End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +779 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.5773195876288659,
|
| 3 |
+
"total_flos": 1.9445707177361142e+18,
|
| 4 |
+
"train_loss": 0.16865910078798022,
|
| 5 |
+
"train_runtime": 17827.0865,
|
| 6 |
+
"train_samples_per_second": 34.802,
|
| 7 |
+
"train_steps_per_second": 0.068
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.5773195876288659,
|
| 3 |
+
"total_flos": 1.9445707177361142e+18,
|
| 4 |
+
"train_loss": 0.16865910078798022,
|
| 5 |
+
"train_runtime": 17827.0865,
|
| 6 |
+
"train_samples_per_second": 34.802,
|
| 7 |
+
"train_steps_per_second": 0.068
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,779 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.921559293793824,
|
| 3 |
+
"best_model_checkpoint": "/kaggle/working/wav2vec2_ASV_deepfake_audio_detection_DF_finetune_frozen_chngd_classifier/checkpoint-40",
|
| 4 |
+
"epoch": 0.5773195876288659,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 140,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.020618556701030927,
|
| 13 |
+
"grad_norm": 88254.15625,
|
| 14 |
+
"learning_rate": 2.066115702479339e-05,
|
| 15 |
+
"loss": 0.6678,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.020618556701030927,
|
| 20 |
+
"eval_EER": 0.07174887892357559,
|
| 21 |
+
"eval_FN": 1044,
|
| 22 |
+
"eval_FP": 131,
|
| 23 |
+
"eval_TN": 17782,
|
| 24 |
+
"eval_TP": 963,
|
| 25 |
+
"eval_accuracy": 0.9410140562248996,
|
| 26 |
+
"eval_auc_roc": 0.9728919807303144,
|
| 27 |
+
"eval_f1": 0.9330636305238397,
|
| 28 |
+
"eval_loss": 0.6581301689147949,
|
| 29 |
+
"eval_min_tDCF": 0.032235667877217884,
|
| 30 |
+
"eval_precision": 0.9380674845291241,
|
| 31 |
+
"eval_recall": 0.9410140562248996,
|
| 32 |
+
"eval_runtime": 566.1185,
|
| 33 |
+
"eval_samples_per_second": 35.187,
|
| 34 |
+
"eval_steps_per_second": 0.276,
|
| 35 |
+
"step": 5
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.041237113402061855,
|
| 39 |
+
"grad_norm": 76395.140625,
|
| 40 |
+
"learning_rate": 4.132231404958678e-05,
|
| 41 |
+
"loss": 0.6124,
|
| 42 |
+
"step": 10
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.041237113402061855,
|
| 46 |
+
"eval_EER": 0.19780767314400322,
|
| 47 |
+
"eval_FN": 1862,
|
| 48 |
+
"eval_FP": 11,
|
| 49 |
+
"eval_TN": 17902,
|
| 50 |
+
"eval_TP": 145,
|
| 51 |
+
"eval_accuracy": 0.9059738955823293,
|
| 52 |
+
"eval_auc_roc": 0.8639963332712217,
|
| 53 |
+
"eval_f1": 0.8680518616812346,
|
| 54 |
+
"eval_loss": 0.5702313184738159,
|
| 55 |
+
"eval_min_tDCF": 0.032200374388851885,
|
| 56 |
+
"eval_precision": 0.9081760352558711,
|
| 57 |
+
"eval_recall": 0.9059738955823293,
|
| 58 |
+
"eval_runtime": 565.0431,
|
| 59 |
+
"eval_samples_per_second": 35.254,
|
| 60 |
+
"eval_steps_per_second": 0.276,
|
| 61 |
+
"step": 10
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.061855670103092786,
|
| 65 |
+
"grad_norm": 61530.01953125,
|
| 66 |
+
"learning_rate": 6.198347107438017e-05,
|
| 67 |
+
"loss": 0.5335,
|
| 68 |
+
"step": 15
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.061855670103092786,
|
| 72 |
+
"eval_EER": 0.3502740408570005,
|
| 73 |
+
"eval_FN": 1959,
|
| 74 |
+
"eval_FP": 1,
|
| 75 |
+
"eval_TN": 17912,
|
| 76 |
+
"eval_TP": 48,
|
| 77 |
+
"eval_accuracy": 0.9016064257028112,
|
| 78 |
+
"eval_auc_roc": 0.692431566834229,
|
| 79 |
+
"eval_f1": 0.8573040422332877,
|
| 80 |
+
"eval_loss": 0.5015577673912048,
|
| 81 |
+
"eval_min_tDCF": 0.0362762570160359,
|
| 82 |
+
"eval_precision": 0.9092907614081319,
|
