roberta-base-ca-cased-ner / trainer_state.json
ccasimiro's picture
upload model
67ba514
{
"best_metric": 0.8867365091915398,
"best_model_checkpoint": "/home/shared/dt01/temutauro/ccasimiro/berta/src/finetuning/ner/roberta-base-ca-cased-ner/checkpoint-2656",
"epoch": 9.998496240601504,
"global_step": 3320,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_accuracy": 0.9859966873884145,
"eval_f1": 0.8615444617784711,
"eval_loss": 0.04728459194302559,
"eval_precision": 0.8486361890126777,
"eval_recall": 0.8748514851485149,
"eval_runtime": 10.4518,
"eval_samples_per_second": 136.723,
"step": 332
},
{
"epoch": 1.51,
"learning_rate": 4.2469879518072294e-05,
"loss": 0.1234,
"step": 500
},
{
"epoch": 2.0,
"eval_accuracy": 0.9875884617866592,
"eval_f1": 0.8732558139534884,
"eval_loss": 0.043166399002075195,
"eval_precision": 0.8550284629981024,
"eval_recall": 0.8922772277227723,
"eval_runtime": 10.4813,
"eval_samples_per_second": 136.338,
"step": 664
},
{
"epoch": 3.0,
"eval_accuracy": 0.988147733872529,
"eval_f1": 0.8809570504020395,
"eval_loss": 0.041024718433618546,
"eval_precision": 0.8725718725718725,
"eval_recall": 0.8895049504950495,
"eval_runtime": 10.4962,
"eval_samples_per_second": 136.145,
"step": 996
},
{
"epoch": 3.01,
"learning_rate": 3.4939759036144585e-05,
"loss": 0.025,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.9883628385209404,
"eval_f1": 0.8852136247292774,
"eval_loss": 0.04351422190666199,
"eval_precision": 0.880187940485513,
"eval_recall": 0.8902970297029703,
"eval_runtime": 10.4692,
"eval_samples_per_second": 136.496,
"step": 1328
},
{
"epoch": 4.52,
"learning_rate": 2.7409638554216873e-05,
"loss": 0.0118,
"step": 1500
},
{
"epoch": 5.0,
"eval_accuracy": 0.9873733571382478,
"eval_f1": 0.8842647641696393,
"eval_loss": 0.05047034099698067,
"eval_precision": 0.8849662832209441,
"eval_recall": 0.8835643564356436,
"eval_runtime": 10.5711,
"eval_samples_per_second": 135.179,
"step": 1660
},
{
"epoch": 6.0,
"eval_accuracy": 0.9880616920131644,
"eval_f1": 0.8832807570977917,
"eval_loss": 0.05266088992357254,
"eval_precision": 0.8794660384766392,
"eval_recall": 0.8871287128712871,
"eval_runtime": 10.4803,
"eval_samples_per_second": 136.351,
"step": 1992
},
{
"epoch": 6.02,
"learning_rate": 1.9879518072289157e-05,
"loss": 0.0073,
"step": 2000
},
{
"epoch": 7.0,
"eval_accuracy": 0.9878896082944353,
"eval_f1": 0.8827694728560189,
"eval_loss": 0.054938483983278275,
"eval_precision": 0.876905041031653,
"eval_recall": 0.8887128712871287,
"eval_runtime": 10.3993,
"eval_samples_per_second": 137.414,
"step": 2324
},
{
"epoch": 7.53,
"learning_rate": 1.2349397590361447e-05,
"loss": 0.0045,
"step": 2500
},
{
"epoch": 8.0,
"eval_accuracy": 0.9880401815483233,
"eval_f1": 0.8867365091915398,
"eval_loss": 0.057308126240968704,
"eval_precision": 0.8851617995264404,
"eval_recall": 0.8883168316831683,
"eval_runtime": 10.4397,
"eval_samples_per_second": 136.881,
"step": 2656
},
{
"epoch": 9.0,
"eval_accuracy": 0.9883843489857815,
"eval_f1": 0.8863502068150484,
"eval_loss": 0.05588913336396217,
"eval_precision": 0.8816614420062696,
"eval_recall": 0.8910891089108911,
"eval_runtime": 10.5436,
"eval_samples_per_second": 135.533,
"step": 2988
},
{
"epoch": 9.04,
"learning_rate": 4.819277108433735e-06,
"loss": 0.003,
"step": 3000
},
{
"epoch": 10.0,
"eval_accuracy": 0.9883413280560993,
"eval_f1": 0.8860410094637224,
"eval_loss": 0.05681716278195381,
"eval_precision": 0.8822143698468787,
"eval_recall": 0.8899009900990099,
"eval_runtime": 10.4599,
"eval_samples_per_second": 136.617,
"step": 3320
},
{
"epoch": 10.0,
"step": 3320,
"total_flos": 0,
"train_runtime": 3184.2872,
"train_samples_per_second": 1.043
}
],
"max_steps": 3320,
"num_train_epochs": 10,
"total_flos": 0,
"trial_name": null,
"trial_params": null
}