facebookresearch/fvcore

Collection of common code that's shared among different research projects in FAIR computer vision team.

What it solves

It provides a shared set of essential, lightweight core utilities and functionality that are commonly used across different computer vision frameworks (such as Detectron2 and PySlowFast) to avoid duplication of effort and ensure consistency.

How it works

It acts as a base library containing type-annotated and benchmarked components, including common PyTorch layers, loss functions, and specialized tools for model analysis and training management.

Who it’s for

Computer vision researchers and developers who are building or using FAIR's computer vision frameworks.

Highlights

  • Hierarchical per-operator FLOP counting tool for model complexity analysis.
  • Recursive parameter counting for neural networks.
  • Stateless, scale-invariant hyperparameter schedulers.
  • Utilities to recompute BatchNorm population statistics.

Related

  • Project
    frgfm/holocron

    A library of high-quality PyTorch implementations of recent deep learning tricks and model architectures for computer vision tasks.

  • Project
    fastai/fastai

    A deep learning library built on PyTorch that provides high-level components for rapid productivity and low-level components for research flexibility.

  • Project
    eduardolundgren/tracking.js

    A lightweight JavaScript library that brings computer vision algorithms like color tracking and face detection directly into the browser.

  • Project
    pytorch/vision

    A PyTorch library providing popular datasets, model architectures, and image transformations for computer vision.

  • Project
    hmorimitsu/ptlflow

    A unified PyTorch Lightning framework for training and evaluating a wide collection of state-of-the-art optical flow estimation models.