| 83 |
+
"eval_recall": 0.9016064257028112,
|
| 84 |
+
"eval_runtime": 574.5762,
|
| 85 |
+
"eval_samples_per_second": 34.669,
|
| 86 |
+
"eval_steps_per_second": 0.272,
|
| 87 |
+
"step": 15
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.08247422680412371,
|
| 91 |
+
"grad_norm": 60310.94921875,
|
| 92 |
+
"learning_rate": 8.264462809917356e-05,
|
| 93 |
+
"loss": 0.4592,
|
| 94 |
+
"step": 20
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.08247422680412371,
|
| 98 |
+
"eval_EER": 0.17736274755577838,
|
| 99 |
+
"eval_FN": 1786,
|
| 100 |
+
"eval_FP": 18,
|
| 101 |
+
"eval_TN": 17895,
|
| 102 |
+
"eval_TP": 221,
|
| 103 |
+
"eval_accuracy": 0.909437751004016,
|
| 104 |
+
"eval_auc_roc": 0.868339391930621,
|
| 105 |
+
"eval_f1": 0.8759229981274053,
|
| 106 |
+
"eval_loss": 0.43352434039115906,
|
| 107 |
+
"eval_min_tDCF": 0.03230474030893547,
|
| 108 |
+
"eval_precision": 0.9108075617102515,
|
| 109 |
+
"eval_recall": 0.909437751004016,
|
| 110 |
+
"eval_runtime": 560.5982,
|
| 111 |
+
"eval_samples_per_second": 35.533,
|
| 112 |
+
"eval_steps_per_second": 0.278,
|
| 113 |
+
"step": 20
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.10309278350515463,
|
| 117 |
+
"grad_norm": 60051.5625,
|
| 118 |
+
"learning_rate": 0.00010330578512396695,
|
| 119 |
+
"loss": 0.3927,
|
| 120 |
+
"step": 25
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.10309278350515463,
|
| 124 |
+
"eval_EER": 0.22969606377675944,
|
| 125 |
+
"eval_FN": 1763,
|
| 126 |
+
"eval_FP": 22,
|
| 127 |
+
"eval_TN": 17891,
|
| 128 |
+
"eval_TP": 244,
|
| 129 |
+
"eval_accuracy": 0.9103915662650602,
|
| 130 |
+
"eval_auc_roc": 0.810980623809521,
|
| 131 |
+
"eval_f1": 0.878150262706421,
|
| 132 |
+
"eval_loss": 0.3780788481235504,
|
| 133 |
+
"eval_min_tDCF": 0.031502623917945206,
|
| 134 |
+
"eval_precision": 0.9110029330061574,
|
| 135 |
+
"eval_recall": 0.9103915662650602,
|
| 136 |
+
"eval_runtime": 558.5071,
|
| 137 |
+
"eval_samples_per_second": 35.667,
|
| 138 |
+
"eval_steps_per_second": 0.279,
|
| 139 |
+
"step": 25
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.12371134020618557,
|
| 143 |
+
"grad_norm": 57445.89453125,
|
| 144 |
+
"learning_rate": 0.00012396694214876033,
|
| 145 |
+
"loss": 0.3231,
|
| 146 |
+
"step": 30
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.12371134020618557,
|
| 150 |
+
"eval_EER": 0.28998505231689087,
|
| 151 |
+
"eval_FN": 1693,
|
| 152 |
+
"eval_FP": 24,
|
| 153 |
+
"eval_TN": 17889,
|
| 154 |
+
"eval_TP": 314,
|
| 155 |
+
"eval_accuracy": 0.9138052208835341,
|
| 156 |
+
"eval_auc_roc": 0.7424168511310174,
|
| 157 |
+
"eval_f1": 0.8850500186402774,
|
| 158 |
+
"eval_loss": 0.320069283246994,
|
| 159 |
+
"eval_min_tDCF": 0.03092566571346294,
|
| 160 |
+
"eval_precision": 0.9150997878637065,
|
| 161 |
+
"eval_recall": 0.9138052208835341,
|
| 162 |
+
"eval_runtime": 556.1583,
|
| 163 |
+
"eval_samples_per_second": 35.817,
|
| 164 |
+
"eval_steps_per_second": 0.28,
|
| 165 |
+
"step": 30
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.14432989690721648,
|
| 169 |
+
"grad_norm": 53125.640625,
|
| 170 |
+
"learning_rate": 0.0001446280991735537,
|
| 171 |
+
"loss": 0.2519,
|
| 172 |
+
"step": 35
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.14432989690721648,
|
| 176 |
+
"eval_EER": 0.11410064773293435,
|
| 177 |
+
"eval_FN": 1576,
|
| 178 |
+
"eval_FP": 26,
|
| 179 |
+
"eval_TN": 17887,
|
| 180 |
+
"eval_TP": 431,
|
| 181 |
+
"eval_accuracy": 0.9195783132530121,
|
| 182 |
+
"eval_auc_roc": 0.9339815808517673,
|
| 183 |
+
