorType([1000, 512], float32)]), Var(fc.bias, ty=TensorType([1000], float32))], (nullptr), [TensorType([1, 1000], float32), TensorType([1000], float32)]), [], (nullptr))"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"mod[\"main\"]"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"type Option[A] {\n",
" Some(A),\n",
" None,\n",
"}\n",
"\n",
"type Tree[A] {\n",
" Rose(A, List[Tree[A]]),\n",
"}\n",
"\n",
"type tensor_float64_t {\n",
" tensor_nil_float64,\n",
" tensor0_float64(float64),\n",
" tensor1_float64(Tensor[(?), float64]),\n",
" tensor2_float64(Tensor[(?, ?), float64]),\n",
" tensor3_float64(Tensor[(?, ?, ?), float64]),\n",
" tensor4_float64(Tensor[(?, ?, ?, ?), float64]),\n",
" tensor5_float64(Tensor[(?, ?, ?, ?, ?), float64]),\n",
" tensor6_float64(Tensor[(?, ?, ?, ?, ?, ?), float64]),\n",
"}\n",
"\n",
"type tensor_uint16_t {\n",
" tensor_nil_uint16,\n",
" tensor0_uint16(uint16),\n",
" tensor1_uint16(Tensor[(?), uint16]),\n",
" tensor2_uint16(Tensor[(?, ?), uint16]),\n",
" tensor3_uint16(Tensor[(?, ?, ?), uint16]),\n",
" tensor4_uint16(Tensor[(?, ?, ?, ?), uint16]),\n",
" tensor5_uint16(Tensor[(?, ?, ?, ?, ?), uint16]),\n",
" tensor6_uint16(Tensor[(?, ?, ?, ?, ?, ?), uint16]),\n",
"}\n",
"\n",
"type tensor_int16_t {\n",
" tensor_nil_int16,\n",
" tensor0_int16(int16),\n",
" tensor1_int16(Tensor[(?), int16]),\n",
" tensor2_int16(Tensor[(?, ?), int16]),\n",
" tensor3_int16(Tensor[(?, ?, ?), int16]),\n",
" tensor4_int16(Tensor[(?, ?, ?, ?), int16]),\n",
" tensor5_int16(Tensor[(?, ?, ?, ?, ?), int16]),\n",
" tensor6_int16(Tensor[(?, ?, ?, ?, ?, ?), int16]),\n",
"}\n",
"\n",
"type List[A] {\n",
" Cons(A, List[A]),\n",
" Nil,\n",
"}\n",
"\n",
"type tensor_float16_t {\n",
" tensor_nil_float16,\n",
" tensor0_float16(float16),\n",
" tensor1_float16(Tensor[(?), float16]),\n",
" tensor2_float16(Tensor[(?, ?), float16]),\n",
" tensor3_float16(Tensor[(?, ?, ?), float16]),\n",
" tensor4_float16(Tensor[(?, ?, ?, ?), float16]),\n",
" tensor5_float16(Tensor[(?, ?, ?, ?, ?), float16]),\n",
" tensor6_float16(Tensor[(?, ?, ?, ?, ?, ?), float16]),\n",
"}\n",
"\n",
"type tensor_uint8_t {\n",
" tensor_nil_uint8,\n",
" tensor0_uint8(uint8),\n",
" tensor1_uint8(Tensor[(?), uint8]),\n",
" tensor2_uint8(Tensor[(?, ?), uint8]),\n",
" tensor3_uint8(Tensor[(?, ?, ?), uint8]),\n",
" tensor4_uint8(Tensor[(?, ?, ?, ?), uint8]),\n",
" tensor5_uint8(Tensor[(?, ?, ?, ?, ?), uint8]),\n",
" tensor6_uint8(Tensor[(?, ?, ?, ?, ?, ?), uint8]),\n",
"}\n",
"\n",
"type tensor_float32_t {\n",
" tensor_nil_float32,\n",
" tensor0_float32(float32),\n",
" tensor1_float32(Tensor[(?), float32]),\n",
" tensor2_float32(Tensor[(?, ?), float32]),\n",
" tensor3_float32(Tensor[(?, ?, ?), float32]),\n",
" tensor4_float32(Tensor[(?, ?, ?, ?), float32]),\n",
" tensor5_float32(Tensor[(?, ?, ?, ?, ?), float32]),\n",
" tensor6_float32(Tensor[(?, ?, ?, ?, ?, ?), float32]),\n",
"}\n",
"\n",
"type tensor_int8_t {\n",
" tensor_nil_int8,\n",
" tensor0_int8(int8),\n",
" tensor1_int8(Tensor[(?), int8]),\n",
" tensor2_int8(Tensor[(?, ?), int8]),\n",
" tensor3_int8(Tensor[(?, ?, ?), int8]),\n",
" tensor4_int8(Tensor[(?, ?, ?, ?), int8]),\n",
" tensor5_int8(Tensor[(?, ?, ?, ?, ?), int8]),\n",
" tensor6_int8(Tensor[(?, ?, ?, ?, ?, ?), int8]),\n",
"}\n",
"\n",
"type tensor_int32_t {\n",
" tensor_nil_int32,\n",
" tensor0_int32(int32),\n",
" tensor1_int32(Tensor[(?), int32]),\n",
" tensor2_int32(Tensor[(?, ?), int32]),\n",
" tensor3_int32(Tensor[(?, ?, ?), int32]),\n",
" tensor4_int32(Tensor[(?, ?, ?, ?), int32]),\n",
" tensor5_int32(Tensor[(?, ?, ?, ?, ?), int32]),\n",
" tensor6_int32(Tensor[(?, ?, ?, ?, ?, ?), int32]),\n",
"}\n",
"\n",
"type tensor_int64_t {\n",
" tensor_nil_int64,\n",
" tensor0_int64(int64),\n",
" tensor1_int64(Tensor[(?), int64]),\n",
" tensor2_int64(Tensor[(?, ?), int64]),\n",
" tensor3_int64(Tensor[(?, ?, ?), int64]),\n",
" tensor4_int64(Tensor[(?, ?, ?, ?), int64]),\n",
" tensor5_int64(Tensor[(?, ?, ?, ?, ?), int64]),\n",
" tensor6_int64(Tensor[(?, ?, ?, ?, ?, ?), int64]),\n",
"}\n",
"\n",
"def @main(%input0: Tensor[(1, 3, 224, 224), float32], %conv1.weight: Tensor[(64, 3, 7, 7), float32], %bn1.weight: Tensor[(64), float32], %bn1.bias: Tensor[(64), float32], %bn1.running_mean: Tensor[(64), float32], %bn1.running_var: Tensor[(64), float32], %layer1.0.conv1.weight: Tensor[(64, 64, 3, 3), float32], %layer1.0.bn1.weight: Tensor[(64), float32], %layer1.0.bn1.bias: Tensor[(64), float32], %layer1.0.bn1.running_mean: Tensor[(64), float32], %layer1.0.bn1.running_var: Tensor[(64), float32], %layer1.0.conv2.weight: Tensor[(64, 64, 3, 3), float32], %layer1.0.bn2.weight: Tensor[(64), float32], %layer1.0.bn2.bias: Tensor[(64), float32], %layer1.0.bn2.running_mean: Tensor[(64), float32], %layer1.0.bn2.running_var: Tensor[(64), float32], %layer1.1.conv1.weight: Tensor[(64, 64, 3, 3), float32], %layer1.1.bn1.weight: Tensor[(64), float32], %layer1.1.bn1.bias: Tensor[(64), float32], %layer1.1.bn1.running_mean: Tensor[(64), float32], %layer1.1.bn1.running_var: Tensor[(64), float32], %layer1.1.conv2.weight: Tensor[(64, 64, 3, 3), float32], %layer1.1.bn2.weight: Tensor[(64), float32], %layer1.1.bn2.bias: Tensor[(64), float32], %layer1.1.bn2.running_mean: Tensor[(64), float32], %layer1.1.bn2.running_var: Tensor[(64), float32], %layer2.0.conv1.weight: Tensor[(128, 64, 3, 3), float32], %layer2.0.bn1.weight: Tensor[(128), float32], %layer2.0.bn1.bias: Tensor[(128), float32], %layer2.0.bn1.running_mean: Tensor[(128), float32], %layer2.0.bn1.running_var: Tensor[(128), float32], %layer2.0.conv2.weight: Tensor[(128, 128, 3, 3), float32], %layer2.0.bn2.weight: Tensor[(128), float32], %layer2.0.bn2.bias: Tensor[(128), float32], %layer2.0.bn2.running_mean: Tensor[(128), float32], %layer2.0.bn2.running_var: Tensor[(128), float32], %layer2.0.downsample.0.weight: Tensor[(128, 64, 1, 1), float32], %layer2.0.downsample.1.weight: Tensor[(128), float32], %layer2.0.downsample.1.bias: Tensor[(128), float32], %layer2.0.downsample.1.running_mean: Tensor[(128), float32], %layer2.0.downsample.1.running_var: Tensor[(128), float32], %layer2.1.conv1.weight: Tensor[(128, 128, 3, 3), float32], %layer2.1.bn1.weight: Tensor[(128), float32], %layer2.1.bn1.bias: Tensor[(128), float32], %layer2.1.bn1.running_mean: Tensor[(128), float32], %layer2.1.bn1.running_var: Tensor[(128), float32], %layer2.1.conv2.weight: Tensor[(128, 128, 3, 3), float32], %layer2.1.bn2.weight: Tensor[(128), float32], %layer2.1.bn2.bias: Tensor[(128), float32], %layer2.1.bn2.running_mean: Tensor[(128), float32], %layer2.1.bn2.running_var: Tensor[(128), float32], %layer3.0.conv1.weight: Tensor[(256, 128, 3, 3), float32], %layer3.0.bn1.weight: Tensor[(256), float32], %layer3.0.bn1.bias: Tensor[(256), float32], %layer3.0.bn1.running_mean: Tensor[(256), float32], %layer3.0.bn1.running_var: Tensor[(256), float32], %layer3.0.conv2.weight: Tensor[(256, 256, 3, 3), float32], %layer3.0.bn2.weight: Tensor[(256), float32], %layer3.0.bn2.bias: Tensor[(256), float32], %layer3.0.bn2.running_mean: Tensor[(256), float32], %layer3.0.bn2.running_var: Tensor[(256), float32], %layer3.0.downsample.0.weight: Tensor[(256, 128, 1, 1), float32], %layer3.0.downsample.1.weight: Tensor[(256), float32], %layer3.0.downsample.1.bias: Tensor[(256), float32], %layer3.0.downsample.1.running_mean: Tensor[(256), float32], %layer3.0.downsample.1.running_var: Tensor[(256), float32], %layer3.1.conv1.weight: Tensor[(256, 256, 3, 3), float32], %layer3.1.bn1.weight: Tensor[(256), float32], %layer3.1.bn1.bias: Tensor[(256), float32], %layer3.1.bn1.running_mean: Tensor[(256), float32], %layer3.1.bn1.running_var: Tensor[(256), float32], %layer3.1.conv2.weight: Tensor[(256, 256, 3, 3), float32], %layer3.1.bn2.weight: Tensor[(256), float32], %layer3.1.bn2.bias: Tensor[(256), float32], %layer3.1.bn2.running_mean: Tensor[(256), float32], %layer3.1.bn2.running_var: