-
@https://github.com/microsoft
- Redmond, WA
Stars
Hackable and optimized Transformers building blocks, supporting a composable construction.
Development repository for the Triton language and compiler
A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch
A framework for few-shot evaluation of language models.
maximal update parametrization (Β΅P)
Simple creation of data classes from dictionaries.
A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)
tiktoken is a fast BPE tokeniser for use with OpenAI's models.
π Volkswagen detects when your tests are being run in a CI server, and makes them pass.
MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.
Fast and memory-efficient exact attention
π€ The largest hub of ready-to-use datasets for AI models with fast, easy-to-use and efficient data manipulation tools
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
[ACL 2022] LinkBERT: A Knowledgeable Language Model π Pretrained with Document Links
Efficient, check-pointed data loading for deep learning with massive data sets.
Read Google Cloud Storage, Azure Blobs, and local paths with the same interface
LASER multilingual sentence embeddings as a pip package
This repository contains implementations and illustrative code to accompany DeepMind publications
π A simple command-line utility for querying and monitoring GPU status
A Deep Learning based project for colorizing and restoring old images (and video!)
Streamlit β A faster way to build and share data apps.
Python Fire is a library for automatically generating command line interfaces (CLIs) from absolutely any Python object.