357 open source projects by facebookresearch

Narwhal and Tusk: A DAG-based Mempool and Efficient BFT Consensus.

Code release for ConvNeXt model

Global Climate Statistical Analysis Library (GCSAL) allows viewing of climate statistics formulated from over 60 years of data acquisition at 3000 locations around the world.

Code accompanying EGO-TOPO: Environment Affordances from Egocentric Video (CVPR 2020)

A library that allows for inference on probabilistic models

Decoupling Dynamics and Reward for Transfer Learning

Development repository for integrating FlexFlow (A distributed deep learning framework that supports flexible parallelization strategies) with DLRM (Deep Learning Recommendation Model)

Predicting Deeper into the Future of Semantic Segmentation

Pytorch implementation of time-domain filterbanks

Python interface for the DIGIT tactile sensor

PyTorch implementation of IRMAE https//arxiv.org/abs/2010.00679

Inferring Body Pose in Egocentric Video via First and Second Person Interactions

Code for SplitNet paper

Recipes are a standard, well supported set of blueprints for machine learning engineers to rapidly train models using the latest research techniques without significant engineering overhead.Specifically, recipes aims to provide- Consistent access to pre-trained SOTA models ready for production- Reference implementations for SOTA research reprodu…

Code release for "Detecting Twenty-thousand Classes using Image-level Supervision".

A method to estimate 3D hand gestures from 3D body motion input in body languages.

Code for Emergent Translation in Multi-Agent Communication

Physical contact plays a critical role in hand and object grasping. By estimating desirable contact then optimizing the hand pose to achieve it, ContactOpt improves the accuracy and realism of estimated hand and object poses.

The code processes URLs in an attempt to consolidate different web addresses that point to the same URL and to remove potentially private and/or sensitive data. It is part of the Facebook URL shares release effort, which is led by Election Research Commission (ERC).

Code for Parameter Prediction for Unseen Deep Architectures (NeurIPS 2021)

The code reproduces the results of the experiments in the paper. In particular, it performs experiments in which machine-learning models are trained that are guaranteed to not leak information about the training data they are trained on.

The Compositional Perturbation Autoencoder (CPA) is a deep generative framework to learn effects of perturbations at the single-cell level. CPA performs OOD predictions of unseen combinations of drugs, learns interpretable embeddings, estimates dose-response curves, and provides uncertainty estimates.

Riemannian Convex Potential Maps

This code implements Prioritized Level Replay, a method for sampling training levels for reinforcement learning agents that exploits the fact that not all levels are equally useful for agents to learn from during training.

Source code, datasets and trained models for the paper Learning Advanced Mathematical Computations from Examples (ICLR 2021), by François Charton, Amaury Hayat (ENPC-Rutgers) and Guillaume Lample

Federated Learning Simulator (FLSim) is a flexible, standalone core library that simulates FL settings with a minimal, easy-to-use API. FLSim is domain-agnostic and accommodates many use cases such as vision and text.

This repo covers the implementation for Labelling unlabelled videos from scratch with multi-modal self-supervision, which learns clusters from multi-modal data in a self-supervised way.

PyTorch original implementation of "Unsupervised Question Decomposition for Question Answering"

Reward Estimation for Variance Reduction in Deep Reinforcement Learning

Code and data to support the paper "PAQ 65 Million Probably-Asked Questions andWhat You Can Do With Them"

Code for SIMMC 2.0: A Task-oriented Dialog Dataset for Immersive Multimodal Conversations

Learning to Perform Local Rewriting for Combinatorial Optimization

Per-Pixel Classification is Not All You Need for Semantic Segmentation (NeurIPS 2021, spotlight)

Reference-less Quality Estimation of Text Simplification Systems

Omnivore: A Single Model for Many Visual Modalities

Visual Coreference Resolution in Visual Dialog using Neural Module Networks

This repository houses the IMPlicature and PRESupposition diagnostic dataset (IMPPRES), consisting of >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), …

Codebase for project about unsupervised skill learning via variational inference and causality.

code for "AttentiveNAS Improving Neural Architecture Search via Attentive Sampling"

Repo reproducing experimental results in "Addressing the Topological Defects of Disentanglement"

Listen to Look: Action Recognition by Previewing Audio (CVPR 2020)

Multi Task RL Baselines

NUANCED is a user-centric conversational recommendation dataset that contains 5.1k annotated dialogues and 26k high-quality user turns.

Project for the paper "A Scalable Approach to Control Diverse Behaviors for Physically Simulated Characters"

This code reproduces the results of the paper, "Measuring Data Leakage in Machine-Learning Models with Fisher Information"

AEPsych is a tool for adaptive experimentation in psychophysics and perception research, built on top of gpytorch and botorch.

Energy-based models for atomic-resolution protein conformations

Official code accompanying the arXiv paper Compressing Multisets with Large Alphabets

Multi-Target Embodied Question Answering

PyTorch implementation of MoCo v3 https//arxiv.org/abs/2104.02057