"eval_f1": 0.8959509050520598,
|
| 184 |
+
"eval_loss": 0.2804459035396576,
|
| 185 |
+
"eval_min_tDCF": 0.029493181501655946,
|
| 186 |
+
"eval_precision": 0.9214521151858079,
|
| 187 |
+
"eval_recall": 0.9195783132530121,
|
| 188 |
+
"eval_runtime": 555.9211,
|
| 189 |
+
"eval_samples_per_second": 35.832,
|
| 190 |
+
"eval_steps_per_second": 0.281,
|
| 191 |
+
"step": 35
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.16494845360824742,
|
| 195 |
+
"grad_norm": 40388.56640625,
|
| 196 |
+
"learning_rate": 0.00016528925619834712,
|
| 197 |
+
"loss": 0.1963,
|
| 198 |
+
"step": 40
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.16494845360824742,
|
| 202 |
+
"eval_EER": 0.08975903614457832,
|
| 203 |
+
"eval_FN": 1256,
|
| 204 |
+
"eval_FP": 47,
|
| 205 |
+
"eval_TN": 17866,
|
| 206 |
+
"eval_TP": 751,
|
| 207 |
+
"eval_accuracy": 0.9345883534136546,
|
| 208 |
+
"eval_auc_roc": 0.9623328065386955,
|
| 209 |
+
"eval_f1": 0.921559293793824,
|
| 210 |
+
"eval_loss": 0.23952148854732513,
|
| 211 |
+
"eval_min_tDCF": 0.028605190825578903,
|
| 212 |
+
"eval_precision": 0.9350002305083934,
|
| 213 |
+
"eval_recall": 0.9345883534136546,
|
| 214 |
+
"eval_runtime": 564.5981,
|
| 215 |
+
"eval_samples_per_second": 35.282,
|
| 216 |
+
"eval_steps_per_second": 0.276,
|
| 217 |
+
"step": 40
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.18556701030927836,
|
| 221 |
+
"grad_norm": 60576.25390625,
|
| 222 |
+
"learning_rate": 0.00018595041322314049,
|
| 223 |
+
"loss": 0.1423,
|
| 224 |
+
"step": 45
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.18556701030927836,
|
| 228 |
+
"eval_EER": 0.09007160122223745,
|
| 229 |
+
"eval_FN": 1880,
|
| 230 |
+
"eval_FP": 16,
|
| 231 |
+
"eval_TN": 17897,
|
| 232 |
+
"eval_TP": 127,
|
| 233 |
+
"eval_accuracy": 0.9048192771084337,
|
| 234 |
+
"eval_auc_roc": 0.9659451424285642,
|
| 235 |
+
"eval_f1": 0.8659131736457947,
|
| 236 |
+
"eval_loss": 0.3794119358062744,
|
| 237 |
+
"eval_min_tDCF": 0.02977514972925526,
|
| 238 |
+
"eval_precision": 0.9032445907005072,
|
| 239 |
+
"eval_recall": 0.9048192771084337,
|
| 240 |
+
"eval_runtime": 565.8875,
|
| 241 |
+
"eval_samples_per_second": 35.201,
|
| 242 |
+
"eval_steps_per_second": 0.276,
|
| 243 |
+
"step": 45
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.20618556701030927,
|
| 247 |
+
"grad_norm": 23454.15625,
|
| 248 |
+
"learning_rate": 0.0002066115702479339,
|
| 249 |
+
"loss": 0.1046,
|
| 250 |
+
"step": 50
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.20618556701030927,
|
| 254 |
+
"eval_EER": 0.07514095907887752,
|
| 255 |
+
"eval_FN": 1371,
|
| 256 |
+
"eval_FP": 50,
|
| 257 |
+
"eval_TN": 17863,
|
| 258 |
+
"eval_TP": 636,
|
| 259 |
+
"eval_accuracy": 0.9286646586345382,
|
| 260 |
+
"eval_auc_roc": 0.9767056579257253,
|
| 261 |
+
"eval_f1": 0.9124369531932773,
|
| 262 |
+
"eval_loss": 0.3193705081939697,
|
| 263 |
+
"eval_min_tDCF": 0.0317551523945207,
|
| 264 |
+
"eval_precision": 0.9285581382596789,
|
| 265 |
+
"eval_recall": 0.9286646586345382,
|
| 266 |
+
"eval_runtime": 567.956,
|
| 267 |
+
"eval_samples_per_second": 35.073,
|
| 268 |
+
"eval_steps_per_second": 0.275,
|
| 269 |
+
"step": 50
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2268041237113402,
|
| 273 |
+
"grad_norm": 24468.62109375,
|
| 274 |
+
"learning_rate": 0.00022727272727272727,
|
| 275 |
+
"loss": 0.0681,
|
| 276 |
+
"step": 55
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.2268041237113402,
|
| 280 |
+
"eval_EER": 0.17090184354758345,
|
| 281 |
+
"eval_FN": 1947,
|
| 282 |
