hanweisen

ECE408

Code base and slides for ECE408：Applied Parallel Programming On GPU.

Cpp-Primer-Plus-6th

《C++ Primer Plus 第6版（中文版）》原书代码、习题答案和个人笔记，仅供学习和交流。

Megatron-LM

Ongoing research training transformer models at scale

image-syncer

Docker image synchronization tool for Docker Registry V2 based services

NeMo

A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)

The-C-20-Masterclass-Source-Code

Source code for the C++ 20 Masterclass on udemy

modern-cpp-tutorial

📚 Modern C++ Tutorial: C++11/14/17/20 On the Fly | https://changkun.de/modern-cpp/

SciencePlots

Matplotlib styles for scientific plotting

open-llms

📋 A list of open LLMs available for commercial use.

llm.c

LLM training in simple, raw C/CUDA

Awesome-Chinese-LLM

整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。

grok-1

Grok open release

Data-Science-Notes

数据科学的笔记以及资料搜集

transformer-debugger

mesh-transformer-jax

Model parallel transformers in JAX and Haiku

pmem-csi

Persistent Memory Container Storage Interface Driver

horovod

Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

AISystem

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术

xla

A machine learning compiler for GPUs, CPUs, and ML accelerators

paper-reading

深度学习经典、新论文逐段精读

kaniko

Build Container Images In Kubernetes

dolma

Data and tools for generating and inspecting OLMo pre-training data.

OLMo

Modeling, training, eval, and inference code for OLMo

zero_nlp

中文nlp解决方案(大模型、数据、模型、训练、推理)

PLMpapers

Must-read Papers on pre-trained language models.

DCGM

NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs

GLM-130B

GLM-130B: An Open Bilingual Pre-Trained Model (ICLR 2023)

autogluon

Fast and Accurate ML in 3 Lines of Code

prml

Repository of notes, code and notebooks in Python for the book Pattern Recognition and Machine Learning by Christopher Bishop

d2l-zh

《动手学深度学习》：面向中文读者、能运行、可讨论。中英文版被70多个国家的500多所大学用于教学。