pytorch-fid (Frechet Inception Distance)
Source: https://github.com/mseitzer/pytorch-fid
Structured fields
{
"archived": false,
"revision": {
"ref": "master",
"pinned": false
},
"checked_at": "2026-10-17T00:00:00.000Z",
"usefulness": "PyTorch port of the official TensorFlow FID implementation, the long-standing distribution-level image quality metric. Established baseline, with known sensitivity to implementation details and sample size.",
"source_type": "repository",
"availability": "available",
"source_owner": "mseitzer (GitHub user)",
"next_check_at": "2026-11-16T00:00:00.000Z",
"known_limitations": [
"Scores differ slightly from the TensorFlow implementation (README notes differences); FID is poorly suited to small sample counts and to single-image judgments (general known property)."
],
"upstream_updated_at": "2024-07-03T06:40:02.000Z"
}Evidence claims
- [sourced_fact] README: a port of the official FID implementation to PyTorch; FID measures similarity between two image datasets; original TensorFlow version is bioinf-jku/TTUR. · locator: README intro source
- [sourced_fact] Repo metadata: Apache-2.0, not archived, last push 2024-07-03. · locator: GitHub API fields source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| code | Apache-2.0 | permitted | permitted | permitted | unknown | Apache-2.0; the Inception weights it downloads have their own terms. |