+
"eval_FP": 4,
|
| 283 |
+
"eval_TN": 17909,
|
| 284 |
+
"eval_TP": 60,
|
| 285 |
+
"eval_accuracy": 0.9020582329317269,
|
| 286 |
+
"eval_auc_roc": 0.9014651616678754,
|
| 287 |
+
"eval_f1": 0.8586333116075927,
|
| 288 |
+
"eval_loss": 0.485874205827713,
|
| 289 |
+
"eval_min_tDCF": 0.037820095750954394,
|
| 290 |
+
"eval_precision": 0.9055263711979262,
|
| 291 |
+
"eval_recall": 0.9020582329317269,
|
| 292 |
+
"eval_runtime": 540.7107,
|
| 293 |
+
"eval_samples_per_second": 36.84,
|
| 294 |
+
"eval_steps_per_second": 0.289,
|
| 295 |
+
"step": 55
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.24742268041237114,
|
| 299 |
+
"grad_norm": 15163.4248046875,
|
| 300 |
+
"learning_rate": 0.00024793388429752067,
|
| 301 |
+
"loss": 0.0473,
|
| 302 |
+
"step": 60
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.24742268041237114,
|
| 306 |
+
"eval_EER": 0.7055306427503634,
|
| 307 |
+
"eval_FN": 1770,
|
| 308 |
+
"eval_FP": 23,
|
| 309 |
+
"eval_TN": 17890,
|
| 310 |
+
"eval_TP": 237,
|
| 311 |
+
"eval_accuracy": 0.9099899598393574,
|
| 312 |
+
"eval_auc_roc": 0.3149452131073315,
|
| 313 |
+
"eval_f1": 0.8774006630754101,
|
| 314 |
+
"eval_loss": 0.5605435967445374,
|
| 315 |
+
"eval_min_tDCF": 0.03815886984734471,
|
| 316 |
+
"eval_precision": 0.9101275606770144,
|
| 317 |
+
"eval_recall": 0.9099899598393574,
|
| 318 |
+
"eval_runtime": 525.717,
|
| 319 |
+
"eval_samples_per_second": 37.891,
|
| 320 |
+
"eval_steps_per_second": 0.297,
|
| 321 |
+
"step": 60
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.26804123711340205,
|
| 325 |
+
"grad_norm": 9380.0947265625,
|
| 326 |
+
"learning_rate": 0.00026859504132231406,
|
| 327 |
+
"loss": 0.0323,
|
| 328 |
+
"step": 65
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.26804123711340205,
|
| 332 |
+
"eval_EER": 0.07025411061248463,
|
| 333 |
+
"eval_FN": 1640,
|
| 334 |
+
"eval_FP": 26,
|
| 335 |
+
"eval_TN": 17887,
|
| 336 |
+
"eval_TP": 367,
|
| 337 |
+
"eval_accuracy": 0.9163654618473895,
|
| 338 |
+
"eval_auc_roc": 0.9790777080085719,
|
| 339 |
+
"eval_f1": 0.8900460479739473,
|
| 340 |
+
"eval_loss": 0.5106881856918335,
|
| 341 |
+
"eval_min_tDCF": 0.03366154733762597,
|
| 342 |
+
"eval_precision": 0.9178100011326927,
|
| 343 |
+
"eval_recall": 0.9163654618473895,
|
| 344 |
+
"eval_runtime": 489.5871,
|
| 345 |
+
"eval_samples_per_second": 40.687,
|
| 346 |
+
"eval_steps_per_second": 0.319,
|
| 347 |
+
"step": 65
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.28865979381443296,
|
| 351 |
+
"grad_norm": 9063.4306640625,
|
| 352 |
+
"learning_rate": 0.0002892561983471074,
|
| 353 |
+
"loss": 0.0339,
|
| 354 |
+
"step": 70
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.28865979381443296,
|
| 358 |
+
"eval_EER": 0.8315894369709501,
|
| 359 |
+
"eval_FN": 1930,
|
| 360 |
+
"eval_FP": 10,
|
| 361 |
+
"eval_TN": 17903,
|
| 362 |
+
"eval_TP": 77,
|
| 363 |
+
"eval_accuracy": 0.9026104417670683,
|
| 364 |
+
"eval_auc_roc": 0.17729458645981183,
|
| 365 |
+
"eval_f1": 0.8604388577635353,
|
| 366 |
+
"eval_loss": 0.8920541405677795,
|
| 367 |
+
"eval_min_tDCF": 0.04347511060142291,
|
| 368 |
+
"eval_precision": 0.900911167660015,
|
| 369 |
+
"eval_recall": 0.9026104417670683,
|
| 370 |
+
"eval_runtime": 488.1086,
|
| 371 |
+
"eval_samples_per_second": 40.811,
|
| 372 |
+
"eval_steps_per_second": 0.32,
|
| 373 |
+
"step": 70
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.30927835051546393,
|
| 377 |
+
"grad_norm": 10245.0908203125,
|
| 378 |
+
"learning_rate": 0.00030991735537190085,
|
| 379 |
