# soumith/ganhacks

starter from "How to Train a GAN?" at NIPS2016

Repository: https://github.com/soumith/ganhacks
Canonical: https://ross.abutalabs.com/products/ganhacks
License Family: other
Last push: 2022-01-09T06:29:17+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": 3554, "days_push": 1697, "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 11616, forks 1643 (observed 2026-08-28T04:10:49.040077+00:00)

## What it is
A curated list of practical tips and tricks for training Generative Adversarial Networks (GANs), originally compiled as a starter for the 'How to Train a GAN?' session at NIPS 2016. It is a documentation-style resource, not software, and is explicitly no longer maintained.

## Use cases
- tips for training GANs
- how to stabilize GAN training
- GAN training tricks like label smoothing
- why my GAN is not converging
- DCGAN best practices
- choosing activation and pooling layers for GANs
- learning GAN loss function pitfalls

## When to choose
- you are learning to train GANs and want a concise checklist of practical tricks
- you need quick reference advice on normalization, label smoothing, or architecture choices for GANs

## When to avoid
- you need up-to-date GAN research or modern techniques (the list is unmaintained since ~2020)
- you need runnable code or a library rather than written guidance

## Facets
- artifact type: learning-resource
- maturity: abandoned
- function: deep-learning, machine-learning, developer-tools
- domain: deep-learning, machine-learning, artificial-intelligence, tutorials
- platform: cross-platform
- tags: gan, generative-adversarial-networks, training-tips, nips2016, guide, documentation

## Member repositories
- soumith/ganhacks (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:49.040077+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 2026-08-29T17:15:27.858276+00:00, confidence not recorded.
  - readme: https://github.com/soumith/ganhacks (fetched 2026-08-28T04:10:49.040077+00:00, sha aeec81c0ebf6)
- Data as of 2026-08-30T08:39:29.467469+00:00.