From 9dba17c6b97e38b2b9916022ba31931ff9a9aa45 Mon Sep 17 00:00:00 2001 From: JIMMY ZHAO Date: Sun, 31 Dec 2023 23:26:58 -0500 Subject: [PATCH] create model compression tools (#452) * create model compression tools add neural-compressor * add AutoGPTQ * AutoAWQ * Update README.md --- README.md | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index 54449c1a..cdc927e1 100644 --- a/README.md +++ b/README.md @@ -19,10 +19,10 @@ This repository contains a curated list of awesome open source libraries that wi | [๐Ÿงต Data Pipeline](#data-pipeline) | [๐Ÿท๏ธ Data Labelling & Synthesis](#data-labelling-and-synthesis) | [๐Ÿ“… Metadata Management](#metadata-management) | | [๐Ÿ—บ๏ธ Computation Distribution](#computation-load-distribution) | [๐Ÿ“ฅ Model Serialisation](#model-serialisation) | [๐Ÿงฎ Optimized Computation](#optimized-computation)| | [๐Ÿ’ธ Data Stream Processing](#data-stream-processing) | [:red_circle: Outlier & Anomaly Detection](#outlier-and-anomaly-detection) | [๐ŸŽ Feature Store](#feature-store) | -| [โš” Adversarial Robustness](#adversarial-robustness) | [๐Ÿ’พ Data Storage Optimisation](#data-storage-optimisation) | [๐Ÿ““ Data Science Notebook](#data-science-notebook) | -| [๐Ÿ”ฅ Neural Search](#neural-search) | [๐Ÿ‘๏ธ Industry-strength Computer Vision](#industry-strength-cv) | [๐Ÿ”  Industry-strength Natural Language Processing](#industry-strength-nlp) | -| [๐Ÿ• Industry-strength Reinforcement Learning](#industry-strength-rl) | [๐Ÿ“Š Industry-strength Visualisation](#industry-strength-visualisation) | [๐Ÿ™Œ Industry-strength Recommender System](#industry-strength-recsys) | -| [๐Ÿ“ˆ Industry-strength Benchmarking & Evaluation](#industry-strength-benchmarking-and-evaluation) | [๐Ÿ’ฐ Commercial Platform](#commercial-platform) | +| [โš” Adversarial Robustness](#adversarial-robustness) | [๐Ÿ’พ Data Storage Optimization](#data-storage-optimisation) | [๐Ÿ““ Data Science Notebook](#data-science-notebook) | +| [๐Ÿ”ฅ Neural Search](#neural-search) | [๐Ÿ”ฉ Model Optimization, Compilation & Compression](#model-optimization-compilation-and-compression) | [๐Ÿ‘๏ธ Industry-strength Computer Vision](#industry-strength-cv) | +| [๐Ÿ”  Industry-strength Natural Language Processing](#industry-strength-nlp) | [๐Ÿ• Industry-strength Reinforcement Learning](#industry-strength-rl) | [๐Ÿ“Š Industry-strength Visualisation](#industry-strength-visualisation) | +| [๐Ÿ™Œ Industry-strength Recommender System](#industry-strength-recsys) | [๐Ÿ“ˆ Industry-strength Benchmarking & Evaluation](#industry-strength-benchmarking-and-evaluation) | [๐Ÿ’ฐ Commercial Platform](#commercial-platform) | ## 10 Min Video Overview @@ -494,6 +494,15 @@ This repository contains a curated list of awesome open source libraries that wi * [NMSLIB](https://github.com/nmslib/nmslib) ![](https://img.shields.io/github/stars/nmslib/nmslib.svg?style=social) - Non-Metric Space Library (NMSLIB): An efficient similarity search library and a toolkit for evaluation of k-NN methods for generic non-metric spaces. * [Qdrant](https://github.com/qdrant/qdrant) ![](https://img.shields.io/github/stars/qdrant/qdrant.svg?style=social) - An open source vector similarity search engine with extended filtering support. + +## Model Optimization, Compilation and Compression +* [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) ![](https://img.shields.io/github/stars/casper-hansen/AutoAWQ.svg?style=social) - AutoAWQ is an easy-to-use package for 4-bit quantized models. +* [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ) ![](https://img.shields.io/github/stars/PanQiWei/AutoGPTQ.svg?style=social) - An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm. +* [AWQ](https://github.com/mit-han-lab/llm-awq) ![](https://img.shields.io/github/stars/mit-han-lab/llm-awq.svg?style=social) - Activation-aware Weight Quantization for LLM Compression and Acceleration. +* [GPTQ](https://github.com/IST-DASLab/gptq) ![](https://img.shields.io/github/stars/IST-DASLab/gptq.svg?style=social) - Accurate Post-training Quantization of Generative Pretrained Transformers. +* [neural-compressor](https://github.com/intel/neural-compressor) ![](https://img.shields.io/github/stars/intel/neural-compressor.svg?style=social) - Intelยฎ Neural Compressor aims to provide popular model compression techniques such as quantization, pruning (sparsity), distillation, and neural architecture search on mainstream frameworks. + + ## Industry Strength CV * [Deep Lake](https://github.com/activeloopai/deeplake) ![](https://img.shields.io/github/stars/activeloopai/deeplake.svg?style=social) - Deep Lake is a data infrastructure optimized for computer vision. * [Detectron2](https://github.com/facebookresearch/detectron2) ![](https://img.shields.io/github/stars/facebookresearch/detectron2.svg?style=social) - Detectron2 is Facebook AI Research's next generation library that provides state-of-the-art detection and segmentation algorithms.