Tensor[(256), float32], %layer4.0.conv1.weight: Tensor[(512, 256, 3, 3), float32], %layer4.0.bn1.weight: Tensor[(512), float32], %layer4.0.bn1.bias: Tensor[(512), float32], %layer4.0.bn1.running_mean: Tensor[(512), float32], %layer4.0.bn1.running_var: Tensor[(512), float32], %layer4.0.conv2.weight: Tensor[(512, 512, 3, 3), float32], %layer4.0.bn2.weight: Tensor[(512), float32], %layer4.0.bn2.bias: Tensor[(512), float32], %layer4.0.bn2.running_mean: Tensor[(512), float32], %layer4.0.bn2.running_var: Tensor[(512), float32], %layer4.0.downsample.0.weight: Tensor[(512, 256, 1, 1), float32], %layer4.0.downsample.1.weight: Tensor[(512), float32], %layer4.0.downsample.1.bias: Tensor[(512), float32], %layer4.0.downsample.1.running_mean: Tensor[(512), float32], %layer4.0.downsample.1.running_var: Tensor[(512), float32], %layer4.1.conv1.weight: Tensor[(512, 512, 3, 3), float32], %layer4.1.bn1.weight: Tensor[(512), float32], %layer4.1.bn1.bias: Tensor[(512), float32], %layer4.1.bn1.running_mean: Tensor[(512), float32], %layer4.1.bn1.running_var: Tensor[(512), float32], %layer4.1.conv2.weight: Tensor[(512, 512, 3, 3), float32], %layer4.1.bn2.weight: Tensor[(512), float32], %layer4.1.bn2.bias: Tensor[(512), float32], %layer4.1.bn2.running_mean: Tensor[(512), float32], %layer4.1.bn2.running_var: Tensor[(512), float32], %fc.weight: Tensor[(1000, 512), float32], %fc.bias: Tensor[(1000), float32]) -> Tensor[(1, 1000), float32] {\n",
" %0 = nn.conv2d(%input0, %conv1.weight, strides=[2, 2], padding=[3, 3, 3, 3], channels=64, kernel_size=[7, 7]) /* ty=Tensor[(1, 64, 112, 112), float32] */;\n",
" %1 = nn.batch_norm(%0, %bn1.weight, %bn1.bias, %bn1.running_mean, %bn1.running_var) /* ty=(Tensor[(1, 64, 112, 112), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %2 = %1.0;\n",
" %3 = nn.relu(%2) /* ty=Tensor[(1, 64, 112, 112), float32] */;\n",
" %4 = nn.max_pool2d(%3, pool_size=[3, 3], strides=[2, 2], padding=[1, 1, 1, 1]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %5 = nn.conv2d(%4, %layer1.0.conv1.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %6 = nn.batch_norm(%5, %layer1.0.bn1.weight, %layer1.0.bn1.bias, %layer1.0.bn1.running_mean, %layer1.0.bn1.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %7 = %6.0;\n",
" %8 = nn.relu(%7) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %9 = nn.conv2d(%8, %layer1.0.conv2.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %10 = nn.batch_norm(%9, %layer1.0.bn2.weight, %layer1.0.bn2.bias, %layer1.0.bn2.running_mean, %layer1.0.bn2.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %11 = %10.0;\n",
" %12 = add(%11, %4) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %13 = nn.relu(%12) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %14 = nn.conv2d(%13, %layer1.1.conv1.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %15 = nn.batch_norm(%14, %layer1.1.bn1.weight, %layer1.1.bn1.bias, %layer1.1.bn1.running_mean, %layer1.1.bn1.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %16 = %15.0;\n",
" %17 = nn.relu(%16) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %18 = nn.conv2d(%17, %layer1.1.conv2.weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %19 = nn.batch_norm(%18, %layer1.1.bn2.weight, %layer1.1.bn2.bias, %layer1.1.bn2.running_mean, %layer1.1.bn2.running_var) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;\n",
" %20 = %19.0;\n",
" %21 = add(%20, %13) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %22 = nn.relu(%21) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %23 = nn.conv2d(%22, %layer2.0.conv1.weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %24 = nn.batch_norm(%23, %layer2.0.bn1.weight, %layer2.0.bn1.bias, %layer2.0.bn1.running_mean, %layer2.0.bn1.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %25 = %24.0;\n",
" %26 = nn.relu(%25) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %27 = nn.conv2d(%26, %layer2.0.conv2.weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %28 = nn.batch_norm(%27, %layer2.0.bn2.weight, %layer2.0.bn2.bias, %layer2.0.bn2.running_mean, %layer2.0.bn2.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %29 = %28.0;\n",
" %30 = nn.conv2d(%22, %layer2.0.downsample.0.weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=128, kernel_size=[1, 1]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %31 = nn.batch_norm(%30, %layer2.0.downsample.1.weight, %layer2.0.downsample.1.bias, %layer2.0.downsample.1.running_mean, %layer2.0.downsample.1.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %32 = %31.0;\n",
" %33 = add(%29, %32) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %34 = nn.relu(%33) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %35 = nn.conv2d(%34, %layer2.1.conv1.weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %36 = nn.batch_norm(%35, %layer2.1.bn1.weight, %layer2.1.bn1.bias, %layer2.1.bn1.running_mean, %layer2.1.bn1.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %37 = %36.0;\n",
" %38 = nn.relu(%37) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %39 = nn.conv2d(%38, %layer2.1.conv2.weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %40 = nn.batch_norm(%39, %layer2.1.bn2.weight, %layer2.1.bn2.bias, %layer2.1.bn2.running_mean, %layer2.1.bn2.running_var) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;\n",
" %41 = %40.0;\n",
" %42 = add(%41, %34) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %43 = nn.relu(%42) /* ty=Tensor[(1, 128, 28, 28), float32] */;\n",
" %44 = nn.conv2d(%43, %layer3.0.conv1.weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %45 = nn.batch_norm(%44, %layer3.0.bn1.weight, %layer3.0.bn1.bias, %layer3.0.bn1.running_mean, %layer3.0.bn1.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %46 = %45.0;\n",
" %47 = nn.relu(%46) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %48 = nn.conv2d(%47, %layer3.0.conv2.weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %49 = nn.batch_norm(%48, %layer3.0.bn2.weight, %layer3.0.bn2.bias, %layer3.0.bn2.running_mean, %layer3.0.bn2.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %50 = %49.0;\n",
" %51 = nn.conv2d(%43, %layer3.0.downsample.0.weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=256, kernel_size=[1, 1]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %52 = nn.batch_norm(%51, %layer3.0.downsample.1.weight, %layer3.0.downsample.1.bias, %layer3.0.downsample.1.running_mean, %layer3.0.downsample.1.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %53 = %52.0;\n",
" %54 = add(%50, %53) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %55 = nn.relu(%54) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %56 = nn.conv2d(%55, %layer3.1.conv1.weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %57 = nn.batch_norm(%56, %layer3.1.bn1.weight, %layer3.1.bn1.bias, %layer3.1.bn1.running_mean, %layer3.1.bn1.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %58 = %57.0;\n",
" %59 = nn.relu(%58) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %60 = nn.conv2d(%59, %layer3.1.conv2.weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %61 = nn.batch_norm(%60, %layer3.1.bn2.weight, %layer3.1.bn2.bias, %layer3.1.bn2.running_mean, %layer3.1.bn2.running_var) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;\n",
" %62 = %61.0;\n",
" %63 = add(%62, %55) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %64 = nn.relu(%63) /* ty=Tensor[(1, 256, 14, 14), float32] */;\n",
" %65 = nn.conv2d(%64, %layer4.0.conv1.weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %66 = nn.batch_norm(%65, %layer4.0.bn1.weight, %layer4.0.bn1.bias, %layer4.0.bn1.running_mean, %layer4.0.bn1.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %67 = %66.0;\n",
" %68 = nn.relu(%67) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %69 = nn.conv2d(%68, %layer4.0.conv2.weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %70 = nn.batch_norm(%69, %layer4.0.bn2.weight, %layer4.0.bn2.bias, %layer4.0.bn2.running_mean, %layer4.0.bn2.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %71 = %70.0;\n",
