small cluster model re-gated on the Spark in one environment: gate_passed against v3; comparison after cycle 4
CI / ledger-and-tools (push) Successful in 19s

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
jungfrauandClaude Fable 5.1 committed 2026-10-02 13:30:10 +02:00
1 parent f37604f87a
commit c7d53646ce
5 files changed
+382 -246

No files matched your search

@@ -1,54 +1,54 @@
{
"per_class": {
"loop_all": {
"tp": 185,
"fp": 18,
"fn": 7,
"precision": 0.9113300492610837,
"recall": 0.9635416666666666,
"f1": 0.9367088607594936,
"mean_mask_iou": 0.8785916632633505,
"n_matched": 185
"tp": 187,
"fp": 13,
"fn": 5,
"precision": 0.935,
"recall": 0.9739583333333334,
"f1": 0.9540816326530612,
"mean_mask_iou": 0.8972832445017794,
"n_matched": 187
},
"pin": {
"tp": 135,
"fp": 14,
"fn": 11,
"precision": 0.9060402684563759,
"recall": 0.9246575342465754,
"f1": 0.9152542372881356,
"mean_mask_iou": 0.9152912872176714,
"n_matched": 135
"tp": 137,
"fp": 9,
"fn": 9,
"precision": 0.9383561643835616,
"recall": 0.9383561643835616,
"f1": 0.9383561643835615,
"mean_mask_iou": 0.9256249811346701,
"n_matched": 137
},
"crystal": {
"tp": 37,
"fp": 15,
"fn": 29,
"precision": 0.7115384615384616,
"recall": 0.5606060606060606,
"f1": 0.6271186440677966,
"mean_mask_iou": 0.7501433709356009,
"n_matched": 37
"tp": 41,
"fp": 24,
"fn": 25,
"precision": 0.6307692307692307,
"recall": 0.6212121212121212,
"f1": 0.6259541984732825,
"mean_mask_iou": 0.7914623790036597,
"n_matched": 41
},
"loop_face": {
"tp": 143,
"fp": 26,
"fn": 29,
"precision": 0.8461538461538461,
"recall": 0.8313953488372093,
"f1": 0.8387096774193549,
"mean_mask_iou": 0.8332595270145433,
"n_matched": 143
"tp": 157,
"fp": 31,
"fn": 15,
"precision": 0.8351063829787234,
"recall": 0.9127906976744186,
"f1": 0.8722222222222221,
"mean_mask_iou": 0.8674352308419995,
"n_matched": 157
},
"ice": {
"tp": 3,
"fp": 5,
"fn": 37,
"precision": 0.375,
"recall": 0.075,
"f1": 0.12499999999999999,
"mean_mask_iou": 0.7736919879138792,
"n_matched": 3
"tp": 5,
"fp": 3,
"fn": 35,
"precision": 0.625,
"recall": 0.125,
"f1": 0.20833333333333334,
"mean_mask_iou": 0.7441614535204797,
"n_matched": 5
},
"needle": {
"tp": 0,
@@ -62,14 +62,14 @@
}
},
"overall": {
"tp": 503,
"fp": 78,
"fn": 114,
"precision": 0.8657487091222031,
"recall": 0.8152350081037277,
"f1": 0.8397328881469115,
"mean_mask_iou": 0.8654796710328908,
"n_matched": 503
"tp": 527,
"fp": 80,
"fn": 90,
"precision": 0.8682042833607908,
"recall": 0.8541329011345219,
"f1": 0.8611111111111112,
"mean_mask_iou": 0.8860734064254818,
"n_matched": 527
},
"dataset": {
"gold_images": 203,
@@ -77,17 +77,82 @@
"missing_predictions": 0,
"iou_threshold": 0.5
},
"environment": {
"host": "mx-spark-01",
"ultralytics": "8.4.160",
"torch": "2.14.0+cu130"
},
"gate": {
"baseline": "/data/user/duan_j/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"f1_delta": 0.0024,
"mask_iou_delta": -0.0492,
"mask_score": 0.7056,
"baseline": "/home/jungfrau/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"f1_delta": 0.0238,
"mask_iou_delta": 0.0021,
"mask_score": 0.7568,
"mask_score_baseline": 0.7547,
"matched_only_iou_delta": -0.0215,
"matched_only_iou_delta": -0.0009,
"class_f1_delta": {
"loop_all": -0.003,
"pin": -0.0104
"loop_all": 0.0144,
"pin": 0.0127
},
"pass": false
"pass_point_rule": true,
