BERT-AI-Vacancy / checkpoint-200 /trainer_state.json
MrAlexGov's picture
Upload folder using huggingface_hub
578704b verified
{
"best_metric": 0.45051154494285583,
"best_model_checkpoint": "BERT-AI-Vacancy/checkpoint-200",
"epoch": 2.0,
"eval_steps": 500,
"global_step": 200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"grad_norm": 8.434496879577637,
"learning_rate": 8.333333333333334e-06,
"loss": 0.6593,
"step": 5
},
{
"epoch": 0.1,
"grad_norm": 7.955150127410889,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.7268,
"step": 10
},
{
"epoch": 0.15,
"grad_norm": 8.903063774108887,
"learning_rate": 2.5e-05,
"loss": 0.6602,
"step": 15
},
{
"epoch": 0.2,
"grad_norm": 12.550924301147461,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.5844,
"step": 20
},
{
"epoch": 0.25,
"grad_norm": 8.15670108795166,
"learning_rate": 4.166666666666667e-05,
"loss": 0.5748,
"step": 25
},
{
"epoch": 0.3,
"grad_norm": 4.381443500518799,
"learning_rate": 5e-05,
"loss": 0.4641,
"step": 30
},
{
"epoch": 0.35,
"grad_norm": 5.676944255828857,
"learning_rate": 4.9074074074074075e-05,
"loss": 0.7415,
"step": 35
},
{
"epoch": 0.4,
"grad_norm": 9.691534996032715,
"learning_rate": 4.814814814814815e-05,
"loss": 0.5794,
"step": 40
},
{
"epoch": 0.45,
"grad_norm": 15.755120277404785,
"learning_rate": 4.722222222222222e-05,
"loss": 0.5693,
"step": 45
},
{
"epoch": 0.5,
"grad_norm": 10.331628799438477,
"learning_rate": 4.62962962962963e-05,
"loss": 0.5301,
"step": 50
},
{
"epoch": 0.55,
"grad_norm": 4.5462141036987305,
"learning_rate": 4.5370370370370374e-05,
"loss": 0.3862,
"step": 55
},
{
"epoch": 0.6,
"grad_norm": 5.572924613952637,
"learning_rate": 4.4444444444444447e-05,
"loss": 0.6586,
"step": 60
},
{
"epoch": 0.65,
"grad_norm": 6.483484745025635,
"learning_rate": 4.351851851851852e-05,
"loss": 0.5363,
"step": 65
},
{
"epoch": 0.7,
"grad_norm": 7.233407974243164,
"learning_rate": 4.259259259259259e-05,
"loss": 0.4871,
"step": 70
},
{
"epoch": 0.75,
"grad_norm": 4.731393814086914,
"learning_rate": 4.166666666666667e-05,
"loss": 0.4605,
"step": 75
},
{
"epoch": 0.8,
"grad_norm": 5.989573001861572,
"learning_rate": 4.074074074074074e-05,
"loss": 0.4152,
"step": 80
},
{
"epoch": 0.85,
"grad_norm": 24.533870697021484,
"learning_rate": 3.981481481481482e-05,
"loss": 0.4753,
"step": 85
},
{
"epoch": 0.9,
"grad_norm": 3.738238573074341,
"learning_rate": 3.888888888888889e-05,
"loss": 0.5456,
"step": 90
},
{
"epoch": 0.95,
"grad_norm": 15.251642227172852,
"learning_rate": 3.7962962962962964e-05,
"loss": 0.3913,
"step": 95
},
{
"epoch": 1.0,
"grad_norm": 11.198273658752441,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.5346,
"step": 100
},
{
"epoch": 1.0,
"eval_accuracy": 0.815,
"eval_auc": 0.9108999999999999,
"eval_f1": 0.8310502283105022,
"eval_loss": 0.4921284019947052,
"eval_precision": 0.7647058823529411,
"eval_recall": 0.91,
"eval_runtime": 29.8706,
"eval_samples_per_second": 6.696,
"eval_steps_per_second": 0.435,
