ruv
e1936c9a24
feat(benchmarks): WiFlow-STD reproduction harness + measurement (a) results (ADR-152 §2.2)
Shipped checkpoint REFUTED (0.08% PCK@20, wrong keypoint normalization);
6 reproducibility defects documented (broken imports, corrupted dataset
tail with float32-max garbage that NaN-poisons fp16 BatchNorm, unreachable
test phase). After repairs, retraining with upstream defaults reproduces
96.09% PCK@20 full-test / 96.61% corruption-free (published 97.25%) on
RTX 5080. Claims graded MEASURED-EQUIVALENT; 2.23M params + ~0.055 GFLOPs
verified. Third-party code/weights/data stay out of tree (gitignored).
Co-Authored-By: claude-flow <ruv@ruv.net>
2026-06-10 18:54:03 -04:00
..
2026-06-10 18:54:03 -04:00
2026-05-17 18:18:07 -04:00
2026-05-31 03:43:14 -04:00
2026-05-12 23:15:10 +00:00
2026-05-25 22:47:48 -04:00
2026-04-25 21:28:13 -04:00
2026-05-11 23:40:55 -04:00
2026-06-10 15:21:09 -04:00
2026-05-24 13:00:38 -04:00
2026-05-12 23:15:10 +00:00
2026-04-25 23:07:52 -04:00
2026-05-23 16:13:28 -04:00
2026-05-24 20:20:25 -04:00
2026-05-25 22:47:48 -04:00
2026-05-17 17:41:39 -04:00
2026-05-23 15:34:48 -04:00
2026-05-23 15:34:48 -04:00
2026-04-25 23:07:52 -04:00
2026-05-31 12:22:53 -04:00
2026-06-02 18:50:39 +02:00
2026-05-29 08:46:12 -04:00
2026-04-25 21:28:13 -04:00
2026-05-17 18:01:44 -04:00
2026-05-25 17:36:40 -04:00
2026-06-03 08:55:36 +02:00
2026-04-25 21:28:13 -04:00
2026-05-28 18:57:08 -04:00
2026-05-23 15:34:48 -04:00