Ecosystem
Tools

PFRL is a deep reinforcement learning library that implements various state-of-the-art deep reinforcement algorithms in Python using PyTorch.

CrypTen is a framework for Privacy Preserving ML. Its goal is to make secure computing techniques accessible to ML practitioners.

The easiest way to use deep metric learning in your application. Modular, flexible, and extensible.

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR).

baal (bayesian active learning) aims to implement active learning using metrics of uncertainty derived from approximations of bayesian posteriors in neural networks.

Polyaxon is a platform for building, training, and monitoring large-scale deep learning applications.

Deep Graph Library (DGL) is a Python package built for easy implementation of graph neural network model family, on top of PyTorch and other frameworks.

A deep learning library for video understanding research. Hosts various video-focused models, datasets, training pipelines and more.

Ray is a fast and simple framework for building and running distributed applications.

A generalizable application framework for segmentation, regression, and classification using PyTorch

FairScale is a PyTorch extension library for high performance and large scale training on one or multiple machines/nodes.

Fast and extensible image augmentation library for different CV tasks like classification, segmentation, object detection and pose estimation.

State-of-the-art Natural Language Processing for PyTorch.

Determined is a platform that helps deep learning teams train models more quickly, easily share GPU resources, and effectively collaborate.

Basic Utilities for PyTorch Natural Language Processing (NLP).

BoTorch is a library for Bayesian Optimization. It provides a modular, extensible interface for composing Bayesian optimization primitives.

Minimalist Neural Machine Translation toolkit for educational purposes

Stable Baselines3 (SB3) is a set of reliable implementations of reinforcement learning algorithms in PyTorch.

Ignite is a high-level library for training neural networks in PyTorch. It helps with writing compact, but full-featured training loops.

A PyTorch framework for deep learning on point clouds.

A PyTorch-based knowledge distillation toolkit for natural language processing

ParlAI is a unified platform for sharing, training, and evaluating dialog models across many tasks.

skorch is a high-level library for PyTorch that provides full scikit-learn compatibility.

Pyro is a universal probabilistic programming language (PPL) written in Python and supported by PyTorch on the backend.

Kornia is a differentiable computer vision library that consists of a set of routines and differentiable modules to solve generic CV problems.

Flair is a very simple framework for state-of-the-art natural language processing (NLP).

ClearML is a full system ML / DL experiment manager, versioning and ML-Ops solution.

Lightly is a computer vision framework for self-supervised learning.

An open source framework for deep learning on satellite and aerial imagery.

TensorLy is a high level API for tensor methods and deep tensorized neural networks in Python that aims to make tensor learning simple.

pystiche is a framework for Neural Style Transfer (NST) built upon PyTorch.

A toolbox for adversarial robustness research. It contains modules for generating adversarial examples and defending against attacks.

PennyLane is a library for quantum ML, automatic differentiation, and optimization of hybrid quantum-classical computations.

Framework for reproducible classification of Alzheimer's Disease

fastai is a library that simplifies training fast and accurate neural nets using modern best practices.

OpenMMLab covers a wide range of computer vision research topics including classification, detection, segmentation, and super-resolution.

ONNX Runtime is a cross-platform inferencing and training accelerator.

Horovod is a distributed training library for deep learning frameworks. Horovod aims to make distributed DL fast and easy to use.

Poutyne is a Keras-like framework for PyTorch and handles much of the boilerplating code needed to train neural networks.

NeMo: a toolkit for conversational AI.

The PopTorch interface library is a simple wrapper for running PyTorch programs directly on Graphcore IPUs.

PyTorch Lightning is a Keras-like ML library for PyTorch. It leaves core training and validation logic to you and automates the rest.

higher is a library which facilitates the implementation of arbitrarily complex gradient-based meta-learning algorithms and nested optimisation loops with near-vanilla PyTorch.

ML Prediction, Planning and Simulation for Self-Driving built on PyTorch.

Hummingbird compiles trained ML models into tensor computation for faster inference.

A unified ensemble framework for PyTorch to improve the performance and robustness of your deep learning model.

PyTorch Geometric Temporal is a temporal (dynamic) extension library for PyTorch Geometric.

TorchDrift is a data and concept drift library for PyTorch. It lets you monitor your PyTorch models to see if they operate within spec.

GPyTorch is a Gaussian process library implemented using PyTorch, designed for creating scalable, flexible Gaussian process models.

PyTorch Geometric is a library for deep learning on irregular input data such as graphs, point clouds, and manifolds.

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

AdaptDL is a resource-adaptive deep learning training and scheduling framework.

A library for state-of-the-art self-supervised learning

Detectron2 is FAIR's next-generation platform for object detection and segmentation.

AllenNLP is an open-source research library built on PyTorch for designing and evaluating deep learning models for NLP.

PyTorch3D provides efficient, reusable components for 3D Computer Vision research with PyTorch.

Captum (“comprehension” in Latin) is an open source, extensible library for model interpretability built on PyTorch.

Flexible and powerful tensor operations for readable and reliable code.

An open source hyperparameter optimization framework to automate hyperparameter search.

Machine learning metrics for distributed, scalable PyTorch applications.

Train PyTorch models with Differential Privacy

Catalyst helps you write compact, but full-featured deep learning and reinforcement learning pipelines with a few lines of code.

A framework for elegantly configuring complex applications.

Forte is a toolkit for building NLP pipelines featuring composable components, convenient data interfaces, and cross-task interaction.

TorchIO is a set of tools to efficiently read, preprocess, sample, augment, and write 3D medical images in deep learning applications written in PyTorch.

PySyft is a Python library for encrypted, privacy preserving deep learning.

MONAI provides domain-optimized foundational capabilities for developing healthcare imaging training workflows.

A powerful and flexible machine learning platform for drug discovery.