For optimization algorithm research and development.
active 2023-08-18 → 2026-08-20 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 1099 days from 2023-08-18 to 2026-08-20. Pushes: 457 total, peak 13 in a day. Pull requests: 242 total, peak 9 in a day. Issues: 38 total, peak 2 in a day. Comments: 803 total, peak 24 in a day. Stars: 351 total, peak 72 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| facebook-github-bot | 1K | 324 | 150 | 557 |
| runame | 352 | 103 | 44 | 109 |
| namgyu-youn | 129 | 0 | 28 | 81 |
| hjmshi | 30 | 0 | 0 | 18 |
| anana10c | 16 | 3 | 2 | 7 |
| meta-codesync[bot] | 15 | 13 | 0 | 2 |
| francois-rozet | 10 | 0 | 1 | 5 |
| jialun-zhang | 8 | 8 | 0 | 0 |
| samsja | 6 | 0 | 2 | 3 |
| SeunghyunSEO | 4 | 0 | 1 | 3 |
| OscarDPan | 4 | 0 | 2 | 2 |
| ExpressGradient | 3 | 0 | 0 | 1 |
| haoyuz | 3 | 0 | 3 | 0 |
| ionutmodo | 3 | 0 | 0 | 2 |
| cyber-phys | 3 | 0 | 2 | 1 |
| wz337 | 3 | 0 | 0 | 0 |
| yipingji | 3 | 0 | 0 | 1 |
| windsornguyen | 2 | 0 | 0 | 1 |
| Vishal-sys-code | 2 | 0 | 0 | 2 |
| shintaro-iwasaki | 2 | 0 | 2 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#275meta-cla[bot]2026-08-20 00:44
- Issue comment#265rohan-anil2026-06-10 08:45Modded GPT pseudoinverse improvement
- Pull request#258runame2026-04-25 16:24
- Issue comment#255meta-codesync[bot]2025-12-16 03:07Sync KL-Shampoo to OSS
- Pull request#255facebook-github-bot2025-12-16 03:07
- Pull request#255runame2025-12-15 17:56
- Issue comment#254meta-codesync[bot]2025-10-14 22:26Revert `ParameterizeEnterExitContext` around Newton Schulz iterations
- Pull request#249facebook-github-bot2025-09-26 16:59
- Issue comment#252facebook-github-bot2025-09-23 22:05Ignore the type check complain on using `polyval()`
- Issue comment#252facebook-github-bot2025-09-23 21:53Ignore the type check complain on using `polyval()`
- Pull request#250facebook-github-bot2025-09-19 22:22
- Issue comment#250facebook-github-bot2025-09-19 17:44Change the default value of `scale_fn` to a private staticmethod for pickleable concern
- Issue comment#248facebook-github-bot2025-09-18 01:24Add test for `LossMetrics` used in `examples`
- Issue comment#248facebook-github-bot2025-09-11 20:06Add test for `LossMetrics` used in `examples`
- Issue comment#249facebook-github-bot2025-09-11 18:42Add custom dtype supports for subclasses of `ShampooPreconditionerConfig`
- Issue comment#248facebook-github-bot2025-09-09 22:54Add test for `LossMetrics` used in `examples`
- Issue comment#248facebook-github-bot2025-09-09 19:48Add test for `LossMetrics` used in `examples`
- Issue comment#247facebook-github-bot2025-09-05 17:57Leverage context manager to perform matrix transpose in Newton-Schulz more systematically
- Issue comment#244facebook-github-bot2025-09-02 21:15Leverage `torch.addmm()` to speed up Newton-Schulz computation
- Pull request#244facebook-github-bot2025-09-02 21:15
- Issue comment#246facebook-github-bot2025-09-02 20:51Update readme example to work (Name of AdamGraftingConfig changed to AdamPreconditionerConfig)
- Pull request#246facebook-github-bot2025-09-02 20:51
- Issue comment#245facebook-github-bot2025-09-02 19:23Expand more `preconditioner_config` choices in `shampoo_grafting_test.py`
- Issue comment#243facebook-github-bot2025-09-02 18:01Update `README.md` according to `GraftingConfig` merge
- Pull request#246CoffeeVampir32025-09-02 05:17
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 351 stars here means stars gained during the window, not the repo's star count.