martinfu918 Goto Github PK

algorithms

算法和数据结构

baby-llama2-chinese

用于从头预训练+SFT一个小参数量的中文LLaMa2的仓库；24G单卡即可运行得到一个具备简单中文问答能力的chat-llama2.

bilivideos

classicnetworks

This is a simple implmentation of some classic neural networks

computer-science-textbooks

Collect some CS textbooks for learning.

coursera-ml-andrewng-notes

吴恩达老师的机器学习课程个人笔记

ctpl

Modern and efficient C++ Thread Pool Library

cuda_freshman

deeplearningsystem

Deep Learning System core principles introduction.

designpattern

利用设计模式实现超市购物

eddie-wang-hackathon2023

Whisper inference with TensorRT-LLM

googletest

GoogleTest - Google Testing and Mocking Framework

how-to-optim-algorithm-in-cuda

how to optimize some algorithm in cuda.

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.

huggingface_toturials

bert-base-chinese example

llama

Inference code for LLaMA models

llama.cpp

Port of Facebook's LLaMA model in C/C++

llama_mark.cpp

llama 2 Inference

llm-action

本项目旨在分享大模型相关技术原理以及实战经验。

mlir-tutorial

Hands-On Practical MLIR Tutorial

mytinystl

Achieve a tiny STL in C++11

nlp_torchtext

oneflow

OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.

openmlsys-zh

《Machine Learning Systems: Design and Implementation》- Chinese Version

torchkeras

Pytorch❤️ Keras 😋😋

transformers-code

手把手带你实战 Huggingface Transformers 课程视频同步更新在B站与YouTube

tvm

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

tvm_mlir_learn

compiler learning resources collect.

vllm

A high-throughput and memory-efficient inference and serving engine for LLMs