"rule": "paired bootstrap",
"bootstrap": {
"frames": 203,
"draws": 2000,
"f1": {
"delta": 0.0238,
"lo": 0.0038,
"hi": 0.044,
"win": 0.987
},
"mask_score": {
"delta": 0.0021,
"lo": -0.0105,
"hi": 0.0142,
"win": 0.628
},
"per_class": {
"crystal": {
"f1": {
"delta": 0.0334,
"lo": -0.0377,
"hi": 0.1081,
"win": 0.821
},
"mask_score": {
"delta": 0.0047,
"lo": -0.0673,
"hi": 0.0765,
"win": 0.552
}
}
}
},
"pass": true,
"environment_note": "verdict computed on mx-spark-01 with candidate and baseline in one environment; the cluster's own verdict (pass=False, F1 0.8397) compared its measurement with a baseline measured on the Spark"
},
"cluster_report": {
"overall": {
"tp": 503,
"fp": 78,
"fn": 114,
"precision": 0.8657487091222031,
"recall": 0.8152350081037277,
"f1": 0.8397328881469115,
"mean_mask_iou": 0.8654796710328908,
"n_matched": 503
},
"gate": {
"baseline": "/data/user/duan_j/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"f1_delta": 0.0024,
"mask_iou_delta": -0.0492,
"mask_score": 0.7056,
"mask_score_baseline": 0.7547,
"matched_only_iou_delta": -0.0215,
"class_f1_delta": {
"loop_all": -0.003,
"pin": -0.0104
},
"pass": false
}
}
}
+77 -17
View File
@@ -1,26 +1,61 @@
{
"overall": {
"tp": 503,
"fp": 78,
"fn": 114,
"precision": 0.8657487091222031,
"recall": 0.8152350081037277,
"f1": 0.8397328881469115,
"mean_mask_iou": 0.8654796710328908,
"n_matched": 503
"tp": 527,
"fp": 80,
"fn": 90,
"precision": 0.8682042833607908,
"recall": 0.8541329011345219,
"f1": 0.8611111111111112,
"mean_mask_iou": 0.8860734064254818,
"n_matched": 527
},
"gate": {
"baseline": "/data/user/duan_j/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"f1_delta": 0.0024,
"mask_iou_delta": -0.0492,
"mask_score": 0.7056,
"baseline": "/home/jungfrau/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"f1_delta": 0.0238,
"mask_iou_delta": 0.0021,
"mask_score": 0.7568,
"mask_score_baseline": 0.7547,
"matched_only_iou_delta": -0.0215,
"matched_only_iou_delta": -0.0009,
"class_f1_delta": {
"loop_all": -0.003,
"pin": -0.0104
"loop_all": 0.0144,
"pin": 0.0127
},
"pass": false
"pass_point_rule": true,
"rule": "paired bootstrap",
"bootstrap": {
"frames": 203,
"draws": 2000,
"f1": {
"delta": 0.0238,
"lo": 0.0038,
"hi": 0.044,
"win": 0.987
},
"mask_score": {
"delta": 0.0021,
"lo": -0.0105,
"hi": 0.0142,
"win": 0.628
},
"per_class": {
"crystal": {
"f1": {
"delta": 0.0334,
"lo": -0.0377,
"hi": 0.1081,
"win": 0.821
},
"mask_score": {
"delta": 0.0047,
"lo": -0.0673,
"hi": 0.0765,
"win": 0.552
}
}
}
},
"pass": true,
"environment_note": "verdict computed on mx-spark-01 with candidate and baseline in one environment; the cluster's own verdict (pass=False, F1 0.8397) compared its measurement with a baseline measured on the Spark"
},
"training": {
"epochs_run": 228,
@@ -34,5 +69,30 @@
"mAP50-95(M)": 0.4995
},
"version": "yolo26s-seg-v3-s-640px-2026-10-01",
"measured_at": "2026-10-01T04:18:50"
"measured_at": "2026-10-02T13:29:54",
"cluster": {
"overall": {
"tp": 503,
"fp": 78,
"fn": 114,
"precision": 0.8657487091222031,
"recall": 0.8152350081037277,
"f1": 0.8397328881469115,
"mean_mask_iou": 0.8654796710328908,
"n_matched": 503
},
"gate": {
"baseline": "/data/user/duan_j/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"f1_delta": 0.0024,
"mask_iou_delta": -0.0492,
