wav2vec2-base-EMOPIA / trainer_state.json
kurosekurose's picture
End of training
896cb83
raw
history blame
6.1 kB
{
"best_metric": 0.7735137939453125,
"best_model_checkpoint": "kurosekurose/wav2vec2-base-EMOPIA/checkpoint-348",
"epoch": 14.994285714285715,
"global_step": 870,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.99,
"learning_rate": 1.9333333333333336e-06,
"loss": 1.3867,
"step": 58
},
{
"epoch": 0.99,
"eval_accuracy": 0.3285714387893677,
"eval_loss": 1.3517659902572632,
"eval_runtime": 46.5999,
"eval_samples_per_second": 1.502,
"eval_steps_per_second": 0.386,
"step": 58
},
{
"epoch": 1.99,
"learning_rate": 3.866666666666667e-06,
"loss": 1.2842,
"step": 116
},
{
"epoch": 1.99,
"eval_accuracy": 0.4571428596973419,
"eval_loss": 1.18719482421875,
"eval_runtime": 46.74,
"eval_samples_per_second": 1.498,
"eval_steps_per_second": 0.385,
"step": 116
},
{
"epoch": 2.99,
"learning_rate": 5.8e-06,
"loss": 1.0725,
"step": 174
},
{
"epoch": 2.99,
"eval_accuracy": 0.5571428537368774,
"eval_loss": 1.0476346015930176,
"eval_runtime": 46.9576,
"eval_samples_per_second": 1.491,
"eval_steps_per_second": 0.383,
"step": 174
},
{
"epoch": 3.99,
"learning_rate": 7.733333333333334e-06,
"loss": 0.9343,
"step": 232
},
{
"epoch": 3.99,
"eval_accuracy": 0.5714285969734192,
"eval_loss": 0.9676682949066162,
"eval_runtime": 46.68,
"eval_samples_per_second": 1.5,
"eval_steps_per_second": 0.386,
"step": 232
},
{
"epoch": 4.99,
"learning_rate": 9.666666666666667e-06,
"loss": 0.8053,
"step": 290
},
{
"epoch": 4.99,
"eval_accuracy": 0.6142857074737549,
"eval_loss": 0.9524909257888794,
"eval_runtime": 46.7102,
"eval_samples_per_second": 1.499,
"eval_steps_per_second": 0.385,
"step": 290
},
{
"epoch": 5.99,
"learning_rate": 9.157894736842105e-06,
"loss": 0.7895,
"step": 348
},
{
"epoch": 5.99,
"eval_accuracy": 0.6857143044471741,
"eval_loss": 0.7735137939453125,
"eval_runtime": 46.9896,
"eval_samples_per_second": 1.49,
"eval_steps_per_second": 0.383,
"step": 348
},
{
"epoch": 6.99,
"learning_rate": 8.140350877192983e-06,
"loss": 0.6867,
"step": 406
},
{
"epoch": 6.99,
"eval_accuracy": 0.6428571343421936,
"eval_loss": 0.8555552363395691,
"eval_runtime": 46.718,
"eval_samples_per_second": 1.498,
"eval_steps_per_second": 0.385,
"step": 406
},
{
"epoch": 7.99,
"learning_rate": 7.140350877192983e-06,
"loss": 0.6218,
"step": 464
},
{
"epoch": 7.99,
"eval_accuracy": 0.6714285612106323,
"eval_loss": 0.8454030156135559,
"eval_runtime": 46.4466,
"eval_samples_per_second": 1.507,
"eval_steps_per_second": 0.388,
"step": 464
},
{
"epoch": 8.99,
"learning_rate": 6.140350877192983e-06,
"loss": 0.558,
"step": 522
},
{
"epoch": 8.99,
"eval_accuracy": 0.6571428775787354,
"eval_loss": 0.8404667973518372,
"eval_runtime": 46.7611,
"eval_samples_per_second": 1.497,
"eval_steps_per_second": 0.385,
"step": 522
},
{
"epoch": 9.99,
"learning_rate": 5.12280701754386e-06,
"loss": 0.5033,
"step": 580
},
{
"epoch": 9.99,
"eval_accuracy": 0.6285714507102966,
"eval_loss": 1.0190438032150269,
"eval_runtime": 47.1353,
"eval_samples_per_second": 1.485,
"eval_steps_per_second": 0.382,
"step": 580
},
{
"epoch": 10.99,
"learning_rate": 4.105263157894737e-06,
"loss": 0.4403,
"step": 638
},
{
"epoch": 10.99,
"eval_accuracy": 0.699999988079071,
"eval_loss": 0.82489013671875,
"eval_runtime": 46.7633,
"eval_samples_per_second": 1.497,
"eval_steps_per_second": 0.385,
"step": 638
},
{
"epoch": 11.99,
"learning_rate": 3.0877192982456146e-06,
"loss": 0.3995,
"step": 696
},
{
"epoch": 11.99,
"eval_accuracy": 0.7142857313156128,
"eval_loss": 0.8996997475624084,
"eval_runtime": 46.7333,
"eval_samples_per_second": 1.498,
"eval_steps_per_second": 0.385,
"step": 696
},
{
"epoch": 12.99,
"learning_rate": 2.0701754385964913e-06,
"loss": 0.3534,
"step": 754
},
{
"epoch": 12.99,
"eval_accuracy": 0.699999988079071,
"eval_loss": 0.9177280068397522,
"eval_runtime": 46.7502,
"eval_samples_per_second": 1.497,
"eval_steps_per_second": 0.385,
"step": 754
},
{
"epoch": 13.99,
"learning_rate": 1.0526315789473685e-06,
"loss": 0.3023,
"step": 812
},
{
"epoch": 13.99,
"eval_accuracy": 0.6571428775787354,
"eval_loss": 0.9544177651405334,
"eval_runtime": 46.7592,
"eval_samples_per_second": 1.497,
"eval_steps_per_second": 0.385,
"step": 812
},
{
"epoch": 14.99,
"learning_rate": 3.508771929824562e-08,
"loss": 0.2752,
"step": 870
},
{
"epoch": 14.99,
"eval_accuracy": 0.699999988079071,
"eval_loss": 0.9346058964729309,
"eval_runtime": 46.4397,
"eval_samples_per_second": 1.507,
"eval_steps_per_second": 0.388,
"step": 870
},
{
"epoch": 14.99,
"step": 870,
"total_flos": 4.99147314409986e+18,
"train_loss": 0.6941862314596944,
"train_runtime": 20182.4464,
"train_samples_per_second": 0.52,
"train_steps_per_second": 0.043
}
],
"max_steps": 870,
"num_train_epochs": 15,
"total_flos": 4.99147314409986e+18,
"trial_name": null,
"trial_params": null
}