minhleduc's picture
E2 run e2_rl_only_seed43 (epoch_4.pt)
58bcb06 verified
Raw History Blame Contribute Delete
2.94 kB
{
"raw": {
"ece": 0.12165459015965463,
"brier": 0.05494365088863375,
"nll": 0.8678970124274492,
"accuracy": 0.7655,
"soft_accuracy": 0.518058865621686,
"mean_confidence": 0.6438454231023788,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.23138969924882985,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.12380825623869898,
"brier": 0.06998325296962624,
"nll": 0.9994215805456043,
"accuracy": 0.735,
"soft_accuracy": 0.4717585042615732,
"mean_confidence": 0.6165745246907075,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.1429442136734724,
"brier": 0.05888521123427381,
"nll": 1.0457728155190125,
"accuracy": 0.725,
"soft_accuracy": 0.43716177746653556,
"mean_confidence": 0.5820557938143611,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.23138969924882985,
"n": 800
},
"noul": {
"ece": 0.09649752428134283,
"brier": 0.0346486350134545,
"nll": 0.4992047068538765,
"accuracy": 0.85,
"soft_accuracy": 0.672222011188666,
"mean_confidence": 0.7535024938980738,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"post_temperature": {
"ece": 0.13698417124152182,
"brier": 0.053501144541695266,
"nll": 0.8650732711162418,
"accuracy": 0.7655,
"soft_accuracy": 0.5108642288148403,
"mean_confidence": 0.6287898935526609,
"mean_target_max": 0.6589462240487337,
"score_mae": 0.2321799295663368,
"n": 2000,
"by_type": {
"choice": {
"ece": 0.14068568497896194,
"brier": 0.06812274806366986,
"nll": 0.9962770284960667,
"accuracy": 0.735,
"soft_accuracy": 0.4645478400836388,
"mean_confidence": 0.6014607128500938,
"mean_target_max": 0.6345381752649943,
"score_mae": null,
"n": 600
},
"score": {
"ece": 0.15858526431024075,
"brier": 0.05696017570102413,
"nll": 1.041912662992254,
"accuracy": 0.725,
"soft_accuracy": 0.43025937089696525,
"mean_confidence": 0.5664147282019257,
"mean_target_max": 0.5928848401084542,
"score_mae": 0.2321799295663368,
"n": 800
},
"noul": {
"ece": 0.11154344946146012,
"brier": 0.03426749947394885,
"nll": 0.4980836579017341,
"accuracy": 0.85,
"soft_accuracy": 0.6646537614365419,
"mean_confidence": 0.7392859613895416,
"mean_target_max": 0.7714361180861791,
"score_mae": null,
"n": 600
}
}
},
"fitted_temperature": {
"type0_bucket1": 1.0622789859771729,
"type2_bucket0": 1.1156989336013794,
"type1_bucket1": 1.069263219833374
}
}