"mask_score": 0.7056,
"mask_score_baseline": 0.7547,
"matched_only_iou_delta": -0.0215,
"class_f1_delta": {
"loop_all": -0.003,
"pin": -0.0104
},
"pass": false
}
}
}
+12 -6
View File
@@ -2,7 +2,7 @@
"id": "2026-09-30_v3-s-640px",
"dataset": "dataset/2026-09-29_v3",
"base_weights": "models/base/yolo26s-seg.pt",
"status": "gate_failed",
"status": "gate_passed",
"created": "2026-09-30T07:17:45",
"created_by": "jungfrau@mx-spark-01",
"hyperparams": {
@@ -24,11 +24,11 @@
},
"version": "yolo26s-seg-v3-s-640px-2026-10-01",
"metrics": {
"f1": 0.8397328881469115,
"mean_mask_iou": 0.8654796710328908,
"gate_pass": false,
"baseline": "/data/user/duan_j/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"measured_at": "2026-10-01T04:18:50"
"f1": 0.8611111111111112,
"mean_mask_iou": 0.8860734064254818,
"gate_pass": true,
"baseline": "/home/jungfrau/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json",
"measured_at": "2026-10-02T13:29:54"
},
"history": [
{
@@ -60,6 +60,12 @@
"at": "2026-10-01T04:18:50",
"by": "duan_j@gpu102",
"note": "{\"baseline\": \"/data/user/duan_j/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json\", \"f1_delta\": 0.0024, \"mask_iou_delta\": -0.0492, \"mask_score\": 0.7056, \"mask_score_baseline\": 0.7547, \"matched_only_iou_delta\": -0.0215, \"class_f1_delta\": {\"loop_all\": -0.003, \"pin\": -0.0104}, \"pass\": false}"
},
{
"status": "gate_passed",
"at": "2026-10-02T13:29:54",
"by": "jungfrau@mx-spark-01",
"note": "re-gated on the Spark in one environment against yolo26n-seg-v3-noflip-mask1-2026-09-30 (the cluster verdict compared two environments); not promoted: four times the parameters, needs a latency check at a beamline"
}
],
"job": 232023,
+6 -1
View File
@@ -17,7 +17,7 @@ Generated by `python tools/ledger.py render`; do not edit by hand.
| 2026-09-29_v3-recipeA | gate_passed | dataset/2026-09-29_v3 | models/base/yolo26n-seg.pt | yolo26n-seg-v3-recipeA-2026-09-29 | 0.831 | 0.881 | yes | 2026-09-29T23:42 |
| 2026-09-30_v3-n-1024px | exported | dataset/2026-09-29_v3 | models/base/yolo26n-seg.pt | - | - | - | - | 2026-09-30T07:17 |
| 2026-09-30_v3-s-1024px | exported | dataset/2026-09-29_v3 | models/base/yolo26s-seg.pt | - | - | - | - | 2026-09-30T07:17 |
| 2026-09-30_v3-s-640px | exported | dataset/2026-09-29_v3 | models/base/yolo26s-seg.pt | - | - | - | - | 2026-09-30T07:17 |
| 2026-09-30_v3-s-640px | gate_passed | dataset/2026-09-29_v3 | models/base/yolo26s-seg.pt | yolo26s-seg-v3-s-640px-2026-10-01 | 0.861 | 0.886 | yes | 2026-10-02T13:29 |
| 2026-10-01_v4-noflip-mask1 | promoted | dataset/2026-10-01_v4 | models/base/yolo26n-seg.pt | yolo26n-seg-v4-noflip-mask1-2026-10-02 | 0.852 | 0.891 | yes | 2026-10-02T13:25 |
| 2026-10-01_v4-recipeA | gate_failed | dataset/2026-10-01_v4 | models/base/yolo26n-seg.pt | yolo26n-seg-v4-recipeA-2026-10-02 | 0.832 | 0.884 | no | 2026-10-02T04:12 |
@@ -120,6 +120,11 @@ States: exported → queued → training → measured → gate_passed → gate_f
## 2026-09-30_v3-s-640px
- 2026-09-30T07:17:45 · exported · jungfrau@mx-spark-01 — cluster: capacity lever, small model at 640 px, default recipe
- 2026-10-01T01:30:13 · queued · duan_j@gpu102
- 2026-10-01T01:30:13 · training · duan_j@gpu102 — train_run.py
- 2026-10-01T04:18:50 · measured · duan_j@gpu102
