minhleduc's picture
E2 run e4_log_only_seed42 (epoch_4.pt)
386c47b verified
Raw History Blame Contribute Delete
3 kB
{
"raw": {
"ece": 0.13457706558704377,
"brier": 0.053072483953451675,
"nll": 0.8642846867744811,
"accuracy": 0.7775,
"soft_accuracy": 0.5214390063285828,
"mean_confidence": 0.6446689553260804,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.21876078649656847,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.13371521070599554,
"brier": 0.06767832974092736,
"nll": 0.9941271715487043,
"accuracy": 0.7533333333333333,
"soft_accuracy": 0.4766030919800202,
"mean_confidence": 0.6196181240181128,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.16226732674986122,
"brier": 0.05451303494450485,
"nll": 1.0379393140226603,
"accuracy": 0.7325,
"soft_accuracy": 0.44023495307192206,
"mean_confidence": 0.5787717271223665,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.21876078649656847,
"n": 800
},
"noul": {
"ece": 0.10408389608065288,
"brier": 0.036545903511238424,
"nll": 0.5029026990026857,
"accuracy": 0.8616666666666667,
"soft_accuracy": 0.6745469916860263,
"mean_confidence": 0.7575827575723331,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"post_temperature": {
"ece": 0.15136478865146635,
"brier": 0.05149772852919583,
"nll": 0.8604798654299229,
"accuracy": 0.7775,
"soft_accuracy": 0.5136348050832749,
"mean_confidence": 0.6285212074667216,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.21874093162477948,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.15615266551574072,
"brier": 0.06530533224266644,
"nll": 0.9888753090736767,
"accuracy": 0.7533333333333333,
"soft_accuracy": 0.4670632988711198,
"mean_confidence": 0.5995675005018711,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.17158496633172038,
"brier": 0.05304105402517052,
"nll": 1.0343197106011213,
"accuracy": 0.7325,
"soft_accuracy": 0.4344353336468339,
"mean_confidence": 0.5658757378160953,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.21874093162477948,
"n": 800
},
"noul": {
"ece": 0.12066448509693146,
"brier": 0.03563235748775898,
"nll": 0.5002979615579048,
"accuracy": 0.8616666666666667,
"soft_accuracy": 0.6658056065440178,
"mean_confidence": 0.7410022072990735,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"fitted_temperature": {
"type0_bucket1": 1.0873889923095703,
"type2_bucket0": 1.1389371156692505,
"type1_bucket1": 1.0593258142471313
}
}