gipf-zero / final-vs-served.json
wuschelschulz's picture
Publish completed overnight evaluation report
55e998f verified
Raw History Blame Contribute Delete
13.9 kB
{
"candidate": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt",
"champion": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/served-champion-d5a7e3bda296.pt",
"hashes": [
"cb19bc298969043987463b267b909ca38c7ead84bab3c0b46571b6ef30e6d00f",
"d5a7e3bda2966a79d22cf34f131976907260e679683d0e2680d935312f41c2f6"
],
"models": [
{
"games": 93916,
"iteration": 2873,
"config": {
"kind": "hexresnet",
"width": 32,
"blocks": 2,
"head": "flat"
}
},
{
"games": 43869,
"iteration": 727,
"config": {
"kind": "resnet",
"width": 32,
"blocks": 2,
"head": "flat"
}
}
],
"wins": 38,
"losses": 42,
"cutoffs": 0,
"unfinished": 0,
"win_rate_decisive": 0.475,
"wilson_95_decisive": [
0.36924733297407575,
0.5830435735027006
],
"candidate_root_simulations": {
"count": 3526,
"mean": 251.41775382870108,
"min": 44,
"max": 10000
},
"champion_root_simulations": {
"count": 3523,
"mean": 243.38575078058471,
"min": 41,
"max": 10000
},
"candidate_search_seconds": {
"count": 3526,
"mean": 0.05023197256244182,
"min": 0.02666553482413292,
"max": 0.06883102655410767
},
"champion_search_seconds": {
"count": 3523,
"mean": 0.05029253078465464,
"min": 0.02774936333298683,
"max": 0.10136080533266068
},
"elapsed_seconds": 354.87011195346713,
"config": {
"candidate": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/selected-final-cb19bc298969.pt",
"champion": "/workspace/gipf/checkpoints/overnight/20260909T235853Z/served-champion-d5a7e3bda296.pt",
"games": 80,
"batch": 1,
"simulations": 10000,
"seed": 950001,
"max_ply": 300,
"seconds": 600,
"output": "/workspace/gipf/reports/overnight/20260909T235853Z/final-vs-served.json",
"device": "cpu",
"threads": 1,
"budget_ms": 50.0
},
"records": [
{
"game": 0,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 44
},
{
"game": 1,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 66
},
{
"game": 2,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 98
},
{
"game": 3,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 40
},
{
"game": 4,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 122
},
{
"game": 5,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 37
},
{
"game": 6,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 39
},
{
"game": 7,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 48
},
{
"game": 8,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 87
},
{
"game": 9,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 59
},
{
"game": 10,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 108
},
{
"game": 11,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 67
},
{
"game": 12,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 75
},
{
"game": 13,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 79
},
{
"game": 14,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 66
},
{
"game": 15,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 111
},
{
"game": 16,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 102
},
{
"game": 17,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 49
},
{
"game": 18,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 92
},
{
"game": 19,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 86
},
{
"game": 20,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 96
},
{
"game": 21,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 84
},
{
"game": 22,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 118
},
{
"game": 23,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 283
},
{
"game": 24,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 46
},
{
"game": 25,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 79
},
{
"game": 26,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 66
},
{
"game": 27,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 63
},
{
"game": 28,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 117
},
{
"game": 29,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 81
},
{
"game": 30,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 87
},
{
"game": 31,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 97
},
{
"game": 32,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 48
},
{
"game": 33,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 64
},
{
"game": 34,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 67
},
{
"game": 35,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 98
},
{
"game": 36,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 67
},
{
"game": 37,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 66
},
{
"game": 38,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 93
},
{
"game": 39,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 73
},
{
"game": 40,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 77
},
{
"game": 41,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 59
},
{
"game": 42,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 66
},
{
"game": 43,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 68
},
{
"game": 44,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 42
},
{
"game": 45,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 80
},
{
"game": 46,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 65
},
{
"game": 47,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 73
},
{
"game": 48,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 37
},
{
"game": 49,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 159
},
{
"game": 50,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 76
},
{
"game": 51,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 62
},
{
"game": 52,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 66
},
{
"game": 53,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 80
},
{
"game": 54,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 70
},
{
"game": 55,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 66
},
{
"game": 56,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 41
},
{
"game": 57,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 62
},
{
"game": 58,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 96
},
{
"game": 59,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 72
},
{
"game": 60,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 79
},
{
"game": 61,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 71
},
{
"game": 62,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 79
},
{
"game": 63,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 73
},
{
"game": 64,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 89
},
{
"game": 65,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 35
},
{
"game": 66,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 59
},
{
"game": 67,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 57
},
{
"game": 68,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 56
},
{
"game": 69,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 84
},
{
"game": 70,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 84
},
{
"game": 71,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 88
},
{
"game": 72,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 49
},
{
"game": 73,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 82
},
{
"game": 74,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 89
},
{
"game": 75,
"candidate_color": -1,
"winner": -1,
"outcome": "win",
"cutoff_reason": null,
"ply": 80
},
{
"game": 76,
"candidate_color": 1,
"winner": -1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 38
},
{
"game": 77,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 115
},
{
"game": 78,
"candidate_color": 1,
"winner": 1,
"outcome": "win",
"cutoff_reason": null,
"ply": 79
},
{
"game": 79,
"candidate_color": -1,
"winner": 1,
"outcome": "loss",
"cutoff_reason": null,
"ply": 73
}
]
}