# facebookresearch/pycls

Codebase for Image Classification Research, written in PyTorch.

Repository: https://github.com/facebookresearch/pycls
Canonical: https://ross.abutalabs.com/products/pycls
Language: Python
License: MIT
License Family: permissive
Archived: true
Last push: 2024-03-20T15:45:40+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2641, "days_push": 896, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2161, forks 239 (observed 2026-08-28T04:06:20.961560+00:00)

## What it is
pycls is a PyTorch-based codebase for image classification research developed by Facebook AI Research. It provides implementations of standard models like ResNet, ResNeXt, EfficientNet, and RegNet, along with tools for studying network design spaces and model populations.

## Use cases
- train image classification models in pytorch
- reproduce regnet and resnet baselines
- run model design space sweeps for neural architecture research
- download pretrained image classification models
- benchmark models across different flop regimes
- implement and evaluate new classification research ideas

## When to choose
- you need a simple, flexible codebase for image classification experiments
- you want to reproduce or build on FAIR papers like RegNet or design space studies
- you need efficient single-machine multi-GPU training for classification models
- you want pretrained baselines across a wide range of compute budgets

## When to avoid
- you need a general-purpose computer vision toolkit beyond classification (e.g., detection or segmentation)
- you want a production inference serving system
- you need a high-level training API with minimal configuration
- you require active development or frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, benchmarking
- domain: computer-vision, image-processing, deep-learning, machine-learning
- platform: python, cross-platform
- tags: image-classification, pytorch, resnet, regnet, efficientnet, model-zoo, neural-architecture-search, facebook-ai-research, research, gpu, linux

## Member repositories
- facebookresearch/pycls (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:20.961560+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:50:15.398537+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/pycls (fetched 2026-08-28T04:06:20.961560+00:00, sha c164c11398c9)
  - registry_pypi: https://pypi.org/pypi/pycls/json (fetched 2026-08-29T10:30:05.115271+00:00, sha 2f24b1239304)
- Data as of 2026-08-30T08:39:29.467469+00:00.