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

Unsupervised text tokenizer focused on computational efficiency

Reference BLEU implementation that auto-downloads test sets and reports a version string to facilitate cross-lab comparisons

Training Transformer-XL on 128 GPUs

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

One Dark theme for IntelliJ IDEA, PhpStorm, PyCharm, RubyMine, WebStorm

Supplementary code for the paper "Hyperbolic Image Embeddings".

Library to scrape and clean web pages to create massive datasets.

Accelerated deep learning R&D

PlayGround: AI Research into Multi-Agent Learning.

Presentations of the advanced topics in optimization

A course on Optimization Methods

OpenChem: Deep Learning toolkit for Computational Chemistry and Drug Design Research

A toolkit for developing and comparing reinforcement learning algorithms.

Soft Actor-Critic

Reinforcement Learning environments based on the 1993 game Doom

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

NIPS 2017 Value Prediction Network

OpenAI Baselines: high-quality implementations of reinforcement learning algorithms

A set of Deep Reinforcement Learning Agents implemented in Tensorflow.