fengxiaoshuai Goto Github PK

3d-headtrack

3D-headtrack

500lines

500 Lines or Less

cnn_model_optimizer

some ways to optimize CNN model

codesamples

Code samples for the CUDA tutorial "CUDA and Applications to Task-based Programming"

cuassembler

An unofficial cuda assembler, for all generations of SASS, hopefully ：）

cuda_decoder

cuda_decoder_op

cuda_encoder

cuda_gemm

A simple high performance CUDA GEMM implementation.

cutt

CUDA Tensor Transpose (cuTT) library

docs

Documentations for PaddlePaddle

flash-attention-minimal

Flash Attention in ~100 lines of CUDA (forward pass only)

gemm_wmma

GEMM by WMMA (tensor core)

halide

a language for fast, portable data-parallel computation

hellogithub

:octocat: 分享 GitHub 上有趣、入门级的开源项目。Share interesting, entry-level open source projects on GitHub.

hellogithub.com

HelloGitHub.com 网站源码

how_to_optimize_in_gpu

This is a series of GPU optimization topics. Here we will introduce how to optimize the CUDA kernel in detail. I will introduce several basic kernel optimizations, including: elementwise, reduce, sgemv, sgemm, etc. The performance of these kernels is basically at or near the theoretical limit.

inference_benchmark

kernel_memory_management

总结整理linux内核的内存管理的资料，包含论文，文章，视频，以及应用程序的内存泄露，内存池相关

messy_test

C++/C++11's usage

models

Pre-trained and Reproduced Deep Learning Models （『飞桨』官方模型库，包含多种学术前沿和工业场景验证的深度学习模型）

ompi

Open MPI main development repository

oneflow

OneFlow is a performance-centered and open-source deep learning framework.

paddle

PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）

paddle-inference-demo

paddlefleetx

Paddle Distributed Training Examples. 飞桨分布式训练示例 Resnet Bert GPT MOE DataParallel ModelParallel PipelineParallel HybridParallel AutoParallel Zero Sharding Recompute GradientMerge Offload AMP DGC LocalSGD Wide&Deep

paddleocr

Awesome multilingual OCR toolkits based on PaddlePaddle （practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices）

paddleutils

Some tools to operate PaddlePaddle model

ppl.kernel.cuda

ppl.nn

A primitive library for neural network