" %72 = nn.conv2d(%64, %layer4.0.downsample.0.weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=512, kernel_size=[1, 1]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %73 = nn.batch_norm(%72, %layer4.0.downsample.1.weight, %layer4.0.downsample.1.bias, %layer4.0.downsample.1.running_mean, %layer4.0.downsample.1.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %74 = %73.0;\n",
" %75 = add(%71, %74) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %76 = nn.relu(%75) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %77 = nn.conv2d(%76, %layer4.1.conv1.weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %78 = nn.batch_norm(%77, %layer4.1.bn1.weight, %layer4.1.bn1.bias, %layer4.1.bn1.running_mean, %layer4.1.bn1.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %79 = %78.0;\n",
" %80 = nn.relu(%79) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %81 = nn.conv2d(%80, %layer4.1.conv2.weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %82 = nn.batch_norm(%81, %layer4.1.bn2.weight, %layer4.1.bn2.bias, %layer4.1.bn2.running_mean, %layer4.1.bn2.running_var) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;\n",
" %83 = %82.0;\n",
" %84 = add(%83, %76) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %85 = nn.relu(%84) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %86 = nn.adaptive_avg_pool2d(%85, output_size=[1, 1]) /* ty=Tensor[(1, 512, 1, 1), float32] */;\n",
" %87 = reshape(%86, newshape=[0, -1, 1, 1]) /* ty=Tensor[(1, 512, 1, 1), float32] */;\n",
" %88 = squeeze(%87, axis=[2, 3]) /* ty=Tensor[(1, 512), float32] */;\n",
" %89 = transpose(%fc.weight, axes=[1, 0]) /* ty=Tensor[(512, 1000), float32] */;\n",
" %90 = transpose(%89, axes=[1, 0]) /* ty=Tensor[(1000, 512), float32] */;\n",
" %91 = nn.dense(%88, %90, units=1000) /* ty=Tensor[(1, 1000), float32] */;\n",
" add(%91, %fc.bias) /* ty=Tensor[(1, 1000), float32] */\n",
"}\n",
"\n"
]
}
],
"source": [
"print(mod)"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [],
"source": [
" if target.device_name == \"vta\":\n",
" # Perform quantization in Relay\n",
" # Note: We set opt_level to 3 in order to fold batch norm\n",
" with tvm.transform.PassContext(opt_level=3):\n",
" with relay.quantize.qconfig(global_scale=8.0, skip_conv_layers=[0]):\n",
" mod = relay.quantize.quantize(mod, params=params)\n",
" else:\n",
" relay_prog = mod[\"main\"]\n",
" \n"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"def @main(%input0: Tensor[(1, 3, 224, 224), float32]) -> Tensor[(1, 1000), float32] {\n",
" %0 = nn.conv2d(%input0, meta[relay.Constant][0] /* ty=Tensor[(64, 3, 7, 7), float32] */, strides=[2, 2], padding=[3, 3, 3, 3], channels=64, kernel_size=[7, 7]) /* ty=Tensor[(1, 64, 112, 112), float32] */;\n",
" %1 = add(%0, meta[relay.Constant][1] /* ty=Tensor[(64, 1, 1), float32] */) /* ty=Tensor[(1, 64, 112, 112), float32] */;\n",
" %2 = nn.relu(%1) /* ty=Tensor[(1, 64, 112, 112), float32] */;\n",
" %3 = nn.max_pool2d(%2, pool_size=[3, 3], strides=[2, 2], padding=[1, 1, 1, 1]) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %4 = annotation.stop_fusion(%3) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %5 = multiply(%4, 16f /* ty=float32 */) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %6 = round(%5) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %7 = clip(%6, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %8 = cast(%7, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %9 = nn.conv2d(%8, meta[relay.Constant][2] /* ty=Tensor[(64, 64, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %10 = add(%9, meta[relay.Constant][3] /* ty=Tensor[(64, 1, 1), int32] */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %11 = nn.relu(%10) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %12 = add(%11, 128 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %13 = right_shift(%12, 8 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %14 = clip(%13, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %15 = cast(%14, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %16 = annotation.stop_fusion(%15) /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %17 = nn.conv2d(%16, meta[relay.Constant][4] /* ty=Tensor[(64, 64, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %18 = add(%17, meta[relay.Constant][5] /* ty=Tensor[(64, 1, 1), int32] */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %19 = add(%18, 64 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %20 = right_shift(%19, 7 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %21 = clip(%20, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %22 = cast(%21, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %23 = annotation.stop_fusion(%22) /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %24 = cast(%23, dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %25 = annotation.stop_fusion(%3) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %26 = multiply(%25, 16f /* ty=float32 */) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %27 = round(%26) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %28 = clip(%27, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), float32] */;\n",
" %29 = cast(%28, dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %30 = add(%24, %29) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %31 = nn.relu(%30) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %32 = clip(%31, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %33 = cast(%32, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %34 = nn.conv2d(%33, meta[relay.Constant][6] /* ty=Tensor[(64, 64, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %35 = add(%34, meta[relay.Constant][7] /* ty=Tensor[(64, 1, 1), int32] */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %36 = nn.relu(%35) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %37 = add(%36, 128 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %38 = right_shift(%37, 8 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %39 = clip(%38, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %40 = cast(%39, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %41 = annotation.stop_fusion(%40) /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %42 = nn.conv2d(%41, meta[relay.Constant][8] /* ty=Tensor[(64, 64, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %43 = add(%42, meta[relay.Constant][9] /* ty=Tensor[(64, 1, 1), int32] */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %44 = add(%43, 32 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %45 = right_shift(%44, 6 /* ty=int32 */) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %46 = clip(%45, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %47 = cast(%46, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %48 = annotation.stop_fusion(%47) /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %49 = cast(%48, dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %50 = cast(%32, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %51 = annotation.stop_fusion(%50) /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %52 = cast(%51, dtype=\"int32\") /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %53 = add(%49, %52) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %54 = nn.relu(%53) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %55 = clip(%54, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 64, 56, 56), int32] */;\n",
