Variational Bayesian PCA for incomplete data with native missingness handling, uncertainty outputs, and C++ accelerated kernels.
active 2026-03-25 → 2026-08-17 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 146 days from 2026-03-25 to 2026-08-17. Pushes: 26 total, peak 4 in a day. Pull requests: 8 total, peak 4 in a day. Issues: 38 total, peak 11 in a day. Comments: 6 total, peak 1 in a day. Stars: 0 total, peak 0 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 |
|---|---|---|---|---|
| jc-macdonald | 42 | 14 | 7 | 4 |
| MacdonaldJoshuaCaleb | 34 | 10 | 0 | 2 |
| Copilot | 2 | 2 | 0 | 0 |
| nightcityblade | 1 | 0 | 1 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#42jc-macdonald2026-04-15 14:38CLI: select command
- Issue comment#55jc-macdonald2026-04-13 21:19Configurable RMS computation timing within CAVI cycle
- Issue#51jc-macdonald2026-04-12 23:26Expose niter_broadprior as a configurable parameter
- Pull request#96jc-macdonald2026-04-12 23:26
- Pull request#96jc-macdonald2026-04-12 23:19
- Pull request#90jc-macdonald2026-04-09 23:37
- Pull request#90jc-macdonald2026-04-09 23:37
- Pull request#88jc-macdonald2026-04-09 22:43
- Pull request#88jc-macdonald2026-04-09 22:09
- Issue#73jc-macdonald2026-04-09 19:50bug: ARD prunes all components when refitting without xprobe on small-d blocks with missing data
- Issue#33jc-macdonald2026-04-08 21:24Replace fragile float == comparisons in tests with pytest.approx / np.testing
- Pull request#56nightcityblade2026-04-08 21:24
- Issue comment#56jc-macdonald2026-04-08 21:11fix: replace fragile float == comparisons with pytest.approx in tests
- Issue#72jc-macdonald2026-04-08 20:38bug: return_best_model=True yields VBPCA with variance_ (Vr) = None
- Pull request#79jc-macdonald2026-04-07 23:53
- Issue#77jc-macdonald2026-04-07 23:37reconstruction_ not set after fit
- Issue#75jc-macdonald2026-04-07 23:36Add K-fold cross-validation with 1-SE rule to model selection
- Issue#75jc-macdonald2026-04-07 23:36Add K-fold cross-validation with 1-SE rule to model selection
- Issue#73jc-macdonald2026-04-07 13:42bug: ARD prunes all components when refitting without xprobe on small-d blocks with missing data
- Issue comment#66jc-macdonald2026-04-06 14:19Tier 4: niter_broadprior profiling (requires #50, #52)
- Issue#64jc-macdonald2026-04-06 14:19Tier 2: ELBO as convergence metric — criterion comparison
- Issue#68jc-macdonald2026-04-06 13:41Tier 6: Scalability frontier — full vs diagonal covariance
- Issue#64jc-macdonald2026-04-06 13:38Tier 2: ELBO as convergence metric — criterion comparison
- Issue#64jc-macdonald2026-04-06 13:38Tier 2: ELBO as convergence metric — criterion comparison
- Issue#63jc-macdonald2026-04-06 13:38Tier 1: Oscillation root cause — RMS timing × pre-centering
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 0 stars here means stars gained during the window, not the repo's star count.