# rosinality/style-based-gan-pytorch

Implementation A Style-Based Generator Architecture for Generative Adversarial Networks in PyTorch

Repository: https://github.com/rosinality/style-based-gan-pytorch
Canonical: https://ross.abutalabs.com/products/style-based-gan-pytorch
Language: Python
License: NOASSERTION
License Family: other
Last push: 2021-08-26T00:34:43+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": 2773, "days_push": 1834, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1110, forks 229 (observed 2026-08-28T04:03:37.448251+00:00)

## What it is
A PyTorch implementation of the StyleGAN paper (A Style-Based Generator Architecture for Generative Adversarial Networks), including training scripts and pretrained checkpoints up to 1024px. It supports LMDB-prepared datasets, style mixing, and R1 regularization with progressive training schedules.

## Use cases
- train a StyleGAN model on my own image dataset
- generate high-resolution faces with StyleGAN in PyTorch
- reproduce the StyleGAN paper results
- download pretrained StyleGAN checkpoints for FFHQ
- experiment with style mixing in generative adversarial networks

## When to choose
- you want a straightforward, readable PyTorch implementation of the original StyleGAN
- you need pretrained checkpoints for FFHQ-style face generation
- you are learning how style-based generators and progressive training work

## When to avoid
- you need StyleGAN2 or StyleGAN3 features and fixes
- you want a production-grade, actively maintained training framework
- you need multi-GPU distributed training out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, data-science
- domain: deep-learning, machine-learning, image-processing, artificial-intelligence
- platform: python
- tags: stylegan, gan, pytorch, generative-models, image-generation, research-code, gpu, linux, docker

## Member repositories
- rosinality/style-based-gan-pytorch (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:37.448251+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 2026-08-30T06:43:28.382773+00:00, confidence not recorded.
  - readme: https://github.com/rosinality/style-based-gan-pytorch (fetched 2026-08-28T04:03:37.448251+00:00, sha 6fbbcbce055a)
- Data as of 2026-08-30T08:39:29.467469+00:00.