" %56 = cast(%55, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %57 = nn.conv2d(%56, meta[relay.Constant][10] /* ty=Tensor[(128, 64, 3, 3), int8] */, strides=[2, 2], padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %58 = add(%57, meta[relay.Constant][11] /* ty=Tensor[(128, 1, 1), int32] */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %59 = nn.relu(%58) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %60 = add(%59, 256 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %61 = right_shift(%60, 9 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %62 = clip(%61, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %63 = cast(%62, dtype=\"int8\") /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %64 = annotation.stop_fusion(%63) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %65 = nn.conv2d(%64, meta[relay.Constant][12] /* ty=Tensor[(128, 128, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %66 = add(%65, meta[relay.Constant][13] /* ty=Tensor[(128, 1, 1), int32] */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %67 = add(%66, 64 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %68 = right_shift(%67, 7 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %69 = clip(%68, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %70 = cast(%69, dtype=\"int8\") /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %71 = annotation.stop_fusion(%70) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %72 = cast(%55, dtype=\"int8\") /* ty=Tensor[(1, 64, 56, 56), int8] */;\n",
" %73 = nn.conv2d(%72, meta[relay.Constant][14] /* ty=Tensor[(128, 64, 1, 1), int8] */, strides=[2, 2], padding=[0, 0, 0, 0], channels=128, kernel_size=[1, 1], out_dtype=\"int32\") /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %74 = add(%73, meta[relay.Constant][15] /* ty=Tensor[(128, 1, 1), int32] */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %75 = add(%74, 64 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %76 = right_shift(%75, 7 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %77 = clip(%76, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %78 = cast(%77, dtype=\"int8\") /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %79 = annotation.stop_fusion(%78) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %80 = add(%71, %79) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %81 = nn.relu(%80) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %82 = clip(%81, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %83 = nn.conv2d(%82, meta[relay.Constant][16] /* ty=Tensor[(128, 128, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %84 = add(%83, meta[relay.Constant][17] /* ty=Tensor[(128, 1, 1), int32] */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %85 = nn.relu(%84) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %86 = add(%85, 128 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %87 = right_shift(%86, 8 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %88 = clip(%87, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %89 = cast(%88, dtype=\"int8\") /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %90 = annotation.stop_fusion(%89) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %91 = nn.conv2d(%90, meta[relay.Constant][18] /* ty=Tensor[(128, 128, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %92 = add(%91, meta[relay.Constant][19] /* ty=Tensor[(128, 1, 1), int32] */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %93 = add(%92, 64 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %94 = right_shift(%93, 7 /* ty=int32 */) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %95 = clip(%94, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 128, 28, 28), int32] */;\n",
" %96 = cast(%95, dtype=\"int8\") /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %97 = annotation.stop_fusion(%96) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %98 = cast(%82, dtype=\"int8\") /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %99 = annotation.stop_fusion(%98) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %100 = cast(%99, dtype=\"int8\") /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %101 = add(%97, %100) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %102 = nn.relu(%101) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %103 = clip(%102, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 128, 28, 28), int8] */;\n",
" %104 = nn.conv2d(%103, meta[relay.Constant][20] /* ty=Tensor[(256, 128, 3, 3), int8] */, strides=[2, 2], padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %105 = add(%104, meta[relay.Constant][21] /* ty=Tensor[(256, 1, 1), int32] */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %106 = nn.relu(%105) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %107 = add(%106, 256 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %108 = right_shift(%107, 9 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %109 = clip(%108, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %110 = cast(%109, dtype=\"int8\") /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %111 = annotation.stop_fusion(%110) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %112 = nn.conv2d(%111, meta[relay.Constant][22] /* ty=Tensor[(256, 256, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %113 = add(%112, meta[relay.Constant][23] /* ty=Tensor[(256, 1, 1), int32] */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %114 = add(%113, 64 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %115 = right_shift(%114, 7 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %116 = clip(%115, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %117 = cast(%116, dtype=\"int8\") /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %118 = annotation.stop_fusion(%117) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %119 = nn.conv2d(%103, meta[relay.Constant][24] /* ty=Tensor[(256, 128, 1, 1), int8] */, strides=[2, 2], padding=[0, 0, 0, 0], channels=256, kernel_size=[1, 1], out_dtype=\"int32\") /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %120 = add(%119, meta[relay.Constant][25] /* ty=Tensor[(256, 1, 1), int32] */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %121 = add(%120, 128 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %122 = right_shift(%121, 8 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %123 = clip(%122, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %124 = cast(%123, dtype=\"int8\") /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %125 = annotation.stop_fusion(%124) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %126 = add(%118, %125) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %127 = nn.relu(%126) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %128 = clip(%127, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %129 = nn.conv2d(%128, meta[relay.Constant][26] /* ty=Tensor[(256, 256, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %130 = add(%129, meta[relay.Constant][27] /* ty=Tensor[(256, 1, 1), int32] */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %131 = nn.relu(%130) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %132 = add(%131, 128 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %133 = right_shift(%132, 8 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %134 = clip(%133, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %135 = cast(%134, dtype=\"int8\") /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %136 = annotation.stop_fusion(%135) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %137 = nn.conv2d(%136, meta[relay.Constant][28] /* ty=Tensor[(256, 256, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %138 = add(%137, meta[relay.Constant][29] /* ty=Tensor[(256, 1, 1), int32] */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %139 = add(%138, 64 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %140 = right_shift(%139, 7 /* ty=int32 */) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %141 = clip(%140, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 256, 14, 14), int32] */;\n",
