minhleduc's picture
E2 run e2_rlce_sigma2_fixed_seed43 (epoch_4.pt)
5e60be5 verified
Raw History Blame Contribute Delete
2.96 kB
{
"raw": {
"ece": 0.05492016673088073,
"brier": 0.07290656918366514,
"nll": 0.9331893428541953,
"accuracy": 0.763,
"soft_accuracy": 0.5495321557377465,
"mean_confidence": 0.710992153853178,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.2455293081885611,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.06423682210346063,
"brier": 0.08723674146583714,
"nll": 1.074158520035368,
"accuracy": 0.735,
"soft_accuracy": 0.5082238023355603,
"mean_confidence": 0.6888738328715166,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.06479341212660074,
"brier": 0.08184851084640059,
"nll": 1.1235608079913073,
"accuracy": 0.72,
"soft_accuracy": 0.47136491038952955,
"mean_confidence": 0.6590714601054788,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.2455293081885611,
"n": 800
},
"noul": {
"ece": 0.04606316069761912,
"brier": 0.0466538080178459,
"nll": 0.5383915454902065,
"accuracy": 0.8483333333333334,
"soft_accuracy": 0.6950635029375554,
"mean_confidence": 0.8023380664984385,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"post_temperature": {
"ece": 0.1465617177784443,
"brier": 0.0529350912007566,
"nll": 0.8636743592321873,
"accuracy": 0.763,
"soft_accuracy": 0.50882018167153,
"mean_confidence": 0.6167448680102825,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.22862462463206612,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.15243854343891144,
"brier": 0.06480050334939733,
"nll": 0.9914538177723686,
"accuracy": 0.735,
"soft_accuracy": 0.46189232205351194,
"mean_confidence": 0.5825614453355471,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.16561398483812806,
"brier": 0.0564608580215554,
"nll": 1.0401638444233685,
"accuracy": 0.72,
"soft_accuracy": 0.42966365384869276,
"mean_confidence": 0.5566691920161247,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.22862462463206612,
"n": 800
},
"noul": {
"ece": 0.11730413367350893,
"brier": 0.03636865662438416,
"nll": 0.5005755871037643,
"accuracy": 0.8483333333333334,
"soft_accuracy": 0.6612900783866644,
"mean_confidence": 0.7310291920105616,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"fitted_temperature": {
"type0_bucket1": 1.6228762865066528,
"type2_bucket0": 1.8150746822357178,
"type1_bucket1": 1.5972267389297485
}
}