Frameworks & Tools
- Dynamic neural networks
- hybrid front end
- Tape based autograd
Open Neural Network Exchange (ONNX) is an open ecosystem that empowers AI developers to choose the right tools as their project evolves. ONNX provides an open source format for AI models, both deep learning and traditional ML. It defines an extensible computation graph model, as well as definitions of built-in operators and standard data types. Currently we focus on the capabilities needed for inferencing (scoring).
ONNX is widely supported and can be found in many frameworks, tools, and hardware. Enabling interoperability between different frameworks and streamlining the path from research to production helps increase the speed of innovation in the AI community.
Tensor Comprehensions accelerates development by automatically generating code from high-level mathematical operations
Glow is a machine learning compiler that accelerates the performance of deep learning frameworks on different hardware platforms
FAISS allows developers to quickly search for embeddings of multimedia documents that are similar to each other
StarSpace is a general purpose neural embedding model that can be applied to many areas including text classification.
Visdom generates rich visualizations of live data to help developers stay on top of their scientific experiments
Libraries, Models & Datasets
Detectron is a high-performance codebase for object detection, covering bounding box and object instance segmentation outputs
DensePose is designed to map all human pixels of an RGB image to a 3D surface-based representation of the human body.
WSL Embeddings allow experimentation with image recognition capabilities based on models trained on large datasets
PyText is a deep-learning based natural language processing (NLP) modeling framework built on PyTorch
FastText is a lightweight library designed to help build scalable solutions for text representation and classification.
Translate is an open source project based on Facebook’s machine translation systems.
ParlAI is a platform that streamlines the process of researching, training, and evaluating conversational models across multiple tasks
Fairseq is a sequence modeling toolkit for training custom models for translation, summarization, and other text generation tasks.
MUSE is a Python library that enables faster development and evaluation of cross-lingual word embeddings and NLP.
VizSeq is a research toolkit for natural language generation (translation, captioning, summarization, etc.)
Wav2letter is an end-to-end Automatic Speech Recognition (ASR) system for transcribing speech
Large dataset for the training of weakly supervised & unsupervised speech models.
ELF is a platform for game research that allows developers to train and test their algorithms in various game environments.
ELF OpenGo is an AI bot from Facebook AI Research (FAIR) that has defeated world champion professional Go players.
House3D is a rich environment containing thousands of 3D scenes of visually realistic houses with full labeling.
Benchmark for PHYsical REasoning that contains a set of simple classical mechanics puzzles in a 2D environment.
TorchCraft is a library that enables AI research on real-time strategy (RTS) games such as StarCraft: Brood War.