" %142 = cast(%141, dtype=\"int8\") /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %143 = annotation.stop_fusion(%142) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %144 = cast(%128, dtype=\"int8\") /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %145 = annotation.stop_fusion(%144) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %146 = cast(%145, dtype=\"int8\") /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %147 = add(%143, %146) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %148 = nn.relu(%147) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %149 = clip(%148, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 256, 14, 14), int8] */;\n",
" %150 = nn.conv2d(%149, meta[relay.Constant][30] /* ty=Tensor[(512, 256, 3, 3), int8] */, strides=[2, 2], padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %151 = add(%150, meta[relay.Constant][31] /* ty=Tensor[(512, 1, 1), int32] */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %152 = nn.relu(%151) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %153 = add(%152, 128 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %154 = right_shift(%153, 8 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %155 = clip(%154, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %156 = cast(%155, dtype=\"int8\") /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %157 = annotation.stop_fusion(%156) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %158 = nn.conv2d(%157, meta[relay.Constant][32] /* ty=Tensor[(512, 512, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %159 = add(%158, meta[relay.Constant][33] /* ty=Tensor[(512, 1, 1), int32] */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %160 = add(%159, 32 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %161 = right_shift(%160, 6 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %162 = clip(%161, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %163 = cast(%162, dtype=\"int8\") /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %164 = annotation.stop_fusion(%163) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %165 = nn.conv2d(%149, meta[relay.Constant][34] /* ty=Tensor[(512, 256, 1, 1), int8] */, strides=[2, 2], padding=[0, 0, 0, 0], channels=512, kernel_size=[1, 1], out_dtype=\"int32\") /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %166 = add(%165, meta[relay.Constant][35] /* ty=Tensor[(512, 1, 1), int32] */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %167 = add(%166, 64 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %168 = right_shift(%167, 7 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %169 = clip(%168, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %170 = cast(%169, dtype=\"int8\") /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %171 = annotation.stop_fusion(%170) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %172 = add(%164, %171) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %173 = nn.relu(%172) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %174 = clip(%173, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %175 = nn.conv2d(%174, meta[relay.Constant][36] /* ty=Tensor[(512, 512, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %176 = add(%175, meta[relay.Constant][37] /* ty=Tensor[(512, 1, 1), int32] */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %177 = nn.relu(%176) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %178 = add(%177, 128 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %179 = right_shift(%178, 8 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %180 = clip(%179, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %181 = cast(%180, dtype=\"int8\") /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %182 = annotation.stop_fusion(%181) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %183 = nn.conv2d(%182, meta[relay.Constant][38] /* ty=Tensor[(512, 512, 3, 3), int8] */, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3], out_dtype=\"int32\") /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %184 = add(%183, meta[relay.Constant][39] /* ty=Tensor[(512, 1, 1), int32] */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %185 = add(%184, 16 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %186 = right_shift(%185, 5 /* ty=int32 */) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %187 = clip(%186, a_min=-127f, a_max=127f) /* ty=Tensor[(1, 512, 7, 7), int32] */;\n",
" %188 = cast(%187, dtype=\"int8\") /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %189 = annotation.stop_fusion(%188) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %190 = cast(%174, dtype=\"int8\") /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %191 = annotation.stop_fusion(%190) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %192 = cast(%191, dtype=\"int8\") /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %193 = add(%189, %192) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %194 = nn.relu(%193) /* ty=Tensor[(1, 512, 7, 7), int8] */;\n",
" %195 = cast(%194, dtype=\"float32\") /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %196 = multiply(%195, 0.0625f /* ty=float32 */) /* ty=Tensor[(1, 512, 7, 7), float32] */;\n",
" %197 = nn.adaptive_avg_pool2d(%196, output_size=[1, 1]) /* ty=Tensor[(1, 512, 1, 1), float32] */;\n",
" %198 = reshape(%197, newshape=[0, -1, 1, 1]) /* ty=Tensor[(1, 512, 1, 1), float32] */;\n",
" %199 = squeeze(%198, axis=[2, 3]) /* ty=Tensor[(1, 512), float32] */;\n",
" %200 = nn.dense(%199, meta[relay.Constant][40] /* ty=Tensor[(1000, 512), float32] */, units=1000) /* ty=Tensor[(1, 1000), float32] */;\n",
" add(%200, meta[relay.Constant][41] /* ty=Tensor[(1000), float32] */) /* ty=Tensor[(1, 1000), float32] */\n",
"}\n",
"\n",
"\n"
]
}
],
"source": [
"print(mod)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"scrolled": true
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"\"-target\" is deprecated, use \"-mtriple\" instead.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 512, 7, 7), 'int8'), ('TENSOR', (512, 512, 3, 3), 'int8'), (1, 1), (1, 1, 1, 1), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 256, 14, 14), 'int8'), ('TENSOR', (512, 256, 3, 3), 'int8'), (2, 2), (1, 1, 1, 1), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 256, 14, 14), 'int8'), ('TENSOR', (256, 256, 3, 3), 'int8'), (1, 1), (1, 1, 1, 1), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 128, 28, 28), 'int8'), ('TENSOR', (256, 128, 3, 3), 'int8'), (2, 2), (1, 1, 1, 1), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 128, 28, 28), 'int8'), ('TENSOR', (128, 128, 3, 3), 'int8'), (1, 1), (1, 1, 1, 1), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 64, 56, 56), 'int8'), ('TENSOR', (128, 64, 3, 3), 'int8'), (2, 2), (1, 1, 1, 1), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 64, 56, 56), 'int8'), ('TENSOR', (64, 64, 3, 3), 'int8'), (1, 1), (1, 1, 1, 1), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 64, 56, 56), 'int8'), ('TENSOR', (128, 64, 1, 1), 'int8'), (2, 2), (0, 0, 0, 0), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 128, 28, 28), 'int8'), ('TENSOR', (256, 128, 1, 1), 'int8'), (2, 2), (0, 0, 0, 0), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n",
"Cannot find config for target=ext_dev -keys=vta,cpu -device=vta -model=sim_1x16_i8w8a32_15_15_18_17, workload=('conv2d_nchw_spatial_pack.arm_cpu', ('TENSOR', (1, 256, 14, 14), 'int8'), ('TENSOR', (512, 256, 1, 1), 'int8'), (2, 2), (0, 0, 0, 0), (1, 1), 'int32'). A fallback configuration is used, which may bring great performance regression.\n"
]
}
],
"source": [
"# Compile Relay program with AlterOpLayout disabled\n",
"if target.device_name != \"vta\":\n",
" with tvm.transform.PassContext(opt_level=3, disabled_pass={\"AlterOpLayout\"}):\n",
" graph, lib, params = relay.build(\n",
" mod, target=target, params=params, target_host=env.target_host\n",
" )\n",
"else:\n",
