357 open source projects by facebookresearch

RLMeta is a light-weight flexible framework for Distributed Reinforcement Learning Research.

Neural Reconstruction for Foveated Rendering and Video Compression using Learned Statistics of Natural Videos

PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs

A suite of tools for managing crowdsourcing tasks from the inception through to data packaging for research use.

Re-Examining Linear Embeddings for High-dimensional Bayesian Optimization

Toy datasets to evaluate algorithms for domain generalization and invariance learning.

Perceiving 3D Human-Object Spatial Arrangements from a Single Image in the Wild

VoxelCNN: Order-Aware Generative Modeling Using the 3D-Craft Dataset

Paper Artifacts for "Aroma: Code Recommendation via Structural Code Search"

An R package for accessing the Facebook Ad Library API

An original implementation of "MetaICL Learning to Learn In Context" by Sewon Min, Mike Lewis, Luke Zettlemoyer and Hannaneh Hajishirzi

Misspelling Oblivious Word Embeddings

Talking with Hands

SIMULEVAL A General Evaluation Toolkit for Simultaneous Translation

VisualEchoes Dataset (ECCV 2020)

Code for Improved Condtional VRNNs for Video Prediction

Code for the paper "UnNatural Language Inference" to appear at ACL 2021 (Long Paper)

The Easy Communications (EasyCom) dataset is a world-first dataset designed to help mitigate the *cocktail party effect* from an augmented-reality (AR) -motivated multi-sensor egocentric world view.

Experiment code for the ACL 2020 paper "Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders".

The official pytorch implementation of our paper "Is Space-Time Attention All You Need for Video Understanding?"

A repository for benchmarking neural vocoders by their quality and speed.

A Dataset for Tuning and Evaluation of Sentence Simplification Models with Multiple Rewriting Transformations

Randomized Value Functions via Multiplicative Normalizing Flows

A thin, highly portable toolkit for dense loop-based computation.

The Python library with command line tools to interact with Dynabench(https://dynabench.org/), such as uploading models.

A collection of algorithms that can do join between two parties while preserving the privacy of keys on which the join happens

DrQ-v2: Improved Data-Augmented Reinforcement Learning

MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning Research

DepthContrast self-supervised learning for 3D

Tooling for the Common Objects In 3D dataset.

Repository for MeshTalk supplemental material and code once the (already approved) 16 GHS captures our lab will make publicly available are released.

Zero-shot dialogue state tracking (DST)

Understanding Training Dynamics of Deep ReLU Networks

This repo will contain replication package for the paper "Feeding Trees to Transformers for Code Completion"

a Lightweight library for sequential learning agents, including reinforcement learning

This repository provides code for "On Interaction Between Augmentations and Corruptions in Natural Corruption Robustness".

Contriever: Unsupervised Dense Information Retrieval with Contrastive Learning

New egocentric synthetic dataset for egocentric 3D human pose estimation

Fairring (FAIR + Herring) is a plug-in for PyTorch that provides a process group for distributed training that outperforms NCCL at large scales

Code for the paper "Learning an Unreferenced Metric for Online Dialogue Evaluation", ACL 2020

The Differentiable Cross-Entropy Method

Code for "DONeRF Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networks"

An official reimplementation of the method described in the INTERSPEECH 2021 paper - Speech Resynthesis from Discrete Disentangled Self-Supervised Representations.

Library for 8-bit optimizers and quantization routines.

Code and data for paper Colorless Green Recurrent Networks Dream Hierarchically

pytorch version of the SAFE neural network

FBPCS (Facebook Private Computation Solutions) leverages secure multi-party computation (MPC) to output aggregated data without making unencrypted, readable data available to the other party or any third parties. Facebook provides impression & opportunity data, and the advertiser provides conversion / outcome data. Both parties have dedicated cl…

Official code Cross-Covariance Image Transformer (XCiT)

We present a framework for training multi-modal deep learning models on unlabelled video data by forcing the network to learn invariances to transformations applied to both the audio and video streams.