+
"loss": 0.0423,
|
| 380 |
+
"step": 75
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.30927835051546393,
|
| 384 |
+
"eval_EER": 0.0732436472346947,
|
| 385 |
+
"eval_FN": 1920,
|
| 386 |
+
"eval_FP": 13,
|
| 387 |
+
"eval_TN": 17900,
|
| 388 |
+
"eval_TP": 87,
|
| 389 |
+
"eval_accuracy": 0.9029618473895582,
|
| 390 |
+
"eval_auc_roc": 0.9753320810313014,
|
| 391 |
+
"eval_f1": 0.8615004047011067,
|
| 392 |
+
"eval_loss": 0.8963676691055298,
|
| 393 |
+
"eval_min_tDCF": 0.03272278811131397,
|
| 394 |
+
"eval_precision": 0.8997903921437517,
|
| 395 |
+
"eval_recall": 0.9029618473895582,
|
| 396 |
+
"eval_runtime": 487.6584,
|
| 397 |
+
"eval_samples_per_second": 40.848,
|
| 398 |
+
"eval_steps_per_second": 0.32,
|
| 399 |
+
"step": 75
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.32989690721649484,
|
| 403 |
+
"grad_norm": 81123.5,
|
| 404 |
+
"learning_rate": 0.00033057851239669424,
|
| 405 |
+
"loss": 0.0456,
|
| 406 |
+
"step": 80
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.32989690721649484,
|
| 410 |
+
"eval_EER": 0.8520179372197283,
|
| 411 |
+
"eval_FN": 1956,
|
| 412 |
+
"eval_FP": 11,
|
| 413 |
+
"eval_TN": 17902,
|
| 414 |
+
"eval_TP": 51,
|
| 415 |
+
"eval_accuracy": 0.9012550200803213,
|
| 416 |
+
"eval_auc_roc": 0.11257390569394102,
|
| 417 |
+
"eval_f1": 0.8573839520076927,
|
| 418 |
+
"eval_loss": 1.0843422412872314,
|
| 419 |
+
"eval_min_tDCF": 0.04780276651882538,
|
| 420 |
+
"eval_precision": 0.8935492259813427,
|
| 421 |
+
"eval_recall": 0.9012550200803213,
|
| 422 |
+
"eval_runtime": 488.2,
|
| 423 |
+
"eval_samples_per_second": 40.803,
|
| 424 |
+
"eval_steps_per_second": 0.32,
|
| 425 |
+
"step": 80
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.35051546391752575,
|
| 429 |
+
"grad_norm": 34188.77734375,
|
| 430 |
+
"learning_rate": 0.00035123966942148763,
|
| 431 |
+
"loss": 0.0712,
|
| 432 |
+
"step": 85
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.35051546391752575,
|
| 436 |
+
"eval_EER": 0.8664673642255027,
|
| 437 |
+
"eval_FN": 1936,
|
| 438 |
+
"eval_FP": 10,
|
| 439 |
+
"eval_TN": 17903,
|
| 440 |
+
"eval_TP": 71,
|
| 441 |
+
"eval_accuracy": 0.9023092369477912,
|
| 442 |
+
"eval_auc_roc": 0.09900188562940444,
|
| 443 |
+
"eval_f1": 0.8597455361705937,
|
| 444 |
+
"eval_loss": 0.8587021827697754,
|
| 445 |
+
"eval_min_tDCF": 0.04799565196239556,
|
| 446 |
+
"eval_precision": 0.899807832232875,
|
| 447 |
+
"eval_recall": 0.9023092369477912,
|
| 448 |
+
"eval_runtime": 491.4372,
|
| 449 |
+
"eval_samples_per_second": 40.534,
|
| 450 |
+
"eval_steps_per_second": 0.317,
|
| 451 |
+
"step": 85
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.3711340206185567,
|
| 455 |
+
"grad_norm": 60596.9375,
|
| 456 |
+
"learning_rate": 0.00037190082644628097,
|
| 457 |
+
"loss": 0.0629,
|
| 458 |
+
"step": 90
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.3711340206185567,
|
| 462 |
+
"eval_EER": 0.08479874951213202,
|
| 463 |
+
"eval_FN": 1424,
|
| 464 |
+
"eval_FP": 36,
|
| 465 |
+
"eval_TN": 17877,
|
| 466 |
+
"eval_TP": 583,
|
| 467 |
+
"eval_accuracy": 0.9267068273092369,
|
| 468 |
+
"eval_auc_roc": 0.9682722568370165,
|
| 469 |
+
"eval_f1": 0.9087037242589197,
|
| 470 |
+
"eval_loss": 0.48098817467689514,
|
| 471 |
+
"eval_min_tDCF": 0.03275128492246656,
|
| 472 |
+
"eval_precision": 0.9277952256157582,
|
| 473 |
+
"eval_recall": 0.9267068273092369,
|
| 474 |
+
"eval_runtime": 488.0171,
|
| 475 |
+
"eval_samples_per_second": 40.818,
|
| 476 |
+
"eval_steps_per_second": 0.32,
|
| 477 |
+
"step": 90
