minhleduc's picture
E2 run e2_rlce_sigma2_fixed_seed44 (epoch_4.pt)
3293e5e verified
Raw History Blame Contribute Delete
2.97 kB
{
"raw": {
"ece": 0.06969134698808194,
"brier": 0.06848202811965921,
"nll": 0.9191458517962601,
"accuracy": 0.769,
"soft_accuracy": 0.5467977187028155,
"mean_confidence": 0.700952344879508,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.23650429474480916,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.06878897140423458,
"brier": 0.08452909949666605,
"nll": 1.0608746462652925,
"accuracy": 0.7433333333333333,
"soft_accuracy": 0.5071625221769015,
"mean_confidence": 0.683260160634915,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.08799861971288919,
"brier": 0.07404621525571883,
"nll": 1.1014315049257128,
"accuracy": 0.735,
"soft_accuracy": 0.4687326293787919,
"mean_confidence": 0.6479248089343309,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.23650429474480916,
"n": 800
},
"noul": {
"ece": 0.051088629961013786,
"brier": 0.04501604056123959,
"nll": 0.5343695198212906,
"accuracy": 0.84,
"soft_accuracy": 0.6905197009940942,
"mean_confidence": 0.7893479103843372,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"post_temperature": {
"ece": 0.15468089970946314,
"brier": 0.051865645535215325,
"nll": 0.8617692014453933,
"accuracy": 0.769,
"soft_accuracy": 0.5089197827856988,
"mean_confidence": 0.6143190894126892,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.22323823970276863,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.17026609112819033,
"brier": 0.063276015626822,
"nll": 0.9874198990563552,
"accuracy": 0.7433333333333333,
"soft_accuracy": 0.4583099640657504,
"mean_confidence": 0.5730672472715378,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.17962630096822976,
"brier": 0.05393934952667223,
"nll": 1.0370397488633172,
"accuracy": 0.735,
"soft_accuracy": 0.4302080630278215,
"mean_confidence": 0.5553737106919289,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.22323823970276863,
"n": 800
},
"noul": {
"ece": 0.1085690744717916,
"brier": 0.03769033678833279,
"nll": 0.5024244406105329,
"accuracy": 0.84,
"soft_accuracy": 0.6644785611828168,
"mean_confidence": 0.7341647698481878,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"fitted_temperature": {
"type1_bucket1": 1.5265471935272217,
"type0_bucket1": 1.6575921773910522,
"type2_bucket0": 1.6345990896224976
}
}