- 2026-10-01T04:18:50 · gate_failed · duan_j@gpu102 — {"baseline": "/data/user/duan_j/AareLC_train/models/registry/yolo26n-seg-v3-noflip-mask1-2026-09-30/gold_report.json", "f1_delta": 0.0024, "mask_iou_delta": -0.0492, "mask_score": 0.7056, "mask_score_baseline": 0.7547, "matched_only_iou_delta": -0.0215, "class_f1_delta": {"loop_all": -0.003, "pin": -0.0104}, "pass": false}
- 2026-10-02T13:29:54 · gate_passed · jungfrau@mx-spark-01 — re-gated on the Spark in one environment against yolo26n-seg-v3-noflip-mask1-2026-09-30 (the cluster verdict compared two environments); not promoted: four times the parameters, needs a latency check at a beamline
## 2026-10-01_v4-noflip-mask1
+166 -166
View File
@@ -671,6 +671,73 @@
"gold_images": 149
}
},
"2026-09-30_v3-s-640px": {
"per_class": {
"loop_all": {
"tp": 134,
"fp": 10,
"fn": 5,
"precision": 0.9305555555555556,
"recall": 0.9640287769784173,
"f1": 0.9469964664310955,
"mean_mask_iou": 0.8941931738466334,
"n_matched": 134
},
"pin": {
"tp": 108,
"fp": 6,
"fn": 8,
"precision": 0.9473684210526315,
"recall": 0.9310344827586207,
"f1": 0.9391304347826087,
"mean_mask_iou": 0.9198485953616634,
"n_matched": 108
},
"crystal": {
"tp": 21,
"fp": 14,
"fn": 19,
"precision": 0.6,
"recall": 0.525,
"f1": 0.56,
"mean_mask_iou": 0.7742300919117516,
"n_matched": 21
},
"loop_face": {
"tp": 110,
"fp": 26,
"fn": 10,
"precision": 0.8088235294117647,
"recall": 0.9166666666666666,
"f1": 0.859375,
"mean_mask_iou": 0.8632894169651606,
"n_matched": 110
},
"ice": {
"tp": 2,
"fp": 1,
"fn": 8,
"precision": 0.6666666666666666,
"recall": 0.2,
"f1": 0.30769230769230765,
"mean_mask_iou": 0.8091053257565746,
"n_matched": 2
}
},
"overall": {
"tp": 375,
"fp": 57,
"fn": 50,
"precision": 0.8680555555555556,
"recall": 0.8823529411764706,
"f1": 0.8751458576429405,
"mean_mask_iou": 0.8853450987795626,
"n_matched": 375
},
"dataset": {
"gold_images": 149
}
},
"2026-10-01_v4-noflip-mask1": {
"per_class": {
"loop_all": {
@@ -804,73 +871,6 @@
"dataset": {
"gold_images": 149
}
},
"yolo26s-seg-v3-s-640px-2026-10-01": {
"per_class": {
"loop_all": {
"tp": 134,
"fp": 10,
"fn": 5,
"precision": 0.9305555555555556,
"recall": 0.9640287769784173,
"f1": 0.9469964664310955,
"mean_mask_iou": 0.8941931738466334,
"n_matched": 134
},
"pin": {
"tp": 108,
"fp": 6,
"fn": 8,
"precision": 0.9473684210526315,
"recall": 0.9310344827586207,
"f1": 0.9391304347826087,
"mean_mask_iou": 0.9198485953616634,
"n_matched": 108
},
"crystal": {
"tp": 21,
"fp": 14,
"fn": 19,
"precision": 0.6,
"recall": 0.525,
"f1": 0.56,
"mean_mask_iou": 0.7742300919117516,
"n_matched": 21
},
"loop_face": {
"tp": 110,
"fp": 26,
"fn": 10,
"precision": 0.8088235294117647,
"recall": 0.9166666666666666,
"f1": 0.859375,
"mean_mask_iou": 0.8632894169651606,
"n_matched": 110
},
"ice": {
"tp": 2,
"fp": 1,
"fn": 8,
"precision": 0.6666666666666666,
"recall": 0.2,
"f1": 0.30769230769230765,
"mean_mask_iou": 0.8091053257565746,
"n_matched": 2
}
},
"overall": {
"tp": 375,
"fp": 57,
"fn": 50,
"precision": 0.8680555555555556,
"recall": 0.8823529411764706,
"f1": 0.8751458576429405,
"mean_mask_iou": 0.8853450987795626,
"n_matched": 375
},
"dataset": {
"gold_images": 149
}
}
},
"core_frames": 149,
@@ -1546,6 +1546,73 @@
"gold_images": 149
}
},
"2026-09-30_v3-s-640px": {
"per_class": {
"loop_all": {
"tp": 134,
"fp": 10,
"fn": 5,