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.3917525773195876,
|
| 481 |
+
"grad_norm": 3116.9609375,
|
| 482 |
+
"learning_rate": 0.0003925619834710744,
|
| 483 |
+
"loss": 0.0477,
|
| 484 |
+
"step": 95
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.3917525773195876,
|
| 488 |
+
"eval_EER": 0.1219226260249616,
|
| 489 |
+
"eval_FN": 1789,
|
| 490 |
+
"eval_FP": 16,
|
| 491 |
+
"eval_TN": 17897,
|
| 492 |
+
"eval_TP": 218,
|
| 493 |
+
"eval_accuracy": 0.9093875502008032,
|
| 494 |
+
"eval_auc_roc": 0.8890115127951516,
|
| 495 |
+
"eval_f1": 0.8756793043607181,
|
| 496 |
+
"eval_loss": 0.9414774179458618,
|
| 497 |
+
"eval_min_tDCF": 0.04080623194802115,
|
| 498 |
+
"eval_precision": 0.9113902477657331,
|
| 499 |
+
"eval_recall": 0.9093875502008032,
|
| 500 |
+
"eval_runtime": 482.559,
|
| 501 |
+
"eval_samples_per_second": 41.28,
|
| 502 |
+
"eval_steps_per_second": 0.323,
|
| 503 |
+
"step": 95
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.41237113402061853,
|
| 507 |
+
"grad_norm": 41389.37109375,
|
| 508 |
+
"learning_rate": 0.0004132231404958678,
|
| 509 |
+
"loss": 0.0484,
|
| 510 |
+
"step": 100
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.41237113402061853,
|
| 514 |
+
"eval_EER": 0.6905829596412563,
|
| 515 |
+
"eval_FN": 1671,
|
| 516 |
+
"eval_FP": 22,
|
| 517 |
+
"eval_TN": 17891,
|
| 518 |
+
"eval_TP": 336,
|
| 519 |
+
"eval_accuracy": 0.9150100401606426,
|
| 520 |
+
"eval_auc_roc": 0.31285679599990995,
|
| 521 |
+
"eval_f1": 0.8872502405426094,
|
| 522 |
+
"eval_loss": 0.7773601412773132,
|
| 523 |
+
"eval_min_tDCF": 0.03829154899736703,
|
| 524 |
+
"eval_precision": 0.916994154680854,
|
| 525 |
+
"eval_recall": 0.9150100401606426,
|
| 526 |
+
"eval_runtime": 493.0782,
|
| 527 |
+
"eval_samples_per_second": 40.399,
|
| 528 |
+
"eval_steps_per_second": 0.316,
|
| 529 |
+
"step": 100
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.4329896907216495,
|
| 533 |
+
"grad_norm": 9161.97265625,
|
| 534 |
+
"learning_rate": 0.00043388429752066115,
|
| 535 |
+
"loss": 0.0449,
|
| 536 |
+
"step": 105
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.4329896907216495,
|
| 540 |
+
"eval_EER": 0.6527154957648229,
|
| 541 |
+
"eval_FN": 1563,
|
| 542 |
+
"eval_FP": 37,
|
| 543 |
+
"eval_TN": 17876,
|
| 544 |
+
"eval_TP": 444,
|
| 545 |
+
"eval_accuracy": 0.9196787148594378,
|
| 546 |
+
"eval_auc_roc": 0.3629036078186794,
|
| 547 |
+
"eval_f1": 0.8966871695287829,
|
| 548 |
+
"eval_loss": 0.39491569995880127,
|
| 549 |
+
"eval_min_tDCF": 0.036267007026237176,
|
| 550 |
+
"eval_precision": 0.9199454809348472,
|
| 551 |
+
"eval_recall": 0.9196787148594378,
|
| 552 |
+
"eval_runtime": 486.0014,
|
| 553 |
+
"eval_samples_per_second": 40.988,
|
| 554 |
+
"eval_steps_per_second": 0.321,
|
| 555 |
+
"step": 105
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.4536082474226804,
|
| 559 |
+
"grad_norm": 139889.5,
|
| 560 |
+
"learning_rate": 0.00045454545454545455,
|
| 561 |
+
"loss": 0.0567,
|
| 562 |
+
"step": 110
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.4536082474226804,
|
| 566 |
+
"eval_EER": 0.21923268560039863,
|
| 567 |
+
"eval_FN": 1467,
|
| 568 |
+
"eval_FP": 63,
|
| 569 |
+
"eval_TN": 17850,
|
| 570 |
+
"eval_TP": 540,
|
| 571 |
+
"eval_accuracy": 0.9231927710843374,
|
| 572 |
+
"eval_auc_roc": 0.8157612037876364,
|
| 573 |
+
"eval_f1": 0.9039825338197379,
|
| 574 |
+
"eval_loss": 0.5852819085121155,
|
| 575 |
+
"eval_min_tDCF": 0.035529632219237356,
|
| 576 |
+
"eval_precision": 0.9211816297183598,
|
| 577 |
+
