codedecde

langchain

🦜🔗 Build context-aware reasoning applications

DeepSpeed

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

streamlit

Streamlit — A faster way to build and share data apps.

python-fire

Python Fire is a library for automatically generating command line interfaces (CLIs) from absolutely any Python object.

loguru

Python logging made (stupidly) simple

datasets

🤗 The largest hub of ready-to-use datasets for ML models with fast, easy-to-use and efficient data manipulation tools

DeOldify

A Deep Learning based project for colorizing and restoring old images (and video!)

volkswagen

:see_no_evil: Volkswagen detects when your tests are being run in a CI server, and makes them pass.

deepmind-research

This repository contains implementations and illustrative code to accompany DeepMind publications

flash-attention

Fast and memory-efficient exact attention

py-spy

Sampling profiler for Python programs

triton

Development repository for the Triton language and compiler

tiktoken

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

apex

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

xformers

Hackable and optimized Transformers building blocks, supporting a composable construction.

lm-evaluation-harness

A framework for few-shot evaluation of language models.

RPA-Python

Python package for doing RPA

trlx

A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)

gpustat

📊 A simple command-line utility for querying and monitoring GPU status

aiofiles

File support for asyncio

DeepSpeed-MII

MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.

dacite

Simple creation of data classes from dictionaries.

mup

maximal update parametrization (µP)

deduplicate-text-datasets

libact

Pool-based active learning in Python

LinkBERT

[ACL 2022] LinkBERT: A Knowledgeable Language Model 😎 Pretrained with Document Links

laserembeddings

LASER multilingual sentence embeddings as a pip package

infinibatch

Efficient, check-pointed data loading for deep learning with massive data sets.

ungoliant

:spider: The pipeline for the OSCAR corpus

blobfile

Read Google Cloud Storage, Azure Blobs, and local paths with the same interface