xpariz10's picture
End of training
9453886
raw
history blame
8.39 kB
{
"best_metric": 0.9642857142857143,
"best_model_checkpoint": "ast-finetuned-audioset-10-10-0.4593_ft_env_aug_0-2\\checkpoint-84",
"epoch": 10.0,
"global_step": 280,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.36,
"learning_rate": 7.142857142857143e-07,
"loss": 2.1288,
"step": 10
},
{
"epoch": 0.71,
"learning_rate": 1.4285714285714286e-06,
"loss": 2.0165,
"step": 20
},
{
"epoch": 1.0,
"eval_accuracy": 0.4642857142857143,
"eval_f1": 0.4710884353741497,
"eval_loss": 1.6251839399337769,
"eval_precision": 0.5373376623376623,
"eval_recall": 0.4642857142857143,
"eval_runtime": 2.5452,
"eval_samples_per_second": 11.001,
"eval_steps_per_second": 5.501,
"step": 28
},
{
"epoch": 1.07,
"learning_rate": 1.984126984126984e-06,
"loss": 1.816,
"step": 30
},
{
"epoch": 1.43,
"learning_rate": 1.9047619047619045e-06,
"loss": 1.5466,
"step": 40
},
{
"epoch": 1.79,
"learning_rate": 1.8253968253968254e-06,
"loss": 1.3702,
"step": 50
},
{
"epoch": 2.0,
"eval_accuracy": 0.8571428571428571,
"eval_f1": 0.8535714285714285,
"eval_loss": 1.0553056001663208,
"eval_precision": 0.8928571428571429,
"eval_recall": 0.8571428571428571,
"eval_runtime": 2.5524,
"eval_samples_per_second": 10.97,
"eval_steps_per_second": 5.485,
"step": 56
},
{
"epoch": 2.14,
"learning_rate": 1.746031746031746e-06,
"loss": 1.2834,
"step": 60
},
{
"epoch": 2.5,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.9612,
"step": 70
},
{
"epoch": 2.86,
"learning_rate": 1.587301587301587e-06,
"loss": 0.8861,
"step": 80
},
{
"epoch": 3.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.6899059414863586,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.5679,
"eval_samples_per_second": 10.904,
"eval_steps_per_second": 5.452,
"step": 84
},
{
"epoch": 3.21,
"learning_rate": 1.5079365079365078e-06,
"loss": 0.7961,
"step": 90
},
{
"epoch": 3.57,
"learning_rate": 1.4285714285714286e-06,
"loss": 0.7187,
"step": 100
},
{
"epoch": 3.93,
"learning_rate": 1.3492063492063493e-06,
"loss": 0.5655,
"step": 110
},
{
"epoch": 4.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.47662997245788574,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.6044,
"eval_samples_per_second": 10.751,
"eval_steps_per_second": 5.376,
"step": 112
},
{
"epoch": 4.29,
"learning_rate": 1.2698412698412697e-06,
"loss": 0.4864,
"step": 120
},
{
"epoch": 4.64,
"learning_rate": 1.1904761904761904e-06,
"loss": 0.4195,
"step": 130
},
{
"epoch": 5.0,
"learning_rate": 1.111111111111111e-06,
"loss": 0.4232,
"step": 140
},
{
"epoch": 5.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.3402732312679291,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.6311,
"eval_samples_per_second": 10.642,
"eval_steps_per_second": 5.321,
"step": 140
},
{
"epoch": 5.36,
"learning_rate": 1.031746031746032e-06,
"loss": 0.3117,
"step": 150
},
{
"epoch": 5.71,
"learning_rate": 9.523809523809522e-07,
"loss": 0.3148,
"step": 160
},
{
"epoch": 6.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.2678673565387726,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.5813,
"eval_samples_per_second": 10.847,
"eval_steps_per_second": 5.424,
"step": 168
},
{
"epoch": 6.07,
"learning_rate": 8.73015873015873e-07,
"loss": 0.3124,
"step": 170
},
{
"epoch": 6.43,
"learning_rate": 7.936507936507936e-07,
"loss": 0.2345,
"step": 180
},
{
"epoch": 6.79,
"learning_rate": 7.142857142857143e-07,
"loss": 0.2335,
"step": 190
},
{
"epoch": 7.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.22391526401042938,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.5943,
"eval_samples_per_second": 10.793,
"eval_steps_per_second": 5.397,
"step": 196
},
{
"epoch": 7.14,
"learning_rate": 6.349206349206349e-07,
"loss": 0.2206,
"step": 200
},
{
"epoch": 7.5,
"learning_rate": 5.555555555555555e-07,
"loss": 0.1858,
"step": 210
},
{
"epoch": 7.86,
"learning_rate": 4.761904761904761e-07,
"loss": 0.176,
"step": 220
},
{
"epoch": 8.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.19793787598609924,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.6109,
"eval_samples_per_second": 10.724,
"eval_steps_per_second": 5.362,
"step": 224
},
{
"epoch": 8.21,
"learning_rate": 3.968253968253968e-07,
"loss": 0.1796,
"step": 230
},
{
"epoch": 8.57,
"learning_rate": 3.1746031746031743e-07,
"loss": 0.1475,
"step": 240
},
{
"epoch": 8.93,
"learning_rate": 2.3809523809523806e-07,
"loss": 0.1624,
"step": 250
},
{
"epoch": 9.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.1823665350675583,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.6135,
"eval_samples_per_second": 10.714,
"eval_steps_per_second": 5.357,
"step": 252
},
{
"epoch": 9.29,
"learning_rate": 1.5873015873015872e-07,
"loss": 0.1532,
"step": 260
},
{
"epoch": 9.64,
"learning_rate": 7.936507936507936e-08,
"loss": 0.1549,
"step": 270
},
{
"epoch": 10.0,
"learning_rate": 0.0,
"loss": 0.1466,
"step": 280
},
{
"epoch": 10.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.17811881005764008,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.5988,
"eval_samples_per_second": 10.774,
"eval_steps_per_second": 5.387,
"step": 280
},
{
"epoch": 10.0,
"step": 280,
"total_flos": 1.5184033474412544e+17,
"train_loss": 0.6554145208426885,
"train_runtime": 635.0581,
"train_samples_per_second": 3.527,
"train_steps_per_second": 0.441
},
{
"epoch": 10.0,
"eval_accuracy": 0.9642857142857143,
"eval_f1": 0.9631083202511775,
"eval_loss": 0.6899059414863586,
"eval_precision": 0.9693877551020408,
"eval_recall": 0.9642857142857143,
"eval_runtime": 2.7776,
"eval_samples_per_second": 10.081,
"eval_steps_per_second": 5.04,
"step": 280
},
{
"epoch": 10.0,
"step": 280,
"test_accuracy": 0.8928571428571429,
"test_f1": 0.8869047619047619,
"test_loss": 0.8998350501060486,
"test_precision": 0.9345238095238095,
"test_recall": 0.8928571428571429,
"test_runtime": 2.9018,
"test_samples_per_second": 9.649,
"test_steps_per_second": 4.825
}
],
"max_steps": 280,
"num_train_epochs": 10,
"total_flos": 1.5184033474412544e+17,
"trial_name": null,
"trial_params": null
}