Stars
My learning notes/codes for ML SYS.
Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation
how to optimize some algorithm in cuda.
High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.
A lightweight data processing framework built on DuckDB and 3FS.
A high-performance distributed file system designed to address the challenges of AI training and inference workloads.
FlashMLA: Efficient MLA decoding kernels
AIInfra（AI 基础设施）指AI系统从底层芯片等硬件，到上层软件栈支持AI大模型训练和推理。
Pretrain, finetune ANY AI model of ANY size on multiple GPUs, TPUs with zero code changes.
Repository hosting code for "Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations" (https://arxiv.org/abs/2402.17152).
An open-source C++ library developed and used at Facebook.
DeepRec is a high-performance recommendation deep learning framework based on TensorFlow. It is hosted in incubation in LF AI & Data Foundation.
A high performance and generic framework for distributed DNN training
Recommendation Algorithm大规模推荐算法库，包含推荐系统经典及最新算法LR、Wide&Deep、DSSM、TDM、MIND、Word2Vec、Bert4Rec、DeepWalk、SSR、AITM，DSIN，SIGN，IPREC、GRU4Rec、Youtube_dnn、NCF、GNN、FM、FFM、DeepFM、DCN、DIN、DIEN、DLRM、MMOE、PLE、ESM…
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
An Industrial Graph Neural Network Framework
Python package built to ease deep learning on graph, on top of existing DL frameworks.
Generate embeddings from large-scale graph-structured data.
Representation learning on large graphs using stochastic graph convolutions.
VIP cheatsheets for Stanford's CS 230 Deep Learning
A Flexible and Powerful Parameter Server for large-scale machine learning
A flexible, high-performance serving system for machine learning models
Apache Spark - A unified analytics engine for large-scale data processing
Notes talking about the design and implementation of Apache Spark
Officially maintained, supported by PaddlePaddle, including CV, NLP, Speech, Rec, TS, big models and so on.