Xiaoyu Zhang

how-to-optim-algorithm-in-cuda. how to optimize some algorithm in cuda.

tvm_mlir_learn. compiler learning resources collect.

Image-processing-algorithm. paper implement

how-to-learn-deep-learning-framework. how to learn PyTorch and OneFlow

Darknet. Source-code notes for AlexeyAB Darknet (legacy)

Keras-Semantic-Segmentation. Keras-Semantic-Segmentation

Image-processing-algorithm-Speed. opencv

model-compression. model compression based on pytorch (1、quantization: 8/4/2bits(dorefa)、ternary/binary value(twn/bnn/xnor-net)；2、 pruning: normal、regular and group convolutional channel pruning；3、 group convolution structure；4、batch-normalization folding for binary value of feature(A))

onnx2X. ONNX2Pytorch

giantpandacv.com. www.giantpandacv.com

onnx_learn. Python

how-to-optimize-gemm. C

ACM_template. Acm_template

ArmNeonOptimization. arm-neon

cv_tools. Python

RWKV-World-HF-Tokenizer. Python

flash-rwkv. Python

yolov3-tiny-onnx-TensorRT. YOLOv3-Tiny ONNX TensorRT 6.0 demo for 13 classes (legacy)

tensorrt-llm-moe. C++

simple-faster-rcnn-explain. Jupyter Notebook

Memory-efficient-Convolution-for-Deep-Neural-Network. C++

run-rwkv-world-4-in-mlc-llm.

pytorch-deform-conv-v2-explain. Python

machine-learning. machine-learning

megatron-lm-parallel-group-playground. Python

oneflow-cifar. Python

Panzhihua-Mi-Yi-Pipa. If you want to purchase Panzhihua Mi Yi Pipa, please contact me.

model_quantization. Python

mlc-llm-code-analysis. Enable everyone to develop, optimize and deploy AI models natively on everyone's devices.

BBuf.

tvm. Open deep learning compiler stack for cpu, gpu and specialized accelerators

cpp_related_tips. 📚 C/C++ 技术面试基础知识总结，包括语言、程序库、数据结构、算法、系统、网络、链接装载库等知识及面试经验、招聘、内推等信息。This repository is a summary of the basic knowledge of recruiting job seekers and beginners in the direction of C/C++ technology, including language, program library, data structure, algorithm, system, network, link loading library, interview experience, recruitment, recommendation, etc.

Awesome-ML-SYS-Tutorial. My learning notes/codes for ML SYS.

FxxkCUDA. Cuda

trl. Train transformer language models with reinforcement learning.

buddy-mlir. An MLIR-Based Ideas Landing Project

GAN-Code. GAN

tvm-cn. TVM Documentation in Chinese Simplified / TVM 中文文档

GLM. Python

opencompass. OpenCompass is an LLM evaluation platform, supporting a wide range of models (LLaMA, LLaMa2, ChatGLM2, ChatGPT, Claude, etc) over 50+ datasets.

vit-profile. Python

msnhnet-onnx. Python

tokenizers-cpp. Universal cross-platform tokenizers binding to HF and sentencepiece

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.

nndeploy. nndeploy是一款模型端到端部署框架。以多端推理以及基于有向无环图模型部署为内核，致力为用户提供跨平台、简单易用、高性能的模型部署体验。

cfx-article-src. C++

transformers. 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

ml-engineering. Machine Learning Engineering Open Book

LargeScale. Python

accelerate. 🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support