" with vta.build_config(opt_level=3, disabled_pass={\"AlterOpLayout\"}):\n",
" lib = relay.build(mod, target=target, params=params, target_host=env.target_host)"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
" # Send the inference library over to the remote RPC server\n",
" from tvm.contrib import util\n",
" temp = util.tempdir()\n",
" lib.export_library(temp.relpath(\"graphlib.tar\"))\n",
" remote.upload(temp.relpath(\"graphlib.tar\"))\n",
" lib = remote.load_module(\"graphlib.tar\")\n",
"\n",
" # Graph runtime\n",
" m = graph_runtime.GraphModule(lib[\"default\"](ctx))"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Module(rpc, 7ff107f8ca08)\n"
]
}
],
"source": [
"print(lib)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## perform the inference"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"# Set inputs\n",
"m.set_input(input_name, tvm.nd.array(input_data))\n",
"# Execute\n",
"m.run()\n",
"# Get outputs\n",
"tvm_output = m.get_output(0)"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<tvm.nd.NDArray shape=(1, 1000), remote[0]:ext_dev(0)>\n",
"array([[ 1.08618414e+00, 3.58600712e+00, 2.04241300e+00,\n",
" 2.79078460e+00, 5.12289047e+00, 6.01200676e+00,\n",
" 5.69327021e+00, 1.08313298e+00, -1.94065928e-01,\n",
" -8.66146684e-02, -5.65413833e-01, 5.66011488e-01,\n",
" 5.59885025e-01, 1.47489047e+00, 1.96457028e+00,\n",
" 1.46113014e+00, -8.97701383e-01, 1.52540177e-01,\n",
" 1.63587523e+00, 2.37933248e-01, -1.31542355e-01,\n",
" -9.69795704e-01, 3.29282761e-01, 2.32619435e-01,\n",
" -4.41796154e-01, 4.63238806e-01, 1.20617115e+00,\n",
" 1.47301626e+00, 3.90578717e-01, 1.77868617e+00,\n",
" 1.65011251e+00, 2.24781528e-01, -4.37062800e-01,\n",
" 2.74081016e+00, 2.77148771e+00, 1.66879857e+00,\n",
" 1.17748499e+00, 1.56333935e+00, 6.96946740e-01,\n",
" 2.54402041e+00, 2.09717464e+00, 1.80453467e+00,\n",
" 1.91088331e+00, 1.79514563e+00, 2.00588584e+00,\n",
" 1.00584984e+00, 2.62914515e+00, -8.12225819e-01,\n",
" 2.45213580e+00, 1.81252468e+00, 2.81083250e+00,\n",
" -1.82743943e+00, 1.27373910e+00, 8.21564078e-01,\n",
" 1.32698283e-01, 1.83107007e+00, -4.42182302e-01,\n",
" 3.22389513e-01, 2.38680935e+00, 4.39417422e-01,\n",
" 2.18580174e+00, 2.41821706e-01, 4.74410146e-01,\n",
" -1.88933223e-01, 8.61847639e-01, 4.79643345e+00,\n",
" 8.51827264e-01, 7.60055244e-01, 1.14879823e+00,\n",
" 2.49305344e+00, 2.09485149e+00, 2.27700591e+00,\n",
" 1.42790467e-01, 6.82145536e-01, 3.42027873e-01,\n",
" 2.45001149e+00, 1.43974936e+00, 1.61738443e+00,\n",
" 1.12437773e+00, 2.76270890e+00, -1.94217896e+00,\n",
" 1.11053538e+00, 2.66064435e-01, -2.31463337e+00,\n",
" 1.66323149e+00, 1.69585621e+00, 2.55509585e-01,\n",
" -9.91764665e-01, -5.15731156e-01, 1.15898907e+00,\n",
" -1.33211493e+00, -8.95358741e-01, 1.10400748e+00,\n",
" -5.04402936e-01, 4.51064676e-01, -2.03192282e+00,\n",
" -2.12016311e-02, -1.33406207e-01, -8.46441448e-01,\n",
" 3.80128324e-01, 6.77205622e-03, -1.53180075e+00,\n",
" -9.10314500e-01, 6.81383014e-01, 7.73328245e-01,\n",
" -3.45612377e-01, 2.65543699e-01, 6.92346048e+00,\n",
" 2.89044809e+00, 3.33969593e+00, 2.78945947e+00,\n",
" 3.47729135e+00, 3.18856144e+00, 2.64913034e+00,\n",
" 2.64445353e+00, 9.92854178e-01, 1.52496099e+00,\n",
" 1.80506968e+00, 1.61883736e+00, 1.33004355e+00,\n",
" 1.48089612e+00, 3.87030959e-01, 1.08933651e+00,\n",
" 2.74953794e+00, 3.60624075e+00, 2.69083428e+00,\n",
" 2.79748464e+00, -3.47478271e-01, 7.00453281e-01,\n",
" 1.48198545e+00, -2.13774785e-01, 9.50405121e-01,\n",
" 2.56706095e+00, 1.24670482e+00, 1.41666031e+00,\n",
" 5.46366632e-01, -7.08506703e-01, 3.17729324e-01,\n",
" 1.71821427e+00, 5.87539554e-01, 5.50138772e-01,\n",
" 1.57895052e+00, 1.12751317e+00, 9.69772220e-01,\n",
" 2.30406940e-01, 4.46623445e-01, -1.25655508e+00,\n",
" -2.19489336e-02, 1.09646179e-01, 2.41094661e+00,\n",
" 1.09990191e+00, -2.70628899e-01, -2.73048663e+00,\n",
" 7.22588181e-01, -1.59288272e-01, 2.62489498e-01,\n",
" -1.07284176e+00, -7.99387753e-01, -6.83991730e-01,\n",
" -1.02515411e+00, -2.80417991e+00, -1.80204630e-01,\n",
" 4.41157967e-01, -8.07245731e-01, -1.09995735e+00,\n",
" -1.24998951e+00, -8.97869706e-01, -2.60487890e+00,\n",
" -6.07123733e-01, -1.49990940e+00, -2.16689730e+00,\n",
" 2.30283424e-01, -1.20377910e+00, -2.29696199e-01,\n",
" -1.16843545e+00, -3.02678180e+00, -1.19537640e+00,\n",
" -2.07507515e+00, 9.04945195e-01, -7.28762865e-01,\n",
" -8.21858421e-02, -4.91953224e-01, -8.55558276e-01,\n",
" -9.34022427e-01, -1.04145277e+00, -1.02066875e+00,\n",
" -1.81945097e+00, -2.52983660e-01, -2.48946190e+00,\n",
" -5.67993104e-01, -1.53946483e+00, -1.17629361e+00,\n",
" -7.84540415e-01, -1.37181079e+00, -2.48251963e+00,\n",
" -1.60107344e-01, -8.28174472e-01, -2.48720670e+00,\n",
" -2.09508014e+00, -4.36627150e-01, -3.02643466e+00,\n",
" 3.89458686e-01, -1.06824636e+00, 1.43833745e+00,\n",
" 1.36150587e+00, -1.63626289e+00, -2.27929664e+00,\n",
" 1.59454662e-02, -1.08112311e+00, 7.92957097e-02,\n",
" -1.75175881e+00, -1.22392440e+00, -1.15495253e+00,\n",
" -1.28531829e-01, -1.94001973e+00, -9.64464426e-01,\n",
" -1.21236193e+00, -3.69648665e-01, -1.05986691e+00,\n",
" -1.15674031e+00, -2.77412534e+00, -1.12587762e+00,\n",
" -2.20160270e+00, 1.31227523e-02, -5.21864772e-01,\n",
" -1.68199074e+00, -2.03161073e+00, -2.22486234e+00,\n",
" -2.87026286e+00, -1.19772601e+00, -5.79524674e-02,\n",
" -1.12180233e+00, -1.95485801e-01, -3.10896039e+00,\n",
" -8.42313468e-01, -1.57053626e+00, -7.51034319e-01,\n",
" -7.17659533e-01, -1.18158960e+00, -1.93714130e+00,\n",
" -1.58646905e+00, -2.09990835e+00, -4.68967348e-01,\n",
" -1.14599538e+00, -4.57070112e+00, -3.04223627e-01,\n",
" -1.74140918e+00, -1.61323619e+00, -5.93328893e-01,\n",
" -1.94160163e+00, -5.63923299e-01, -8.28243196e-01,\n",
" -2.82759571e+00, -4.53982413e-01, -5.46679914e-01,\n",
" -3.05702996e+00, -2.50537467e+00, -1.32533324e+00,\n",
" -1.14837635e+00, 3.11215490e-01, -1.70880973e+00,\n",
" -1.62900102e+00, -3.58738589e+00, -2.08900660e-01,\n",
" -1.43031132e+00, -6.04722500e-02, -6.46871328e-01,\n",
" -2.00499082e+00, -1.97122955e+00, -2.07128882e+00,\n",
" -1.18170369e+00, -1.77189672e+00, 1.29241556e-01,\n",
" -1.85393822e+00, -1.06033826e+00, -2.33754182e+00,\n",
" -2.09594655e+00, -4.14108366e-01, -9.28400218e-01,\n",
" 7.64854193e-01, 6.82094812e-01, 6.77053034e-01,\n",
" 1.09528506e+00, 3.70565914e-02, 1.04173112e+00,\n",
" 1.25884982e-02, -1.15088427e+00, -8.33884597e-01,\n",
" -4.93242532e-01, -1.71342266e+00, -2.67303872e+00,\n",
" -1.90955043e+00, -8.69829834e-01, -1.80522478e+00,\n",
" -1.91985035e+00, -4.96761203e-01, -5.29186308e-01,\n",
" -3.28707957e+00, 1.35370612e+00, -2.86374390e-01,\n",
" 8.40017557e-01, 1.13399994e+00, 9.98689592e-01,\n",
" 1.74621344e+00, 2.07226321e-01, 6.08601272e-01,\n",
" 9.34968531e-01, 1.59688771e+00, 1.14034593e+00,\n",
" 5.41304529e-01, 3.86263728e+00, 1.54349637e+00,\n",
" 1.91221261e+00, 2.40769243e+00, 3.87992525e+00,\n",
" 3.08658600e+00, 2.41656590e+00, 2.83994389e+00,\n",
" 3.34692717e+00, 1.35415208e+00, 3.82013470e-01,\n",
" -6.26180768e-01, -1.03839242e+00, -7.08445728e-01,\n",
" 1.12548327e+00, 1.95702538e-04, 4.06242907e-01,\n",
" 3.99857616e+00, 1.66841197e+00, 3.90678453e+00,\n",
" 7.76442528e-01, 1.06665039e+00, 1.30748332e-01,\n",
" 2.54443765e-01, 1.04806781e+00, 5.49456000e-01,\n",
" 2.20009089e-02, 3.53348315e-01, 2.00871721e-01,\n",
" -2.23133636e+00, -2.08690691e+00, -1.70284367e+00,\n",
" -1.00411975e+00, -1.55421302e-01, 2.47023374e-01,\n",
" -1.25203884e+00, -2.06214237e+00, -1.49636281e+00,\n",
" -2.02013588e+00, -9.78285074e-01, -1.20530856e+00,\n",
" -1.48999465e+00, -1.03236246e+00, -5.19484699e-01,\n",
" -1.54678226e+00, -2.35162020e+00, 5.99346519e-01,\n",
" -2.67057754e-02, 2.97991663e-01, 2.40855277e-01,\n",
" -3.29041183e-01, 2.68818617e+00, 7.32163429e-01,\n",
" 2.11805630e+00, -1.76404428e+00, -1.42859256e+00,\n",
