mx/AareLC_train
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 3 Wiki Activity
66 Commits 1 Branch 7 Tags
main
Commit Graph
10 Commits
Author SHA1 Message Date
jungfrauandClaude Fable 5.1 8c3c538244 gold: 285 frames (batch 8 added by mount: 83 corrected frames, 2-per-mount cap, 20 siblings held out); every model re-measured
CI / ledger-and-tools (push) Successful in 20s
Details 
231 frames unseen by every model (core 148), 82 crystals on them. v4-noflip-mask1 stays
the best: F1 0.855, mask score 0.798; v3-noflip-mask1 0.848 / 0.793, within noise;
the small cluster model now clearly loses on masks (-0.021, won in 0 % of draws).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-05 15:49:30 +02:00
jungfrauandClaude Fable 5.1 e8804c7762 gold: 202 frames (tpi_14688 dropped: same image as tpi_14689), mount groups and light states; comparison drawn by mount
CI / ledger-and-tools (push) Successful in 18s
Details 
11 mounts hold 40 of the 202 frames (largest 12); 6 saturated and 29 dark frames.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-02 15:48:02 +02:00
GotthardGandClaude Fable 5.1 831367d35d noise-aware gate, frozen core of unseen frames, and a cycle script for the Spark 
A paired bootstrap over the 149 unseen gold frames (2026-10-02) showed that a
difference between two models carries about +-0.023 of noise in F1, +-0.018 in
the mask score and +-0.10 in crystal F1, while the gate's tolerance was 0.005.

- tools/gate_stats.py: per-frame counts and a paired bootstrap (pure Python).
- train_run.py: when the baseline's gold predictions exist, the gate requires
  the candidate to beat it on F1 or the mask score in 90 % of the draws and
  not to lose the other in more than 90 %; the point rule is the fallback and
  is always reported (pass_point_rule).
- tools/compare_unseen.py replaces the Spark's ~/fair_gate_all.py: all models
  on the frames none trained on, on the frozen core (data/gold/unseen_core.json,
  149 frames) and on all unseen frames once the gold set grows, with the
  interval and win share against the active model.
- measure_weights.py keeps a version's gate verdict when it re-measures it.
- tools/spark_cycle.sh: gold growth, export, re-measure, train, gate, compare,
  push, with what cycles 3 and 4 taught built in.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-02 13:25:45 +02:00
jungfrau 7688cfa713 dataset: 2026-10-01_v4 (1610 train / 285 val, 493 new pool images) 2026-10-02 13:25:44 +02:00
jungfrau a6da2aa4aa dataset: 2026-09-29_v3 (1181 train / 210 val, 366 new pool images) 2026-09-29 22:41:12 +02:00
jungfrau 039cdcba1c dataset: 2026-09-29_v2 (876 train / 156 val, 315 new pool images) 2026-09-29 17:58:38 +02:00
jungfrau 8898f4212a dataset: 2026-09-28_alc-v1-humanonly (476 train / 84 val, 7 new pool images) 2026-09-28 20:01:45 +02:00
jungfrau 489cc22dc3 dataset: 2026-09-28_alc-v1 (606 train / 108 val, 714 new pool images) 2026-09-28 16:49:58 +02:00
GotthardGandClaude Fable 5.1 4698dbeaf9 training loop: run ledger, studio export importer with an image pool, train_run with gold gate and auto-promotion, SLURM template, CI
CI / ledger-and-tools (push) Successful in 17s
Details 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-28 16:06:06 +02:00
duan_j 391c357c84 migration and splitting AareLC 2026-04-14 16:07:31 +02:00