minhleduc's picture
E2 run e4_log_sph_seed42 (epoch_4.pt)
77351f3 verified
Raw History Blame Contribute Delete
2.95 kB
{
"raw": {
"ece": 0.13688878096640109,
"brier": 0.05333632830852582,
"nll": 0.8655621116682887,
"accuracy": 0.7745,
"soft_accuracy": 0.5185397755280137,
"mean_confidence": 0.6392160383313894,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.2247641178953927,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.1493766143421332,
"brier": 0.06701280505378539,
"nll": 0.9947354776660601,
"accuracy": 0.755,
"soft_accuracy": 0.47321102102597556,
"mean_confidence": 0.6113708821932474,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.16540088813751935,
"brier": 0.05504629719041077,
"nll": 1.0401792683731765,
"accuracy": 0.73625,
"soft_accuracy": 0.4377902174182236,
"mean_confidence": 0.5740311346575617,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.2247641178953927,
"n": 800
},
"noul": {
"ece": 0.10002204368511836,
"brier": 0.03737989305408632,
"nll": 0.5035658700640003,
"accuracy": 0.845,
"soft_accuracy": 0.671534607509772,
"mean_confidence": 0.7539743993679683,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"post_temperature": {
"ece": 0.15061623822152614,
"brier": 0.052181859376831695,
"nll": 0.8627431136853992,
"accuracy": 0.7745,
"soft_accuracy": 0.5119791107401251,
"mean_confidence": 0.625746834486723,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.2250523244496435,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.16907053758700694,
"brier": 0.06521383934873787,
"nll": 0.9906951957133909,
"accuracy": 0.755,
"soft_accuracy": 0.4637976591785749,
"mean_confidence": 0.5920462933182716,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.17583330307155845,
"brier": 0.053951102091104985,
"nll": 1.0373363818135113,
"accuracy": 0.73625,
"soft_accuracy": 0.4329697924293578,
"mean_confidence": 0.5634086280316114,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.2250523244496435,
"n": 800
},
"noul": {
"ece": 0.11166821002960206,
"brier": 0.0367908891192278,
"nll": 0.5020000074865917,
"accuracy": 0.845,
"soft_accuracy": 0.6655063200493654,
"mean_confidence": 0.7425649842619896,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"fitted_temperature": {
"type0_bucket1": 1.0848711729049683,
"type2_bucket0": 1.0921294689178467,
"type1_bucket1": 1.0488768815994263
}
}