" -2.94283724e+00, -1.48987889e+00, -1.74058318e+00,\n",
" -1.10061526e+00, 1.05532385e-01, -2.53243208e+00,\n",
" -1.84635425e+00, -2.39082530e-01, 1.09810865e+00,\n",
" -5.90419769e-01, -9.05582964e-01, 2.02747107e-01,\n",
" -1.51103735e+00, -2.06983066e+00, -8.34027886e-01,\n",
" -1.55088413e+00, -8.99944425e-01, 6.79159537e-03,\n",
" -1.85121489e+00, -1.50112903e+00, -1.47589350e+00,\n",
" -1.72867334e+00, -2.10732651e+00, 1.03799403e+00,\n",
" 3.97320914e+00, 2.81198001e+00, 2.30052257e+00,\n",
" 2.30658793e+00, 2.30695629e+00, 2.20866251e+00,\n",
" 2.69618654e+00, 3.07287169e+00, -2.56410748e-01,\n",
" -7.35531032e-01, -2.70364976e+00, -4.32497114e-01,\n",
" 9.19001341e-01, -1.91649902e+00, -1.79021513e+00,\n",
" -1.71204492e-01, -1.60249770e-01, -2.73476934e+00,\n",
" -3.05478573e+00, 9.47945416e-01, 1.71766126e+00,\n",
" 1.82603046e-01, 1.41272521e+00, 2.45719433e-01,\n",
" -5.39486647e-01, -1.89158690e+00, -5.88643312e-01,\n",
" -4.55190614e-02, 6.82070971e-01, 2.08072686e+00,\n",
" -8.67420733e-02, 7.27240324e-01, -4.98709381e-01,\n",
" -1.36455429e+00, -1.21921027e+00, 8.35032165e-01,\n",
" -3.62131625e-01, 1.28271067e+00, 1.07553816e+00,\n",
" 2.00005221e+00, -9.58178282e-01, -1.66674173e+00,\n",
" -2.21298546e-01, -4.47881296e-02, -9.17688847e-01,\n",
" -4.85686570e-01, -8.55692089e-01, -1.34389937e-01,\n",
" 8.78246486e-01, -3.34650922e+00, 2.74024105e+00,\n",
" 2.28237224e+00, -2.14688873e+00, 1.72798470e-01,\n",
" -1.25110471e+00, -1.08559489e+00, 1.32716095e+00,\n",
" 4.34288949e-01, 5.51602602e-01, -4.25514281e-01,\n",
" -3.03428221e+00, 3.32969218e-01, -5.85350931e-01,\n",
" -1.01127481e+00, -1.13758612e+00, 2.99821973e+00,\n",
" 2.36347169e-01, -3.05448294e-01, 2.52388477e+00,\n",
" 5.11164248e-01, 4.82690275e-01, 1.82521828e-02,\n",
" 1.69068468e+00, 2.53376842e+00, 4.07392919e-01,\n",
" -2.90624261e+00, -2.62598562e+00, -1.50837755e+00,\n",
" 1.15478611e+00, 4.62593555e-01, -3.77383918e-01,\n",
" -4.54136401e-01, -5.01974821e-01, -6.60325885e-01,\n",
" 1.64890563e+00, -9.92922068e-01, -4.37342882e+00,\n",
" -1.76332748e+00, 8.67795527e-01, -1.22817285e-01,\n",
" 8.17923486e-01, 2.62077355e+00, -5.14370620e-01,\n",
" 1.42861617e+00, -8.89410079e-01, 2.82288074e-01,\n",
" 2.86746889e-01, -1.59284472e-01, 3.06442451e+00,\n",
" 3.08767509e+00, 1.10410511e+00, 5.84134638e-01,\n",
" 9.96468961e-01, -1.60095942e+00, 8.00808489e-01,\n",
" -1.17358029e+00, -1.73136264e-01, 1.14211357e+00,\n",
" -6.64759398e-01, -1.38795450e-01, -9.28050578e-02,\n",
" -1.00550234e+00, 1.83907902e+00, -7.27077186e-01,\n",
" -3.25605959e-01, 1.27522230e-01, 8.20962429e-01,\n",
" 1.53685296e+00, 1.07783592e+00, -6.43890500e-01,\n",
" -2.69787741e+00, -2.53440714e+00, 5.89859128e-01,\n",
" -5.44066668e-01, 4.96726274e-01, -9.72725898e-02,\n",
" -1.09674740e+00, -9.26089168e-01, -1.03492486e+00,\n",
" 5.92053495e-02, -1.72262740e+00, 2.95496881e-02,\n",
" -1.52114466e-01, 1.88301849e+00, -1.24823391e+00,\n",
" -6.95580065e-01, -1.09528470e+00, -1.11678982e+00,\n",
" 1.61060169e-01, -5.14116228e-01, 1.14424551e+00,\n",
" -1.17884445e+00, -1.62069273e+00, 1.15921688e+00,\n",
" -1.93254483e+00, -1.18327582e+00, -1.61414790e+00,\n",
" -9.07000065e-01, 1.40650189e+00, 3.51846385e+00,\n",
" -2.22942972e+00, 6.23425376e-03, 1.30355918e+00,\n",
" -2.59207189e-01, 1.41317892e+00, 1.21975076e+00,\n",
" -5.61854523e-03, -2.80561876e+00, -1.93194425e+00,\n",
" 1.79248798e+00, -3.49373937e+00, -1.27135968e+00,\n",
" 9.39924896e-01, -1.80615366e-01, -2.32698250e+00,\n",
" -3.05012751e+00, -6.95001602e-01, -5.00427067e-01,\n",
" 5.66623509e-02, 1.73128045e+00, 9.36404228e-01,\n",
" -1.75694168e+00, 2.81710553e+00, 1.46450055e+00,\n",
" -1.25955689e+00, -5.08831799e-01, -4.34357561e-02,\n",
" 6.77392364e-01, -1.76032722e+00, -1.51608109e+00,\n",
" 6.83466315e-01, -8.54872346e-01, 5.87889493e-01,\n",
" -1.40195870e+00, 2.58034039e+00, -1.87670493e+00,\n",
" -1.60772192e+00, 3.85116905e-01, -6.16014421e-01,\n",
" -2.32660726e-01, -6.02841139e-01, 4.28686857e-01,\n",
" -1.56341994e+00, -1.86496055e+00, -1.05299592e+00,\n",
" -7.12690651e-01, -2.46136498e+00, 1.86990905e+00,\n",
" -5.13823986e-01, 5.03967345e-01, -1.76901722e+00,\n",
" -2.19511509e-01, -7.83396810e-02, 7.94978261e-01,\n",
" 1.37682825e-01, -1.27755415e+00, 6.63644612e-01,\n",
" 9.54980791e-01, 1.97708070e-01, 1.87124002e+00,\n",
" 1.87821043e+00, -4.69974428e-01, -1.12402463e+00,\n",
" -3.18121600e+00, 1.03094232e+00, 7.65274763e-01,\n",
" 2.62565637e+00, -9.58706796e-01, 1.93622518e+00,\n",
" -2.33188128e+00, -4.79165167e-01, 6.44829988e+00,\n",
" -2.47202611e+00, 1.26634550e+00, -1.30976355e+00,\n",
" 1.13209270e-01, 2.43186593e+00, -1.29538023e+00,\n",
" 3.66333872e-01, 1.34910965e+00, 7.76115537e-01,\n",
" 1.77355200e-01, 1.99346089e+00, 2.71920621e-01,\n",
" -1.00989282e+00, -1.98029423e+00, 2.27880776e-01,\n",
" -2.38420868e+00, -1.31710851e+00, -6.37629986e-01,\n",
" 6.39242709e-01, 7.14427680e-02, 2.82906055e-01,\n",
" 1.63520253e+00, -3.16152573e+00, -4.16788489e-01,\n",
" -5.05552471e-01, 1.47144735e+00, 1.66715786e-01,\n",
" 7.78935291e-03, 3.49167609e+00, -7.94411421e-01,\n",
" -1.72711682e+00, 4.59302485e-01, 3.55083513e+00,\n",
" -2.33258295e+00, 1.70016158e+00, 8.90555561e-01,\n",
" -3.42013299e-01, 4.12573308e-01, 1.19436765e+00,\n",
" -1.20788920e+00, -2.30027962e+00, 9.88094449e-01,\n",
" -1.95416367e+00, 7.93366432e-01, -1.63555586e+00,\n",
" -4.73215967e-01, 1.20251060e+00, 5.61390698e-01,\n",
" -8.28601599e-01, -2.05484200e+00, 7.42964745e-01,\n",
" -1.47961900e-01, 1.38295853e+00, 1.28272176e-01,\n",
" 8.96114349e-01, 2.00291246e-01, -6.04499653e-02,\n",
" 8.73708367e-01, -1.35136700e+00, -3.32257837e-01,\n",
" -8.05843771e-02, 1.58672857e+00, 3.06422234e+00,\n",
" -2.50524282e+00, -1.93878329e+00, 3.45390892e+00,\n",
" -1.27112937e+00, 1.42587197e+00, -5.96717358e-01,\n",
" 4.46699917e-01, 4.72907752e-01, 2.88197566e-02,\n",
" 1.16810751e+00, 9.55489874e-02, -1.48410499e-01,\n",
" -2.72217417e+00, 1.33178687e+00, -1.65416276e+00,\n",
" -3.28158069e+00, -2.59980820e-02, 1.23474574e+00,\n",
" -1.94041923e-01, -1.47240829e+00, 3.32517505e+00,\n",
" 2.01005554e+00, -1.15649569e+00, 7.00166374e-02,\n",
" -1.09931970e+00, 1.93688691e+00, 3.34086776e-01,\n",
" -1.37803209e+00, 7.59569705e-01, 2.08346963e+00,\n",
" -2.47377157e+00, 7.11220920e-01, 1.10764730e+00,\n",
" -1.00727522e+00, -8.94211233e-02, 1.16281521e+00,\n",
" 1.28872931e+00, 2.33569527e+00, -2.68954086e+00,\n",
" 1.68462288e+00, -1.65096605e+00, 8.13466251e-01,\n",
" -2.57265878e+00, 8.13374281e-01, 5.00084341e-01,\n",
" 2.21961784e+00, 4.90845710e-01, 3.52778387e+00,\n",
" 1.06868827e+00, -1.39897001e+00, 3.33065540e-01,\n",
" -1.66987145e+00, 1.30672359e+00, 2.47979403e+00,\n",
" -2.65644789e+00, 4.72160876e-01, 4.88310277e-01,\n",
" -1.79450822e+00, 2.26955843e+00, -3.41013288e+00,\n",
" 1.58357954e+00, 1.73186862e+00, 2.02063274e+00,\n",
" -4.41434383e-01, -2.28582644e+00, 6.93113327e-01,\n",
" 3.23371339e+00, -1.13124955e+00, 3.21346074e-02,\n",
" -6.50638521e-01, -8.73156011e-01, 2.73187049e-02,\n",
" 8.38176906e-02, -4.76620868e-02, 1.66702068e+00,\n",
" 1.05954051e+00, -1.89402997e+00, 1.87468922e+00,\n",
" 2.03380442e+00, -1.85234264e-01, -1.19531035e+00,\n",
" 6.52794421e-01, -2.38493586e+00, 9.01854634e-01,\n",
" -5.86835563e-01, -4.04116124e-01, 7.44021416e-01,\n",
" -1.87741232e+00, 1.50937176e+00, 7.37908110e-02,\n",
" -5.43003559e-01, -1.89494491e+00, 1.88077593e+00,\n",
" -8.33676219e-01, 1.92328608e+00, 1.71817744e+00,\n",
" 1.40228546e+00, 8.00139368e-01, 1.60305774e+00,\n",
" 1.50285590e+00, 1.03710675e+00, -1.29266623e-02,\n",
" 1.65807104e+00, 4.58535403e-01, -3.08411777e-01,\n",
" -1.80658817e+00, -1.26245463e+00, 1.79194403e+00,\n",
" 3.05525565e+00, 1.90686536e+00, -2.98485827e+00,\n",
" 2.01937795e-01, -7.63000667e-01, -6.01673186e-01,\n",
