# junyanz/pytorch-CycleGAN-and-pix2pix

Image-to-Image Translation in PyTorch

Repository: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix
Canonical: https://ross.abutalabs.com/products/pytorch-cyclegan-and-pix2pix
Language: Python
License: NOASSERTION
License Family: other
Topics: pytorch, gan, cyclegan, pix2pix, deep-learning, computer-vision, computer-graphics, image-manipulation, image-generation, generative-adversarial-network, gans
Last push: 2025-08-06T17:58:38+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 35, release rhythm 35, longevity 100
- inputs: {"age_days": 3424, "days_push": 392, "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 25232, forks 6569 (observed 2026-08-28T04:11:38.072069+00:00)

## What it is
Official PyTorch implementations of CycleGAN and pix2pix for paired and unpaired image-to-image translation. It includes training and testing scripts, pretrained models, and support for multi-GPU DDP training with PyTorch 2.4+.

## Use cases
- translate photos between styles without paired training data
- convert sketches or edges to realistic images
- train a pix2pix model on my own paired dataset
- turn horses into zebras with cyclegan
- reproduce GAN image-to-image translation paper results
- learn how GANs are implemented in PyTorch

## When to choose
- you need well-tested reference implementations of CycleGAN or pix2pix
- you want pretrained models for common translation tasks
- you need a codebase to adapt for custom image translation research

## When to avoid
- you need fast one-step translation - the authors recommend img2img-turbo instead
- you want diffusion-based image editing rather than GANs
- you need a production inference service rather than research code

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, image-processing, computer-vision, graphics
- domain: deep-learning, computer-vision, image-processing, machine-learning, graphics
- platform: python, windows, cross-platform
- tags: gan, cyclegan, pix2pix, image-to-image-translation, pytorch, generative-models, research-code, unpaired-translation, gpu, linux, macos

## Member repositories
- junyanz/pytorch-CycleGAN-and-pix2pix (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:38.072069+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 2026-08-29T16:56:04.912022+00:00, confidence not recorded.
  - readme: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix (fetched 2026-08-28T04:11:38.072069+00:00, sha 3f8df6d3d484)
- Data as of 2026-08-30T08:39:29.467469+00:00.