tiandiao123

import inspect
from typing import Any, Callable, Dict, List, Optional, Tuple, Union

import PIL.Image
import torch
from transformers import CLIPTextModel, CLIPTextModelWithProjection, CLIPTokenizer
from diffusers import StableDiffusionXLImg2ImgPipeline

from diffusers.image_processor import PipelineImageInput, VaeImageProcessor
from diffusers.loaders import FromSingleFileMixin, StableDiffusionXLLoraLoaderMixin, TextualInversionLoaderMixin

tiandiao123

#include <cublas_v2.h>
#include <cstdint>
#include <cuda.h>
#include <cuda_runtime.h>
#include <cuda_fp16.h>
#include <iostream>
#include <torch/torch.h>
#include <torch/types.h>
#include <c10/util/Half.h>

tiandiao123

#include <cublas_v2.h>
#include <cstdint>
#include <cuda.h>
#include <cuda_runtime.h>
#include <cuda_fp16.h>
#include <iostream>
#include <torch/torch.h>
#include <torch/types.h>
#include <c10/util/Half.h>

tiandiao123

#include <cublas_v2.h>
#include <cstdint>
#include <cuda.h>
#include <cuda_runtime.h>
#include <cuda_fp16.h>
#include <iostream>
#include <torch/torch.h>
#include <torch/types.h>
#include <c10/util/Half.h>

tiandiao123

#include <cublas_v2.h>
#include <cstdint>
#include <cuda.h>
#include <cuda_runtime.h>
#include <cuda_fp16.h>
#include <iostream>
#include <torch/torch.h>

#include "cutlass/cutlass.h"
#include "cutlass/gemm/device/gemm_splitk_parallel.h"

tiandiao123

#include <torch/extension.h>
#include <cutlass/gemm/gemm.h>
#include <cutlass/epilogue/thread/linear_combination.h>

torch::Tensor bmm_fp16_fp16_f32(torch::Tensor A, torch::Tensor B, float alpha) {
    int batch_size = A.size(0);
    int M = A.size(1);
    int N = B.size(1);
    int K = A.size(2);

tiandiao123

#include <iostream>
#include "cutlass/cutlass.h"
#include "cutlass/gemm/device/gemm.h"
#include "cutlass/util/host_tensor.h"
#include "cutlass/util/reference/host/tensor_compare.h"
#include "cutlass/util/reference/host/tensor_copy.h"
#include "cutlass/util/reference/host/tensor_fill.h"
#include "cutlass/util/tensor_view_io.h"
#include "helper.h"

tiandiao123

import torch
from inference import CaiInferenceConfig, convert_to_ds_model, recover_from_ds_model
from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer
from argparse import ArgumentParser
import time
import torch
from torch.profiler import profile, record_function, ProfilerActivity

parser = ArgumentParser()
parser.add_argument("--name", default="bigscience/bloom-560m", type=str, help="model_name")

tiandiao123

import torch
import time
from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer
from argparse import ArgumentParser
from transformers import LlamaForCausalLM, LlamaTokenizer
from inference import CaiInferenceConfig, convert_to_ds_model, recover_from_ds_model
from torch.profiler import profile, record_function, ProfilerActivity
from types import MethodType
from typing import Optional, Sequence, Tuple, Union
import torch

tiandiao123

import pytest
import torch

import triton
import triton.language as tl


@triton.jit
def max_fn(x, y):
    return tl.math.max(x, y)