All Workflows CI

Management

Job queue

33 workflow runs

Actor

All actors duan_j gotthard_g

Status

All status Success Failure Canceled Skipped Waiting Running Blocked Canceling

Branch

All branches main

dedup study: suppressing nested duplicate outlines on the unseen gold frames

CI #33:Commit ddaa5a99e6 pushed by gotthard_g

main

2026-10-05 22:25:41 +02:00

20s

View workflow file

gold: 285 frames (batch 8 added by mount: 83 corrected frames, 2-per-mount cap, 20 siblings held out); every model re-measured

CI #32:Commit 8c3c538244 pushed by gotthard_g

main

2026-10-05 15:48:56 +02:00

20s

View workflow file

gold: 202 frames (tpi_14688 dropped: same image as tpi_14689), mount groups and light states; comparison drawn by mount

CI #31:Commit e8804c7762 pushed by gotthard_g

main

2026-10-02 15:47:38 +02:00

18s

View workflow file

gate: bootstrap by mount, light states and excluded frames of the gold set

CI #30:Commit 82f0d9099b pushed by gotthard_g

main

2026-10-02 15:46:54 +02:00

19s

View workflow file

small cluster model re-gated on the Spark in one environment: gate_passed against v3; comparison after cycle 4

CI #29:Commit c7d53646ce pushed by gotthard_g

main

2026-10-02 13:29:42 +02:00

19s

View workflow file

run 2026-10-01_v4-noflip-mask1: promoted; comparison updated

CI #28:Commit f37604f87a pushed by gotthard_g

main

2026-10-02 13:28:37 +02:00

19s

View workflow file

yolo26s-seg-v3-s-640px-2026-10-01

CI #27:Commit a9192406d3 pushed by duan_j

main

2026-10-02 10:48:13 +02:00

22s

View workflow file

measure dir for yolo26n-seg-v3-noflip-mask1-2026-09-30; unseen-frame comparison includes the promoted v3 model

CI #26:Commit 28cbfa889d pushed by gotthard_g

main

2026-09-30 13:57:30 +02:00

25s

View workflow file

run 2026-09-29_v3-noflip-mask1: promoted; comparison updated

CI #25:Commit 2956303ffe pushed by gotthard_g

main

2026-09-30 13:55:43 +02:00

21s

View workflow file

add ra train script

CI #24:Commit 9941d8d456 pushed by duan_j

main

2026-09-30 11:45:49 +02:00

17s

View workflow file

add train script on merlin

CI #23:Commit 8ff20f9f84 pushed by duan_j

main

2026-09-30 10:41:10 +02:00

18s

View workflow file

cluster runs on dataset v3: resolution and model-size levers

CI #22:Commit f1a945528e pushed by gotthard_g

main

2026-09-30 07:18:16 +02:00

28s

View workflow file

ledger: 2026-09-29_alc-v1-noflip-mask1 deployed at the beamlines (weights only)

CI #21:Commit 486b077f24 pushed by gotthard_g

main

2026-09-29 20:47:23 +02:00

18s

View workflow file

experiments: loop-type classifier, first cross-validated test (broken loops AUROC 0.985)

CI #20:Commit b4d86d3f31 pushed by gotthard_g

main

2026-09-29 18:18:52 +02:00

18s

View workflow file

all models re-measured on the 203-frame gold set; comparison on the unseen frames updated

CI #19:Commit 0ec5138409 pushed by gotthard_g

main

2026-09-29 18:00:13 +02:00

18s

View workflow file

slurm: base weights via LFS, and the cluster run 2026-09-29_v2-noflip-mask1 spelled out

CI #18:Commit acaf09483e pushed by gotthard_g

main

2026-09-29 17:58:48 +02:00

17s

View workflow file

run 2026-09-29_v2-recipeA: gate_failed; comparison updated

CI #17:Commit 3d39d40560 pushed by gotthard_g

main

2026-09-29 17:58:04 +02:00

18s

View workflow file

recipe: no horizontal flip and full-resolution masks by default (winner of the 2026-09-28/29 comparison)

CI #16:Commit babe197cb2 pushed by gotthard_g

main

2026-09-29 11:13:45 +02:00

16s

View workflow file

run 2026-09-29_alc-v1-noflip-mask1: promoted (same detection as A, tightest masks); loop switches to it

CI #15:Commit e7a02cd1ea pushed by gotthard_g

main

2026-09-29 11:13:18 +02:00

18s

View workflow file

run 2026-09-29_alc-v1-noflip-mask1: gate_passed, comparison updated

CI #14:Commit 703bfc2e11 pushed by gotthard_g

main

2026-09-29 10:37:06 +02:00

18s

View workflow file

ignore Ultralytics label caches inside datasets

CI #13:Commit 308b4c7704 pushed by gotthard_g

main

2026-09-29 00:35:32 +02:00

19s

View workflow file

experiments 2026-09-28: comparison on the unseen gold frames

CI #12:Commit d2d525487c pushed by gotthard_g

main

2026-09-29 00:35:03 +02:00

19s

View workflow file

train_run: --set KEY=VALUE overrides for Ultralytics train arguments, recorded in the ledger

CI #11:Commit 1bdf1fc6b1 pushed by gotthard_g

main

2026-09-28 21:01:58 +02:00

19s

View workflow file

run 2026-09-28_alc-v1-humanonly: gate passed, not promoted (A/B on clean gold: masks no tighter without accepted-unchanged labels, crystal worse)

CI #10:Commit ed208bff95 pushed by gotthard_g

main

2026-09-28 20:43:56 +02:00

17s

View workflow file

run 2026-09-28_alc-v1-humanonly: exported (twin without accepted-unchanged labels)

CI #9:Commit 84b5092766 pushed by gotthard_g

main

2026-09-28 20:01:25 +02:00

19s

View workflow file

runs/README.md: render

CI #8:Commit 46bfcb7ea5 pushed by gotthard_g

main

2026-09-28 19:17:52 +02:00

18s

View workflow file

run 2026-09-28_alc-v1: promoted by the gate after the mask-term fix (F1 +0.045, mask score 0.678 -> 0.710)

CI #7:Commit 7bfd799c26 pushed by gotthard_g

main

2026-09-28 19:17:49 +02:00

18s

View workflow file

gate: mask term over all gold objects (unmatched = 0), so finding more objects is not punished; tests

CI #6:Commit 3794330fc3 pushed by gotthard_g

main

2026-09-28 19:16:27 +02:00

18s

View workflow file

run 2026-09-28_alc-v1: gate_failed (F1 +0.045, mask IoU -0.022 vs the deployed model)

CI #5:Commit e8475684d4 pushed by gotthard_g

main

2026-09-28 19:12:48 +02:00

17s

View workflow file

run 2026-09-28_alc-v1: exported

CI #4:Commit e0b685507f pushed by gotthard_g

main

2026-09-28 16:53:32 +02:00

18s

View workflow file