# DeepVoltaire/AutoAugment

Unofficial implementation of the ImageNet, CIFAR 10 and SVHN Augmentation Policies learned by AutoAugment using pillow

Repository: https://github.com/DeepVoltaire/AutoAugment
Canonical: https://ross.abutalabs.com/products/autoaugment
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2023-05-16T16:40:38+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3011, "days_push": 1205, "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 1493, forks 202 (observed 2026-08-28T04:04:52.755431+00:00)

## What it is
An unofficial Python implementation of the AutoAugment data augmentation policies learned for ImageNet, CIFAR-10, and SVHN, built on Pillow. It provides drop-in transforms for PyTorch data pipelines to improve image classifier accuracy.

## Use cases
- apply autoaugment policies to pytorch image training
- improve image classifier accuracy with data augmentation
- use imagenet augmentation policy on custom dataset
- augment cifar10 training images
- reproduce autoaugment paper results

## When to choose
- you train image classification models in PyTorch and want the AutoAugment policies as a simple transform
- you want a lightweight Pillow-based implementation without heavy dependencies

## When to avoid
- you need augmentation policies for non-image modalities
- you need actively maintained features or support for newer torchvision APIs
- you want automatically learned policies rather than the fixed published ones

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning, machine-learning
- platform: python
- tags: data-augmentation, autoaugment, pytorch, pillow, image-classification

## Member repositories
- DeepVoltaire/AutoAugment (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:52.755431+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 2026-08-30T04:33:25.524317+00:00, confidence not recorded.
  - readme: https://github.com/DeepVoltaire/AutoAugment (fetched 2026-08-28T04:04:52.755431+00:00, sha 2b7d0c6e9fdb)
- Data as of 2026-08-30T08:39:29.467469+00:00.