Alexnet_no_optim

graph(%0 : Float(1, 3, 224, 224)
      %1 : Float(64, 3, 11, 11)
      %2 : Float(64)
      %3 : Float(192, 64, 5, 5)
      %4 : Float(192)
      %5 : Float(384, 192, 3, 3)
      %6 : Float(384)
      %7 : Float(256, 384, 3, 3)
      %8 : Float(256)
      %9 : Float(256, 256, 3, 3)
      %10 : Float(256)
      %11 : Float(4096, 9216)
      %12 : Float(4096)
      %13 : Float(4096, 4096)
      %14 : Float(4096)
      %15 : Float(1000, 4096)
      %16 : Float(1000)) {
  %17 : Float(1, 64, 55, 55), %18 : Float(1, 363, 3025), %19 : Float(0) = thnn_conv2d_forward[kernel_size=[11, 11], stride=[4, 4], padding=[2, 2]](%0, %1, %2), scope: AlexNet/Sequential[features]/Conv2d[0]
  %20 : Float(1, 64, 55, 55) = thnn_conv2d[kernel_size=[11, 11], stride=[4, 4], padding=[2, 2]](%0, %1, %2), scope: AlexNet/Sequential[features]/Conv2d[0]
  %21 : Float(1, 64, 55, 55) = _convolution_nogroup[stride=[4, 4], padding=[2, 2], dilation=[1, 1], transposed=0, output_padding=[0, 0]](%0, %1, %2), scope: AlexNet/Sequential[features]/Conv2d[0]
  %22 : Float(1, 64, 55, 55) = _convolution[stride=[4, 4], padding=[2, 2], dilation=[1, 1], transposed=0, output_padding=[0, 0], groups=1, benchmark=0, deterministic=0, cudnn_enabled=1](%0, %1, %2), scope: AlexNet/Sequential[features]/Conv2d[0]
  %23 : Float(1, 64, 55, 55) = threshold_forward[threshold={0}, value={0}](%22), scope: AlexNet/Sequential[features]/ReLU[1]
  %24 : Float(1, 64, 55, 55) = threshold[threshold={0}, value={0}](%23), scope: AlexNet/Sequential[features]/ReLU[1]
  %25 : Float(1, 64, 27, 27), %26 : Long(1, 64, 27, 27) = max_pool2d_forward[kernel_size=[3, 3], stride=[2, 2], padding=[0, 0], dilation=[1, 1], ceil_mode=0](%24), scope: AlexNet/Sequential[features]/MaxPool2d[2]
  %27 : Float(1, 64, 27, 27), %28 : Long(1, 64, 27, 27) = max_pool2d[kernel_size=[3, 3], stride=[2, 2], padding=[0, 0], dilation=[1, 1], ceil_mode=0](%24), scope: AlexNet/Sequential[features]/MaxPool2d[2]
  %29 : Float(1, 192, 27, 27), %30 : Float(1, 1600, 729), %31 : Float(0) = thnn_conv2d_forward[kernel_size=[5, 5], stride=[1, 1], padding=[2, 2]](%27, %3, %4), scope: AlexNet/Sequential[features]/Conv2d[3]
  %32 : Float(1, 192, 27, 27) = thnn_conv2d[kernel_size=[5, 5], stride=[1, 1], padding=[2, 2]](%27, %3, %4), scope: AlexNet/Sequential[features]/Conv2d[3]
  %33 : Float(1, 192, 27, 27) = _convolution_nogroup[stride=[1, 1], padding=[2, 2], dilation=[1, 1], transposed=0, output_padding=[0, 0]](%27, %3, %4), scope: AlexNet/Sequential[features]/Conv2d[3]
  %34 : Float(1, 192, 27, 27) = _convolution[stride=[1, 1], padding=[2, 2], dilation=[1, 1], transposed=0, output_padding=[0, 0], groups=1, benchmark=0, deterministic=0, cudnn_enabled=1](%27, %3, %4), scope: AlexNet/Sequential[features]/Conv2d[3]
  %35 : Float(1, 192, 27, 27) = threshold_forward[threshold={0}, value={0}](%34), scope: AlexNet/Sequential[features]/ReLU[4]
