# kakaobrain/fast-autoaugment

Official Implementation of 'Fast AutoAugment' in PyTorch.

Repository: https://github.com/kakaobrain/fast-autoaugment
Canonical: https://ross.abutalabs.com/products/fast-autoaugment
Language: Python
License: MIT
License Family: permissive
Topics: deep-learning, convolutional-neural-networks, pytorch, augmentation, image-classification, computer-vision, distributed, cnn, automl, automated-machine-learning
Last push: 2021-06-16T17:01:28+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2681, "days_push": 1904, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1605, forks 197 (observed 2026-08-28T04:05:10.346247+00:00)

## What it is
Official PyTorch implementation of Fast AutoAugment (NeurIPS 2019), which learns image augmentation policies via density-matching search. It reduces augmentation policy search time by orders of magnitude compared to AutoAugment while achieving comparable accuracy.

## Use cases
- search for image augmentation policies for CIFAR-10/100 training
- speed up AutoAugment policy search on ImageNet
- improve image classification accuracy with learned augmentation
- reproduce Fast AutoAugment paper results
- apply learned augmentation policies to CNN training in PyTorch

## When to choose
- you train CNNs in PyTorch and want AutoAugment-quality policies with far less search compute
- you need the official reference implementation or pretrained checkpoints from the paper

## When to avoid
- you use TensorFlow or non-PyTorch frameworks
- you need actively maintained tooling - the repo has not seen releases since 2021
- you work outside image classification, e.g. NLP or detection

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, image-processing, data-science
- domain: deep-learning, computer-vision, image-processing
- platform: python
- tags: autoaugment, data-augmentation, automl, pytorch, image-classification, neurips-2019, cnn, gpu, linux

## Member repositories
- kakaobrain/fast-autoaugment (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:10.346247+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 2026-08-30T03:52:17.198335+00:00, confidence not recorded.
  - readme: https://github.com/kakaobrain/fast-autoaugment (fetched 2026-08-28T04:05:10.346247+00:00, sha 51b45b25456c)
- Data as of 2026-08-30T08:39:29.467469+00:00.