wav2vec2-base-ft-fake-detection / trainer_state.json
alexandreacff's picture
End of training
1a62985 verified
raw
history blame
5.07 kB
{
"best_metric": 0.9906542056074766,
"best_model_checkpoint": "wav2vec2-base-ft-fake-detection/checkpoint-165",
"epoch": 4.925373134328359,
"eval_steps": 500,
"global_step": 165,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.29850746268656714,
"grad_norm": 0.7558884620666504,
"learning_rate": 1.7647058823529414e-05,
"loss": 0.6889,
"step": 10
},
{
"epoch": 0.5970149253731343,
"grad_norm": 0.29062432050704956,
"learning_rate": 2.939189189189189e-05,
"loss": 0.5495,
"step": 20
},
{
"epoch": 0.8955223880597015,
"grad_norm": 0.2452060431241989,
"learning_rate": 2.7364864864864865e-05,
"loss": 0.4897,
"step": 30
},
{
"epoch": 0.9850746268656716,
"eval_accuracy": 0.0,
"eval_loss": 1.3924530744552612,
"eval_runtime": 17.8422,
"eval_samples_per_second": 59.97,
"eval_steps_per_second": 3.755,
"step": 33
},
{
"epoch": 1.1940298507462686,
"grad_norm": 0.7244436740875244,
"learning_rate": 2.533783783783784e-05,
"loss": 0.4327,
"step": 40
},
{
"epoch": 1.4925373134328357,
"grad_norm": 1.3644465208053589,
"learning_rate": 2.331081081081081e-05,
"loss": 0.4158,
"step": 50
},
{
"epoch": 1.7910447761194028,
"grad_norm": 3.455235719680786,
"learning_rate": 2.1283783783783782e-05,
"loss": 0.3905,
"step": 60
},
{
"epoch": 2.0,
"eval_accuracy": 0.7953271028037383,
"eval_loss": 0.6337738037109375,
"eval_runtime": 20.1073,
"eval_samples_per_second": 53.214,
"eval_steps_per_second": 3.332,
"step": 67
},
{
"epoch": 2.08955223880597,
"grad_norm": 2.2601675987243652,
"learning_rate": 1.9256756756756756e-05,
"loss": 0.3468,
"step": 70
},
{
"epoch": 2.388059701492537,
"grad_norm": 1.6426903009414673,
"learning_rate": 1.722972972972973e-05,
"loss": 0.3186,
"step": 80
},
{
"epoch": 2.6865671641791042,
"grad_norm": 3.29968523979187,
"learning_rate": 1.5202702702702704e-05,
"loss": 0.3155,
"step": 90
},
{
"epoch": 2.9850746268656714,
"grad_norm": 1.9062788486480713,
"learning_rate": 1.3175675675675676e-05,
"loss": 0.3139,
"step": 100
},
{
"epoch": 2.9850746268656714,
"eval_accuracy": 0.9710280373831776,
"eval_loss": 0.4036500155925751,
"eval_runtime": 20.2064,
"eval_samples_per_second": 52.954,
"eval_steps_per_second": 3.316,
"step": 100
},
{
"epoch": 3.283582089552239,
"grad_norm": 2.5272302627563477,
"learning_rate": 1.1148648648648649e-05,
"loss": 0.297,
"step": 110
},
{
"epoch": 3.582089552238806,
"grad_norm": 2.7984001636505127,
"learning_rate": 9.121621621621622e-06,
"loss": 0.286,
"step": 120
},
{
"epoch": 3.8805970149253732,
"grad_norm": 5.105321884155273,
"learning_rate": 7.0945945945945946e-06,
"loss": 0.2777,
"step": 130
},
{
"epoch": 4.0,
"eval_accuracy": 0.9887850467289719,
"eval_loss": 0.3067161440849304,
"eval_runtime": 17.3626,
"eval_samples_per_second": 61.627,
"eval_steps_per_second": 3.859,
"step": 134
},
{
"epoch": 4.17910447761194,
"grad_norm": 2.5902462005615234,
"learning_rate": 5.067567567567567e-06,
"loss": 0.2757,
"step": 140
},
{
"epoch": 4.477611940298507,
"grad_norm": 3.8617169857025146,
"learning_rate": 3.040540540540541e-06,
"loss": 0.271,
"step": 150
},
{
"epoch": 4.776119402985074,
"grad_norm": 2.8858726024627686,
"learning_rate": 1.0135135135135136e-06,
"loss": 0.2455,
"step": 160
},
{
"epoch": 4.925373134328359,
"eval_accuracy": 0.9906542056074766,
"eval_loss": 0.2779688239097595,
"eval_runtime": 17.3475,
"eval_samples_per_second": 61.68,
"eval_steps_per_second": 3.862,
"step": 165
},
{
"epoch": 4.925373134328359,
"step": 165,
"total_flos": 1.9134126152064e+17,
"train_loss": 0.3664692994320031,
"train_runtime": 350.8638,
"train_samples_per_second": 60.95,
"train_steps_per_second": 0.47
}
],
"logging_steps": 10,
"max_steps": 165,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.9134126152064e+17,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}