minhleduc's picture
e2_rlce_sigma1_fixed_seed43: test_eval_v2.json
630bd53 verified
Raw History Blame Contribute Delete
2.97 kB
{
"raw": {
"ece": 0.08768837297707795,
"brier": 0.06656727136412512,
"nll": 0.892609149595024,
"accuracy": 0.7615,
"soft_accuracy": 0.5317894631270319,
"mean_confidence": 0.6775622668862343,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.24285939731518738,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.08936514576276143,
"brier": 0.08215777321688468,
"nll": 1.0255133604661872,
"accuracy": 0.73,
"soft_accuracy": 0.48657992063711086,
"mean_confidence": 0.6510106047491232,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.10235176857560872,
"brier": 0.07003681775919177,
"nll": 1.0722642303165049,
"accuracy": 0.72375,
"soft_accuracy": 0.45388750651385634,
"mean_confidence": 0.623098932094872,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.24285939731518738,
"n": 800
},
"noul": {
"ece": 0.06660163760185242,
"brier": 0.04635070765127668,
"nll": 0.5201648310952199,
"accuracy": 0.8433333333333334,
"soft_accuracy": 0.680868281101187,
"mean_confidence": 0.7767317087451617,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"post_temperature": {
"ece": 0.1403249937593937,
"brier": 0.05714447159732477,
"nll": 0.8703827581712976,
"accuracy": 0.7615,
"soft_accuracy": 0.5068629283048213,
"mean_confidence": 0.6229879706948995,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.239818540236447,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.1397575196127097,
"brier": 0.07136482296457568,
"nll": 1.0011731689423322,
"accuracy": 0.73,
"soft_accuracy": 0.46079019087056317,
"mean_confidence": 0.5944797937572003,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.1615325337834656,
"brier": 0.05838666493040364,
"nll": 1.045272371917963,
"accuracy": 0.72375,
"soft_accuracy": 0.4283607154991478,
"mean_confidence": 0.5635718555748462,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.239818540236447,
"n": 800
},
"noul": {
"ece": 0.11261570334434509,
"brier": 0.04126786245263536,
"nll": 0.5064061957380424,
"accuracy": 0.8433333333333334,
"soft_accuracy": 0.6576052828133107,
"mean_confidence": 0.7307176344593366,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"fitted_temperature": {
"type0_bucket1": 1.2617459297180176,
"type2_bucket0": 1.441089391708374,
"type1_bucket1": 1.2855894565582275
}
}