"precision": 0.9305555555555556,
"recall": 0.9640287769784173,
"f1": 0.9469964664310955,
"mean_mask_iou": 0.8941931738466334,
"n_matched": 134
},
"pin": {
"tp": 108,
"fp": 6,
"fn": 8,
"precision": 0.9473684210526315,
"recall": 0.9310344827586207,
"f1": 0.9391304347826087,
"mean_mask_iou": 0.9198485953616634,
"n_matched": 108
},
"crystal": {
"tp": 21,
"fp": 14,
"fn": 19,
"precision": 0.6,
"recall": 0.525,
"f1": 0.56,
"mean_mask_iou": 0.7742300919117516,
"n_matched": 21
},
"loop_face": {
"tp": 110,
"fp": 26,
"fn": 10,
"precision": 0.8088235294117647,
"recall": 0.9166666666666666,
"f1": 0.859375,
"mean_mask_iou": 0.8632894169651606,
"n_matched": 110
},
"ice": {
"tp": 2,
"fp": 1,
"fn": 8,
"precision": 0.6666666666666666,
"recall": 0.2,
"f1": 0.30769230769230765,
"mean_mask_iou": 0.8091053257565746,
"n_matched": 2
}
},
"overall": {
"tp": 375,
"fp": 57,
"fn": 50,
"precision": 0.8680555555555556,
"recall": 0.8823529411764706,
"f1": 0.8751458576429405,
"mean_mask_iou": 0.8853450987795626,
"n_matched": 375
},
"dataset": {
"gold_images": 149
}
},
"2026-10-01_v4-noflip-mask1": {
"per_class": {
"loop_all": {
@@ -1679,73 +1746,6 @@
"dataset": {
"gold_images": 149
}
},
"yolo26s-seg-v3-s-640px-2026-10-01": {
"per_class": {
"loop_all": {
"tp": 134,
"fp": 10,
"fn": 5,
"precision": 0.9305555555555556,
"recall": 0.9640287769784173,
"f1": 0.9469964664310955,
"mean_mask_iou": 0.8941931738466334,
"n_matched": 134
},
"pin": {
"tp": 108,
"fp": 6,
"fn": 8,
"precision": 0.9473684210526315,
"recall": 0.9310344827586207,
"f1": 0.9391304347826087,
"mean_mask_iou": 0.9198485953616634,
"n_matched": 108
},
"crystal": {
"tp": 21,
"fp": 14,
"fn": 19,
"precision": 0.6,
"recall": 0.525,
"f1": 0.56,
"mean_mask_iou": 0.7742300919117516,
"n_matched": 21
},
"loop_face": {
"tp": 110,
"fp": 26,
"fn": 10,
"precision": 0.8088235294117647,
"recall": 0.9166666666666666,
"f1": 0.859375,
"mean_mask_iou": 0.8632894169651606,
"n_matched": 110
},
"ice": {
"tp": 2,
"fp": 1,
"fn": 8,
"precision": 0.6666666666666666,
"recall": 0.2,
"f1": 0.30769230769230765,
"mean_mask_iou": 0.8091053257565746,
"n_matched": 2
}
},
"overall": {
"tp": 375,
"fp": 57,
"fn": 50,
"precision": 0.8680555555555556,
"recall": 0.8823529411764706,
"f1": 0.8751458576429405,
"mean_mask_iou": 0.8853450987795626,
"n_matched": 375
},
"dataset": {
"gold_images": 149
}
}
},
"active": "yolo26n-seg-v4-noflip-mask1-2026-10-02",
@@ -2070,6 +2070,38 @@
}
}
},
"2026-09-30_v3-s-640px": {
"frames": 149,
"draws": 2000,
"f1": {
"delta": 0.0042,
"lo": -0.0183,
"hi": 0.0242,
"win": 0.649
},
"mask_score": {
"delta": -0.011,
"lo": -0.0278,
"hi": 0.0042,
"win": 0.082
},
"per_class": {
"crystal": {
"f1": {
"delta": -0.001,
"lo": -0.1061,
"hi": 0.1018,
"win": 0.458
},
"mask_score": {
"delta": -0.055,
"lo": -0.1347,
"hi": 0.0031,
"win": 0.032
}
}
}
},
"2026-10-01_v4-recipeA": {
"frames": 149,
"draws": 2000,
@@ -2101,38 +2133,6 @@
}
}
}
},
"yolo26s-seg-v3-s-640px-2026-10-01": {
"frames": 149,
"draws": 2000,
"f1": {
"delta": 0.0042,
"lo": -0.0183,
"hi": 0.0242,
"win": 0.649
},
"mask_score": {
"delta": -0.011,
"lo": -0.0278,
"hi": 0.0042,
"win": 0.082
},
"per_class": {
"crystal": {
"f1": {
"delta": -0.001,
"lo": -0.1061,
"hi": 0.1018,
"win": 0.458
},
"mask_score": {
"delta": -0.055,
"lo": -0.1347,
"hi": 0.0031,
"win": 0.032
}
}
}
}
}
}