"eval_recall": 0.9231927710843374,
|
| 578 |
+
"eval_runtime": 486.1743,
|
| 579 |
+
"eval_samples_per_second": 40.973,
|
| 580 |
+
"eval_steps_per_second": 0.321,
|
| 581 |
+
"step": 110
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.4742268041237113,
|
| 585 |
+
"grad_norm": 34995.62890625,
|
| 586 |
+
"learning_rate": 0.000475206611570248,
|
| 587 |
+
"loss": 0.0416,
|
| 588 |
+
"step": 115
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 0.4742268041237113,
|
| 592 |
+
"eval_EER": 0.763328350772297,
|
| 593 |
+
"eval_FN": 1912,
|
| 594 |
+
"eval_FP": 8,
|
| 595 |
+
"eval_TN": 17905,
|
| 596 |
+
"eval_TP": 95,
|
| 597 |
+
"eval_accuracy": 0.9036144578313253,
|
| 598 |
+
"eval_auc_roc": 0.25487012449671276,
|
| 599 |
+
"eval_f1": 0.8625587543925759,
|
| 600 |
+
"eval_loss": 0.7031072974205017,
|
| 601 |
+
"eval_min_tDCF": 0.04081736781756234,
|
| 602 |
+
"eval_precision": 0.9054126400558561,
|
| 603 |
+
"eval_recall": 0.9036144578313253,
|
| 604 |
+
"eval_runtime": 485.8129,
|
| 605 |
+
"eval_samples_per_second": 41.003,
|
| 606 |
+
"eval_steps_per_second": 0.321,
|
| 607 |
+
"step": 115
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.4948453608247423,
|
| 611 |
+
"grad_norm": 13682.4638671875,
|
| 612 |
+
"learning_rate": 0.0004958677685950413,
|
| 613 |
+
"loss": 0.1778,
|
| 614 |
+
"step": 120
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.4948453608247423,
|
| 618 |
+
"eval_EER": 0.7389138016939923,
|
| 619 |
+
"eval_FN": 1924,
|
| 620 |
+
"eval_FP": 3,
|
| 621 |
+
"eval_TN": 17910,
|
| 622 |
+
"eval_TP": 83,
|
| 623 |
+
"eval_accuracy": 0.9032630522088353,
|
| 624 |
+
"eval_auc_roc": 0.2838427308695789,
|
| 625 |
+
"eval_f1": 0.8613309785081884,
|
| 626 |
+
"eval_loss": 0.5440215468406677,
|
| 627 |
+
"eval_min_tDCF": 0.039758951190511656,
|
| 628 |
+
"eval_precision": 0.9092537776701568,
|
| 629 |
+
"eval_recall": 0.9032630522088353,
|
| 630 |
+
"eval_runtime": 486.772,
|
| 631 |
+
"eval_samples_per_second": 40.923,
|
| 632 |
+
"eval_steps_per_second": 0.32,
|
| 633 |
+
"step": 120
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.5154639175257731,
|
| 637 |
+
"grad_norm": 4219.48876953125,
|
| 638 |
+
"learning_rate": 0.0004981634527089073,
|
| 639 |
+
"loss": 0.036,
|
| 640 |
+
"step": 125
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.5154639175257731,
|
| 644 |
+
"eval_EER": 0.1538247720702691,
|
| 645 |
+
"eval_FN": 1651,
|
| 646 |
+
"eval_FP": 20,
|
| 647 |
+
"eval_TN": 17893,
|
| 648 |
+
"eval_TP": 356,
|
| 649 |
+
"eval_accuracy": 0.9161144578313253,
|
| 650 |
+
"eval_auc_roc": 0.9077825250210763,
|
| 651 |
+
"eval_f1": 0.8892338338691552,
|
| 652 |
+
"eval_loss": 0.582485020160675,
|
| 653 |
+
"eval_min_tDCF": 0.03658143992258881,
|
| 654 |
+
"eval_precision": 0.9186759600892138,
|
| 655 |
+
"eval_recall": 0.9161144578313253,
|
| 656 |
+
"eval_runtime": 490.0237,
|
| 657 |
+
"eval_samples_per_second": 40.651,
|
| 658 |
+
"eval_steps_per_second": 0.318,
|
| 659 |
+
"step": 125
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.5360824742268041,
|
| 663 |
+
"grad_norm": 164861.5625,
|
| 664 |
+
"learning_rate": 0.0004958677685950413,
|
| 665 |
+
"loss": 0.0797,
|
| 666 |
+
"step": 130
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 0.5360824742268041,
|
| 670 |
+
"eval_EER": 0.22364541543815544,
|
| 671 |
+
"eval_FN": 1934,
|
| 672 |
+
"eval_FP": 5,
|
| 673 |
+
"eval_TN": 17908,
|
| 674 |
+
"eval_TP": 73,
|
| 675 |
+
"eval_accuracy": 0.9026606425702811,
|
| 676 |
+
"eval_auc_roc": 0.7720901258034772,
|
| 677 |
+
