minhleduc's picture
E2 run e2_rl_only_seed44 (epoch_4.pt)
da6643e verified
Raw History Blame Contribute Delete
3 kB
{
"raw": {
"ece": 0.14623528042435646,
"brier": 0.05187493308873255,
"nll": 0.8620833359397948,
"accuracy": 0.782,
"soft_accuracy": 0.5165476535595953,
"mean_confidence": 0.6357647009789944,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.22164019856252706,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.1406357887387276,
"brier": 0.06430449025351814,
"nll": 0.9907977162860334,
"accuracy": 0.7466666666666667,
"soft_accuracy": 0.47202380614976086,
"mean_confidence": 0.6060308781266213,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.1810961751639843,
"brier": 0.05559757824222288,
"nll": 1.0381295581767336,
"accuracy": 0.75125,
"soft_accuracy": 0.43400188169442117,
"mean_confidence": 0.5704369857162237,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.22164019856252706,
"n": 800
},
"noul": {
"ece": 0.10573116928339003,
"brier": 0.0344818490526265,
"nll": 0.4986406592776378,
"accuracy": 0.8583333333333333,
"soft_accuracy": 0.6711325301229953,
"mean_confidence": 0.7526021441817283,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"post_temperature": {
"ece": 0.15405899201333523,
"brier": 0.05127986825817003,
"nll": 0.8608446306493134,
"accuracy": 0.782,
"soft_accuracy": 0.5126817736849189,
"mean_confidence": 0.6279409993737936,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.22208344614249653,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.15592424268523852,
"brier": 0.06307169253218187,
"nll": 0.9881872447952628,
"accuracy": 0.7466666666666667,
"soft_accuracy": 0.4644742188602686,
"mean_confidence": 0.5907424199084441,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.18456893473863603,
"brier": 0.05520905020358327,
"nll": 1.037406365359202,
"accuracy": 0.75125,
"soft_accuracy": 0.4323760567605495,
"mean_confidence": 0.5668293686956167,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.22208344614249653,
"n": 800
},
"noul": {
"ece": 0.11171158452828726,
"brier": 0.034249134723607216,
"nll": 0.49808637022351226,
"accuracy": 0.8583333333333333,
"soft_accuracy": 0.6679636177420616,
"mean_confidence": 0.7466217530767123,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"fitted_temperature": {
"type1_bucket1": 1.0158109664916992,
"type0_bucket1": 1.0667140483856201,
"type2_bucket0": 1.0462324619293213
}
}