" -1.70703030e+00, -7.26590633e-01, 2.29800129e+00,\n",
" 1.67319036e+00, -4.77491140e-01, 1.84200013e+00,\n",
" 2.43693635e-01, -8.34791481e-01, -3.37798089e-01,\n",
" -9.48283792e-01, -6.22405946e-01, -1.78150237e+00,\n",
" 1.23711586e+00, -1.37317026e+00, -1.15404069e+00,\n",
" 1.58743632e+00, 1.58258605e+00, 1.84869957e+00,\n",
" -2.27265191e+00, -6.47241950e-01, 7.16710567e-01,\n",
" -3.35976958e+00, 2.10211921e+00, -1.25926805e+00,\n",
" 1.00763321e-01, -4.54020917e-01, 1.45979857e+00,\n",
" -2.53317326e-01, -3.45974612e+00, 2.59195352e+00,\n",
" -5.87325513e-01, -2.71071696e+00, -1.14105523e+00,\n",
" -2.00959516e+00, -4.72351164e-01, 5.33669829e-01,\n",
" 4.06048834e-01, 1.26865700e-01, -8.99718761e-01,\n",
" 2.95253348e+00, -1.92177546e+00, -1.82954943e+00,\n",
" -1.00271082e+00, -1.15991127e+00, -7.15413749e-01,\n",
" -6.08443797e-01, 1.54210913e+00, 3.29932868e-01,\n",
" 6.46673322e-01, -3.16645116e-01, -5.44970214e-01,\n",
" 1.33709788e+00, -3.31854522e-01, -2.52866093e-02,\n",
" 7.03483284e-01, 3.39909482e+00, 4.95852089e+00,\n",
" 5.70806086e-01, -3.37446952e+00, -3.59201618e-03,\n",
" -1.39897689e-01, -3.36632341e-01, 1.90125847e+00,\n",
" 3.59996223e+00, 6.19568408e-01, -6.32025659e-01,\n",
" 1.50538492e+00, -4.28654337e+00, -1.53590822e+00,\n",
" 1.20004463e+00, 6.33485079e-01, -1.99690461e+00,\n",
" 2.58403361e-01, -1.04410827e+00, -1.48701441e+00,\n",
" -1.92449307e+00, -2.66258454e+00, -9.56860125e-01,\n",
" -1.29206252e+00, 9.42504823e-01, -3.58749032e-01,\n",
" 1.03755164e+00, -1.62184131e+00, 1.59799910e+00,\n",
" 5.54946810e-02, -3.81573486e+00, -8.37609708e-01,\n",
" 2.55720727e-02, -9.65045571e-01, -2.72325826e+00,\n",
" 3.14082050e+00, -2.13606939e-01, 1.08747587e-01,\n",
" 1.64792323e+00, -7.40568489e-02, 7.32264221e-01,\n",
" 1.30035400e+00, -1.44593418e-01, -1.24217224e+00,\n",
" 1.84440557e-02, -5.87913513e-01, -2.76239157e-01,\n",
" -9.74993825e-01, 3.23910880e+00, 1.56754959e+00,\n",
" 1.63032234e-01, -2.26810980e+00, 3.08372080e-01,\n",
" -1.26275218e+00, 2.38262820e+00, 1.77280918e-01,\n",
" -6.12160504e-01, -1.18261969e+00, 1.06038773e+00,\n",
" -6.61281884e-01, 2.56689310e+00, -4.24892366e-01,\n",
" -1.60467124e+00, 9.23135281e-01, 5.09784281e-01,\n",
" -2.36816093e-01, 5.49685061e-01, 3.72588813e-01,\n",
" -7.05751538e-01, 2.24120522e+00, -2.93325353e+00,\n",
" -4.03847575e-01, -1.17595446e+00, -3.13691795e-01,\n",
" -5.22548445e-02, 4.02303159e-01, 5.13418317e-01,\n",
" -3.38978544e-02, -8.05717766e-01, -2.12521458e+00,\n",
" -1.44194162e+00, -7.00470507e-01, -2.57672811e+00,\n",
" 7.86034912e-02, -1.48267210e+00, 8.90361249e-01,\n",
" -7.15771854e-01, -3.80829461e-02, 4.78945732e-01,\n",
" -1.30621874e+00, -1.69138300e+00, -7.94364512e-01,\n",
" -1.53714645e+00, 1.80786395e+00, -9.40447330e-01,\n",
" -1.76979446e+00, -7.40339100e-01, -2.98443532e+00,\n",
" -1.34205759e+00, -1.67450964e+00, -2.46101165e+00,\n",
" -5.52049518e-01, -1.95667374e+00, 8.24635148e-01,\n",
" 3.13193351e-01, 2.35134795e-01, 7.60618389e-01,\n",
" 1.55331349e+00, 7.76964307e-01, -1.99027658e+00,\n",
" -3.07744354e-01, 2.12018919e+00, -2.81869531e+00,\n",
" -3.04112345e-01, -1.98528573e-01, -1.90984106e+00,\n",
" -1.21275973e+00, 5.06801844e-01, -8.89652550e-01,\n",
" 6.22948468e-01, -2.17541170e+00, -3.79369545e+00,\n",
" -6.77980036e-02, 1.37057900e+00, 6.11521482e-01,\n",
" 5.70312917e-01, -1.52763176e+00, 3.89497638e+00,\n",
" -4.54609171e-02, 3.64303732e+00, 6.63903713e-01,\n",
" 1.66248333e+00, 7.42205322e-01, 3.17934799e+00,\n",
" 2.14342809e+00, -2.61262152e-02, 2.03321719e+00,\n",
" 8.19472671e-02, 2.73765445e-01, 3.76390529e+00,\n",
" 1.59334075e+00, 9.30227816e-01, 2.49806151e-01,\n",
" 2.30522466e+00, 1.06439626e+00, -6.16410375e-01,\n",
" 3.22261155e-01, -4.35784191e-01, 8.28886867e-01,\n",
" -3.44528270e+00, -2.20949578e+00, -1.76651394e+00,\n",
" -1.59534466e+00, 1.17536557e+00, 7.72589147e-01,\n",
" 1.66956329e+00]], dtype=float32)"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"tvm_output"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"tensor([[[[ 1.7219e-01, -4.9744e-01, 1.9181e+00, ..., 3.4058e-01,\n",
" 2.3876e+00, -6.2053e-01],\n",
" [ 1.2227e-02, 2.4274e+00, 2.8559e-02, ..., -4.6603e-01,\n",
" -7.3615e-01, 6.9074e-01],\n",
" [-7.2072e-01, -1.7811e+00, -6.8173e-01, ..., 2.3840e+00,\n",
" 2.5641e-02, -1.0845e-01],\n",
" ...,\n",
" [ 1.0349e+00, -9.1062e-01, 3.8786e-01, ..., -3.3641e-01,\n",
" -6.2294e-01, -7.1930e-01],\n",
" [ 7.6752e-01, 8.4012e-01, 2.4716e-01, ..., 3.7769e-01,\n",
" -1.9019e+00, 1.1545e-01],\n",
" [ 1.3290e+00, -3.9752e-01, -1.1462e+00, ..., -5.1818e-01,\n",
" 1.9632e+00, 2.5788e+00]],\n",
"\n",
" [[-7.2132e-01, 3.2107e-01, -5.3535e-01, ..., 2.2049e-03,\n",
" -2.5831e-01, -7.9573e-01],\n",
" [-9.4956e-01, -9.1265e-02, 1.9193e+00, ..., 1.5525e+00,\n",
" -4.9848e-02, 2.9720e-01],\n",
" [ 5.8233e-01, 9.6106e-01, -6.4952e-01, ..., 3.1290e-01,\n",
" 3.2880e-01, -1.5676e+00],\n",
" ...,\n",
" [ 2.0057e+00, 7.6539e-01, -6.7423e-01, ..., -1.8337e-01,\n",
" 3.3970e-01, 2.8306e-01],\n",
" [ 7.6145e-01, -3.3815e-01, 2.7822e-01, ..., -4.0741e-01,\n",
" 6.5746e-01, -7.9156e-01],\n",
" [-1.2608e+00, 1.2614e+00, 2.0711e-01, ..., -1.2500e+00,\n",
" 8.3432e-01, -2.9766e-01]],\n",
"\n",
" [[ 1.2557e+00, -8.8738e-01, 3.8425e-01, ..., 3.2118e-01,\n",
" 2.2620e-01, 3.2907e-01],\n",
" [-6.8088e-01, -1.8085e+00, -1.2791e-01, ..., -1.6832e-01,\n",
" 9.0862e-01, -1.0912e+00],\n",
" [-1.9298e+00, -1.1261e+00, -3.1442e-01, ..., -3.6195e-02,\n",
" 1.3986e-01, 1.0616e+00],\n",
" ...,\n",
" [-4.0836e-03, -7.4391e-01, 8.2932e-01, ..., 2.0618e+00,\n",
" 2.8723e-02, 3.2699e-01],\n",
" [-5.1492e-01, -7.4563e-01, -7.9411e-01, ..., -1.4049e+00,\n",
" -1.5328e-01, -1.1750e+00],\n",
" [ 6.4362e-01, 5.0400e-01, -1.4239e+00, ..., 8.7916e-01,\n",
" -4.5287e-01, -1.2845e+00]]]])"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"input_data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Checking VTA operations\n",
"\n",
"It seems that there is no operations sent to VTA."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [],
"source": [
"# Graph runtime\n",
"m = graph_runtime.GraphModule(lib[\"default\"](ctx))"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [],
"source": [
"m.set_input(input_name, tvm.nd.array(input_data))"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [],
"source": [
"# Perform inference and gather execution statistics\n",
"# More on: :py:method:`tvm.runtime.Module.time_evaluator`\n",
"num = 4 # number of times we run module for a single measurement\n",
"rep = 3 # number of measurements (we derive std dev from this)\n",
"timer = m.module.time_evaluator(\"run\", ctx, number=num, repeat=rep)\n"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"Execution statistics:\n",
"\tinp_load_nbytes : 0\n",
"\twgt_load_nbytes : 0\n",
"\tacc_load_nbytes : 0\n",
"\tuop_load_nbytes : 0\n",
"\tout_store_nbytes: 0\n",
"\tgemm_counter : 0\n",
"\talu_counter : 0\n"
]
}
],
"source": [
"if env.TARGET in [\"sim\", \"tsim\"]:\n",
" simulator.clear_stats()\n",
" timer()\n",
" sim_stats = simulator.stats()\n",
" print(\"\\nExecution statistics:\")\n",
" for k, v in sim_stats.items():\n",
" # Since we execute the workload many times, we need to normalize stats\n",
" # Note that there is always one warm up run\n",
" # Therefore we divide the overall stats by (num * rep + 1)\n",
" print(\"\\t{:<16}: {:>16}\".format(k, v // (num * rep + 1)))\n",
"else:\n",
" tcost = timer()\n",
" std = np.std(tcost.results) * 1000\n",
" mean = tcost.mean * 1000\n",
" print(\"\\nPerformed inference in %.2fms (std = %.2f) for %d samples\" % (mean, std, env.BATCH))\n",
" print(\"Average per sample inference time: %.2fms\" % (mean / env.BATCH))\n"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.7"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment