# brain-research/self-attention-gan

Repository: https://github.com/brain-research/self-attention-gan
Canonical: https://ross.abutalabs.com/products/self-attention-gan
Language: Python
License: Apache-2.0
License Family: permissive
Archived: true
Last push: 2019-08-19T23:34:45+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2996, "days_push": 2571, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1019, forks 176 (observed 2026-08-28T04:03:15.127793+00:00)

## What it is
A TensorFlow implementation of Self-Attention GANs for reproducing results from the paper 'Self-Attention Generative Adversarial Networks' by Zhang, Goodfellow et al. It includes training and evaluation scripts for ImageNet with self-attention layers in generator and discriminator networks.

## Use cases
- reproduce sagan imagenet results
- train a self-attention gan
- generate images with a gan
- study self-attention in generative models
- implement spectral normalization in tensorflow

## When to choose
- you need a reference implementation of the SAGAN paper
- you want to experiment with self-attention GANs in TensorFlow 1.x
- you are doing research on generative adversarial networks

## When to avoid
- you need a maintained or modern framework
- you use TensorFlow 2.x or PyTorch
- you lack multi-GPU resources since training takes weeks

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing
- domain: deep-learning, machine-learning, image-processing
- platform: python
- tags: gan, generative-adversarial-networks, self-attention, tensorflow, research-code, image-generation, spectral-normalization, linux, gpu

## Member repositories
- brain-research/self-attention-gan (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:15.127793+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 2026-08-30T07:09:21.921361+00:00, confidence not recorded.
  - readme: https://github.com/brain-research/self-attention-gan (fetched 2026-08-28T04:03:15.127793+00:00, sha a5a7c46fa1c0)
- Data as of 2026-08-30T08:39:29.467469+00:00.