"step": 100
},
{
"epoch": 1.05,
"grad_norm": 5.278227806091309,
"learning_rate": 3.611111111111111e-05,
"loss": 0.4063,
"step": 105
},
{
"epoch": 1.1,
"grad_norm": 8.036142349243164,
"learning_rate": 3.518518518518519e-05,
"loss": 0.4221,
"step": 110
},
{
"epoch": 1.15,
"grad_norm": 9.602019309997559,
"learning_rate": 3.425925925925926e-05,
"loss": 0.495,
"step": 115
},
{
"epoch": 1.2,
"grad_norm": 8.248969078063965,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.2828,
"step": 120
},
{
"epoch": 1.25,
"grad_norm": 4.021635055541992,
"learning_rate": 3.240740740740741e-05,
"loss": 0.131,
"step": 125
},
{
"epoch": 1.3,
"grad_norm": 10.284868240356445,
"learning_rate": 3.148148148148148e-05,
"loss": 0.5376,
"step": 130
},
{
"epoch": 1.35,
"grad_norm": 11.141060829162598,
"learning_rate": 3.055555555555556e-05,
"loss": 0.7297,
"step": 135
},
{
"epoch": 1.4,
"grad_norm": 14.501387596130371,
"learning_rate": 2.962962962962963e-05,
"loss": 0.2696,
"step": 140
},
{
"epoch": 1.45,
"grad_norm": 9.892600059509277,
"learning_rate": 2.8703703703703706e-05,
"loss": 0.3433,
"step": 145
},
{
"epoch": 1.5,
"grad_norm": 4.6811137199401855,
"learning_rate": 2.777777777777778e-05,
"loss": 0.5032,
"step": 150
},
{
"epoch": 1.55,
"grad_norm": 1.6397993564605713,
"learning_rate": 2.6851851851851855e-05,
"loss": 0.2801,
"step": 155
},
{
"epoch": 1.6,
"grad_norm": 5.426936149597168,
"learning_rate": 2.5925925925925925e-05,
"loss": 0.5105,
"step": 160
},
{
"epoch": 1.65,
"grad_norm": 2.963355779647827,
"learning_rate": 2.5e-05,
"loss": 0.5286,
"step": 165
},
{
"epoch": 1.7,
"grad_norm": 6.488004684448242,
"learning_rate": 2.4074074074074074e-05,
"loss": 0.674,
"step": 170
},
{
"epoch": 1.75,
"grad_norm": 11.268607139587402,
"learning_rate": 2.314814814814815e-05,
"loss": 0.6404,
"step": 175
},
{
"epoch": 1.8,
"grad_norm": 5.970183849334717,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.4052,
"step": 180
},
{
"epoch": 1.85,
"grad_norm": 6.261168003082275,
"learning_rate": 2.1296296296296296e-05,
"loss": 0.4296,
"step": 185
},
{
"epoch": 1.9,
"grad_norm": 4.376778602600098,
"learning_rate": 2.037037037037037e-05,
"loss": 0.203,
"step": 190
},
{
"epoch": 1.95,
"grad_norm": 4.808477878570557,
"learning_rate": 1.9444444444444445e-05,
"loss": 0.2788,
"step": 195
},
{
"epoch": 2.0,
"grad_norm": 3.679661512374878,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.1991,
"step": 200
},
{
"epoch": 2.0,
"eval_accuracy": 0.835,
"eval_auc": 0.9207,
"eval_f1": 0.8405797101449275,
"eval_loss": 0.45051154494285583,
"eval_precision": 0.8130841121495327,
"eval_recall": 0.87,
"eval_runtime": 30.0166,
"eval_samples_per_second": 6.663,
"eval_steps_per_second": 0.433,
"step": 200
}
],
"logging_steps": 5,
"max_steps": 300,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 5,
"early_stopping_threshold": 0.01
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 105244422144000.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}