  %36 : Float(1, 192, 27, 27) = threshold[threshold={0}, value={0}](%35), scope: AlexNet/Sequential[features]/ReLU[4]
  %37 : Float(1, 192, 13, 13), %38 : Long(1, 192, 13, 13) = max_pool2d_forward[kernel_size=[3, 3], stride=[2, 2], padding=[0, 0], dilation=[1, 1], ceil_mode=0](%36), scope: AlexNet/Sequential[features]/MaxPool2d[5]
  %39 : Float(1, 192, 13, 13), %40 : Long(1, 192, 13, 13) = max_pool2d[kernel_size=[3, 3], stride=[2, 2], padding=[0, 0], dilation=[1, 1], ceil_mode=0](%36), scope: AlexNet/Sequential[features]/MaxPool2d[5]
  %41 : Float(1, 384, 13, 13), %42 : Float(1, 1728, 169), %43 : Float(0) = thnn_conv2d_forward[kernel_size=[3, 3], stride=[1, 1], padding=[1, 1]](%39, %5, %6), scope: AlexNet/Sequential[features]/Conv2d[6]
  %44 : Float(1, 384, 13, 13) = thnn_conv2d[kernel_size=[3, 3], stride=[1, 1], padding=[1, 1]](%39, %5, %6), scope: AlexNet/Sequential[features]/Conv2d[6]
  %45 : Float(1, 384, 13, 13) = _convolution_nogroup[stride=[1, 1], padding=[1, 1], dilation=[1, 1], transposed=0, output_padding=[0, 0]](%39, %5, %6), scope: AlexNet/Sequential[features]/Conv2d[6]
  %46 : Float(1, 384, 13, 13) = _convolution[stride=[1, 1], padding=[1, 1], dilation=[1, 1], transposed=0, output_padding=[0, 0], groups=1, benchmark=0, deterministic=0, cudnn_enabled=1](%39, %5, %6), scope: AlexNet/Sequential[features]/Conv2d[6]
  %47 : Float(1, 384, 13, 13) = threshold_forward[threshold={0}, value={0}](%46), scope: AlexNet/Sequential[features]/ReLU[7]
  %48 : Float(1, 384, 13, 13) = threshold[threshold={0}, value={0}](%47), scope: AlexNet/Sequential[features]/ReLU[7]
  %49 : Float(1, 256, 13, 13), %50 : Float(1, 3456, 169), %51 : Float(0) = thnn_conv2d_forward[kernel_size=[3, 3], stride=[1, 1], padding=[1, 1]](%48, %7, %8), scope: AlexNet/Sequential[features]/Conv2d[8]
  %52 : Float(1, 256, 13, 13) = thnn_conv2d[kernel_size=[3, 3], stride=[1, 1], padding=[1, 1]](%48, %7, %8), scope: AlexNet/Sequential[features]/Conv2d[8]
  %53 : Float(1, 256, 13, 13) = _convolution_nogroup[stride=[1, 1], padding=[1, 1], dilation=[1, 1], transposed=0, output_padding=[0, 0]](%48, %7, %8), scope: AlexNet/Sequential[features]/Conv2d[8]
  %54 : Float(1, 256, 13, 13) = _convolution[stride=[1, 1], padding=[1, 1], dilation=[1, 1], transposed=0, output_padding=[0, 0], groups=1, benchmark=0, deterministic=0, cudnn_enabled=1](%48, %7, %8), scope: AlexNet/Sequential[features]/Conv2d[8]
  %55 : Float(1, 256, 13, 13) = threshold_forward[threshold={0}, value={0}](%54), scope: AlexNet/Sequential[features]/ReLU[9]
  %56 : Float(1, 256, 13, 13) = threshold[threshold={0}, value={0}](%55), scope: AlexNet/Sequential[features]/ReLU[9]
  %57 : Float(1, 256, 13, 13), %58 : Float(1, 2304, 169), %59 : Float(0) = thnn_conv2d_forward[kernel_size=[3, 3], stride=[1, 1], padding=[1, 1]](%56, %9, %10), scope: AlexNet/Sequential[features]/Conv2d[10]