"eval_f1": 0.8601190952116461,
|
| 678 |
+
"eval_loss": 0.5863870978355408,
|
| 679 |
+
"eval_min_tDCF": 0.04074244164850256,
|
| 680 |
+
"eval_precision": 0.9058918578876928,
|
| 681 |
+
"eval_recall": 0.9026606425702811,
|
| 682 |
+
"eval_runtime": 485.4127,
|
| 683 |
+
"eval_samples_per_second": 41.037,
|
| 684 |
+
"eval_steps_per_second": 0.321,
|
| 685 |
+
"step": 130
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 0.5567010309278351,
|
| 689 |
+
"grad_norm": 8241.765625,
|
| 690 |
+
"learning_rate": 0.0004935720844811754,
|
| 691 |
+
"loss": 0.0669,
|
| 692 |
+
"step": 135
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.5567010309278351,
|
| 696 |
+
"eval_EER": 0.15970094107930716,
|
| 697 |
+
"eval_FN": 1915,
|
| 698 |
+
"eval_FP": 5,
|
| 699 |
+
"eval_TN": 17908,
|
| 700 |
+
"eval_TP": 92,
|
| 701 |
+
"eval_accuracy": 0.9036144578313253,
|
| 702 |
+
"eval_auc_roc": 0.879100700164842,
|
| 703 |
+
"eval_f1": 0.862304584131794,
|
| 704 |
+
"eval_loss": 0.4263610243797302,
|
| 705 |
+
"eval_min_tDCF": 0.03697872218629872,
|
| 706 |
+
"eval_precision": 0.9079348320075219,
|
| 707 |
+
"eval_recall": 0.9036144578313253,
|
| 708 |
+
"eval_runtime": 487.6391,
|
| 709 |
+
"eval_samples_per_second": 40.85,
|
| 710 |
+
"eval_steps_per_second": 0.32,
|
| 711 |
+
"step": 135
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.5773195876288659,
|
| 715 |
+
"grad_norm": 10060.2216796875,
|
| 716 |
+
"learning_rate": 0.0004912764003673095,
|
| 717 |
+
"loss": 0.0353,
|
| 718 |
+
"step": 140
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.5773195876288659,
|
| 722 |
+
"eval_EER": 0.16392625809663627,
|
| 723 |
+
"eval_FN": 1677,
|
| 724 |
+
"eval_FP": 24,
|
| 725 |
+
"eval_TN": 17889,
|
| 726 |
+
"eval_TP": 330,
|
| 727 |
+
"eval_accuracy": 0.9146084337349397,
|
| 728 |
+
"eval_auc_roc": 0.9188998000105197,
|
| 729 |
+
"eval_f1": 0.8865990450587786,
|
| 730 |
+
"eval_loss": 0.6355106830596924,
|
| 731 |
+
"eval_min_tDCF": 0.03572232573699304,
|
| 732 |
+
"eval_precision": 0.9160949134664114,
|
| 733 |
+
"eval_recall": 0.9146084337349397,
|
| 734 |
+
"eval_runtime": 487.3528,
|
| 735 |
+
"eval_samples_per_second": 40.874,
|
| 736 |
+
"eval_steps_per_second": 0.32,
|
| 737 |
+
"step": 140
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5773195876288659,
|
| 741 |
+
"step": 140,
|
| 742 |
+
"total_flos": 1.9445707177361142e+18,
|
| 743 |
+
"train_loss": 0.16865910078798022,
|
| 744 |
+
"train_runtime": 17827.0865,
|
| 745 |
+
"train_samples_per_second": 34.802,
|
| 746 |
+
"train_steps_per_second": 0.068
|
| 747 |
+
}
|
| 748 |
+
],
|
| 749 |
+
"logging_steps": 5,
|
| 750 |
+
"max_steps": 1210,
|
| 751 |
+
"num_input_tokens_seen": 0,
|
| 752 |
+
"num_train_epochs": 5,
|
| 753 |
+
"save_steps": 10,
|
| 754 |
+
"stateful_callbacks": {
|
| 755 |
+
"EarlyStoppingCallback": {
|
| 756 |
+
"args": {
|
| 757 |
+
"early_stopping_patience": 20,
|
| 758 |
+
"early_stopping_threshold": 0.0
|
| 759 |
+
},
|
| 760 |
+
"attributes": {
|
| 761 |
+
"early_stopping_patience_counter": 0
|
| 762 |
+
}
|
| 763 |
+
},
|
| 764 |
+
"TrainerControl": {
|
| 765 |
+
"args": {
|
| 766 |
+
"should_epoch_stop": false,
|
| 767 |
+
"should_evaluate": false,
|
| 768 |
+
"should_log": false,
|
| 769 |
+
"should_save": true,
|
| 770 |
+
"should_training_stop": true
|
| 771 |
+
},
|
| 772 |
+
"attributes": {}
|
| 773 |
+
}
|
| 774 |
+
},
|
| 775 |
+
"total_flos": 1.9445707177361142e+18,
|
| 776 |
+
"train_batch_size": 128,
|
| 777 |
+
"trial_name": null,
|
| 778 |
+
"trial_params": null
|
| 779 |
+
}
|