facebookresearch/moco
The MoCo paper, executable
Facebook AI Research's PyTorch port of the MoCo method, offering the training code behind the arXiv:1911.05722 paper.
Not currently ranked — collecting fresh signals.
star history
What it does — Provides PyTorch code for the MoCo method introduced in arXiv:1911.05722. Released by Facebook AI Research, the repository has drawn over 5,000 stars.
The interesting bit — Third-party rewrites often drift; because this comes from the same research group, it is the implementation most researchers treat as the ground-truth baseline.
Key highlights —
- Direct from Facebook AI Research
- Tied to arXiv paper 1911.05722
- 5,138 stars
- PyTorch implementation
Caveats —
- The provided description is minimal; expect to consult the paper for methodology and exact settings.
Verdict — Essential if you are benchmarking against MoCo or extending its training pipeline; otherwise, it is a narrow, paper-specific artifact.
Frequently asked
- What is facebookresearch/moco?
- Facebook AI Research's PyTorch port of the MoCo method, offering the training code behind the arXiv:1911.05722 paper.
- Is moco open source?
- Yes — facebookresearch/moco is an open-source project tracked on heatdrop.
- How popular is moco?
- facebookresearch/moco has 5.1k stars on GitHub.
- Where can I find moco?
- facebookresearch/moco is on GitHub at https://github.com/facebookresearch/moco.