  %60 : Float(1, 256, 13, 13) = thnn_conv2d[kernel_size=[3, 3], stride=[1, 1], padding=[1, 1]](%56, %9, %10), scope: AlexNet/Sequential[features]/Conv2d[10]
  %61 : Float(1, 256, 13, 13) = _convolution_nogroup[stride=[1, 1], padding=[1, 1], dilation=[1, 1], transposed=0, output_padding=[0, 0]](%56, %9, %10), scope: AlexNet/Sequential[features]/Conv2d[10]
  %62 : Float(1, 256, 13, 13) = _convolution[stride=[1, 1], padding=[1, 1], dilation=[1, 1], transposed=0, output_padding=[0, 0], groups=1, benchmark=0, deterministic=0, cudnn_enabled=1](%56, %9, %10), scope: AlexNet/Sequential[features]/Conv2d[10]
  %63 : Float(1, 256, 13, 13) = threshold_forward[threshold={0}, value={0}](%62), scope: AlexNet/Sequential[features]/ReLU[11]
  %64 : Float(1, 256, 13, 13) = threshold[threshold={0}, value={0}](%63), scope: AlexNet/Sequential[features]/ReLU[11]
  %65 : Float(1, 256, 6, 6), %66 : Long(1, 256, 6, 6) = max_pool2d_forward[kernel_size=[3, 3], stride=[2, 2], padding=[0, 0], dilation=[1, 1], ceil_mode=0](%64), scope: AlexNet/Sequential[features]/MaxPool2d[12]
  %67 : Float(1, 256, 6, 6), %68 : Long(1, 256, 6, 6) = max_pool2d[kernel_size=[3, 3], stride=[2, 2], padding=[0, 0], dilation=[1, 1], ceil_mode=0](%64), scope: AlexNet/Sequential[features]/MaxPool2d[12]
  %69 : Float(1, 9216) = view[size=[1, 9216]](%67), scope: AlexNet
  %70 : Float(1, 9216), %71 : Handle = ^Dropout(0.5, False, False)(%69), scope: AlexNet/Sequential[classifier]/Dropout[0]
  %72 : Float(9216!, 4096!) = t(%11), scope: AlexNet/Sequential[classifier]/Linear[1]
  %73 : Float(1, 4096) = expand[size=[1, 4096]](%12), scope: AlexNet/Sequential[classifier]/Linear[1]
  %74 : Float(1, 4096) = addmm[beta={1}, alpha={1}](%73, %70, %72), scope: AlexNet/Sequential[classifier]/Linear[1]
  %75 : Float(1, 4096) = threshold_forward[threshold={0}, value={0}](%74), scope: AlexNet/Sequential[classifier]/ReLU[2]
  %76 : Float(1, 4096) = threshold[threshold={0}, value={0}](%75), scope: AlexNet/Sequential[classifier]/ReLU[2]
  %77 : Float(1, 4096), %78 : Handle = ^Dropout(0.5, False, False)(%76), scope: AlexNet/Sequential[classifier]/Dropout[3]
  %79 : Float(4096!, 4096!) = t(%13), scope: AlexNet/Sequential[classifier]/Linear[4]
  %80 : Float(1, 4096) = expand[size=[1, 4096]](%14), scope: AlexNet/Sequential[classifier]/Linear[4]
  %81 : Float(1, 4096) = addmm[beta={1}, alpha={1}](%80, %77, %79), scope: AlexNet/Sequential[classifier]/Linear[4]
  %82 : Float(1, 4096) = threshold_forward[threshold={0}, value={0}](%81), scope: AlexNet/Sequential[classifier]/ReLU[5]
  %83 : Float(1, 4096) = threshold[threshold={0}, value={0}](%82), scope: AlexNet/Sequential[classifier]/ReLU[5]
  %84 : Float(4096!, 1000!) = t(%15), scope: AlexNet/Sequential[classifier]/Linear[6]
  %85 : Float(1, 1000) = expand[size=[1, 1000]](%16), scope: AlexNet/Sequential[classifier]/Linear[6]
  %86 : Float(1, 1000) = addmm[beta={1}, alpha={1}](%85, %83, %84), scope: AlexNet/Sequential[classifier]/Linear[6]
  return (%86);
}