Upload trainer_state.json
Browse files- trainer_state.json +992 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,992 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 198.91205211726384,
|
| 5 |
+
"global_step": 91500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 3.26,
|
| 12 |
+
"learning_rate": 5.988000000000001e-05,
|
| 13 |
+
"loss": 6.0553,
|
| 14 |
+
"step": 1500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 3.26,
|
| 18 |
+
"eval_cer": 0.7651548792050215,
|
| 19 |
+
"eval_loss": 3.0141708850860596,
|
| 20 |
+
"eval_runtime": 204.505,
|
| 21 |
+
"eval_samples_per_second": 21.858,
|
| 22 |
+
"eval_steps_per_second": 1.369,
|
| 23 |
+
"eval_wer": 0.9928897917724734,
|
| 24 |
+
"step": 1500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 6.52,
|
| 28 |
+
"learning_rate": 7.911466666666667e-05,
|
| 29 |
+
"loss": 2.3839,
|
| 30 |
+
"step": 3000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 6.52,
|
| 34 |
+
"eval_cer": 0.19668038836815765,
|
| 35 |
+
"eval_loss": 0.5387711524963379,
|
| 36 |
+
"eval_runtime": 202.7107,
|
| 37 |
+
"eval_samples_per_second": 22.051,
|
| 38 |
+
"eval_steps_per_second": 1.381,
|
| 39 |
+
"eval_wer": 0.45625045345715737,
|
| 40 |
+
"step": 3000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 9.78,
|
| 44 |
+
"learning_rate": 7.778133333333334e-05,
|
| 45 |
+
"loss": 1.3597,
|
| 46 |
+
"step": 4500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 9.78,
|
| 50 |
+
"eval_cer": 0.15966779880705206,
|
| 51 |
+
"eval_loss": 0.4122415781021118,
|
| 52 |
+
"eval_runtime": 202.9557,
|
| 53 |
+
"eval_samples_per_second": 22.025,
|
| 54 |
+
"eval_steps_per_second": 1.38,
|
| 55 |
+
"eval_wer": 0.36247551331350214,
|
| 56 |
+
"step": 4500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 13.04,
|
| 60 |
+
"learning_rate": 7.644800000000001e-05,
|
| 61 |
+
"loss": 1.2294,
|
| 62 |
+
"step": 6000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 13.04,
|
| 66 |
+
"eval_cer": 0.15757353911885075,
|
| 67 |
+
"eval_loss": 0.36426475644111633,
|
| 68 |
+
"eval_runtime": 204.7716,
|
| 69 |
+
"eval_samples_per_second": 21.829,
|
| 70 |
+
"eval_steps_per_second": 1.367,
|
| 71 |
+
"eval_wer": 0.3416164840745846,
|
| 72 |
+
"step": 6000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 16.3,
|
| 76 |
+
"learning_rate": 7.511644444444445e-05,
|
| 77 |
+
"loss": 1.1672,
|
| 78 |
+
"step": 7500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 16.3,
|
| 82 |
+
"eval_cer": 0.1442218835135557,
|
| 83 |
+
"eval_loss": 0.34255123138427734,
|
| 84 |
+
"eval_runtime": 202.492,
|
| 85 |
+
"eval_samples_per_second": 22.075,
|
| 86 |
+
"eval_steps_per_second": 1.383,
|
| 87 |
+
"eval_wer": 0.3202314445331205,
|
| 88 |
+
"step": 7500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 19.56,
|
| 92 |
+
"learning_rate": 7.378311111111112e-05,
|
| 93 |
+
"loss": 1.1288,
|
| 94 |
+
"step": 9000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 19.56,
|
| 98 |
+
"eval_cer": 0.1378970992403058,
|
| 99 |
+
"eval_loss": 0.31974485516548157,
|
| 100 |
+
"eval_runtime": 204.9353,
|
| 101 |
+
"eval_samples_per_second": 21.812,
|
| 102 |
+
"eval_steps_per_second": 1.366,
|
| 103 |
+
"eval_wer": 0.3065551766669085,
|
| 104 |
+
"step": 9000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 22.83,
|
| 108 |
+
"learning_rate": 7.245155555555557e-05,
|
| 109 |
+
"loss": 1.0864,
|
| 110 |
+
"step": 10500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 22.83,
|
| 114 |
+
"eval_cer": 0.13394260888350154,
|
| 115 |
+
"eval_loss": 0.30638357996940613,
|
| 116 |
+
"eval_runtime": 205.7125,
|
| 117 |
+
"eval_samples_per_second": 21.729,
|
| 118 |
+
"eval_steps_per_second": 1.361,
|
| 119 |
+
"eval_wer": 0.29559965174490316,
|
| 120 |
+
"step": 10500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 26.09,
|
| 124 |
+
"learning_rate": 7.111822222222222e-05,
|
| 125 |
+
"loss": 1.0493,
|
| 126 |
+
"step": 12000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 26.09,
|
| 130 |
+
"eval_cer": 0.1299701162944205,
|
| 131 |
+
"eval_loss": 0.29015836119651794,
|
| 132 |
+
"eval_runtime": 205.4201,
|
| 133 |
+
"eval_samples_per_second": 21.76,
|
| 134 |
+
"eval_steps_per_second": 1.363,
|
| 135 |
+
"eval_wer": 0.2849162011173184,
|
| 136 |
+
"step": 12000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 29.35,
|
| 140 |
+
"learning_rate": 6.97848888888889e-05,
|
| 141 |
+
"loss": 1.0193,
|
| 142 |
+
"step": 13500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 29.35,
|
| 146 |
+
"eval_cer": 0.12554156715432713,
|
| 147 |
+
"eval_loss": 0.27857705950737,
|
| 148 |
+
"eval_runtime": 204.6144,
|
| 149 |
+
"eval_samples_per_second": 21.846,
|
| 150 |
+
"eval_steps_per_second": 1.368,
|
| 151 |
+
"eval_wer": 0.2710222738155699,
|
| 152 |
+
"step": 13500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 32.61,
|
| 156 |
+
"learning_rate": 6.845333333333334e-05,
|
| 157 |
+
"loss": 0.991,
|
| 158 |
+
"step": 15000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 32.61,
|
| 162 |
+
"eval_cer": 0.12242118022634807,
|
| 163 |
+
"eval_loss": 0.27255550026893616,
|
| 164 |
+
"eval_runtime": 205.3956,
|
| 165 |
+
"eval_samples_per_second": 21.763,
|
| 166 |
+
"eval_steps_per_second": 1.363,
|
| 167 |
+
"eval_wer": 0.2632046724225495,
|
| 168 |
+
"step": 15000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 35.87,
|
| 172 |
+
"learning_rate": 6.712000000000001e-05,
|
| 173 |
+
"loss": 0.965,
|
| 174 |
+
"step": 16500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 35.87,
|
| 178 |
+
"eval_cer": 0.1227452204073305,
|
| 179 |
+
"eval_loss": 0.2663341462612152,
|
| 180 |
+
"eval_runtime": 207.401,
|
| 181 |
+
"eval_samples_per_second": 21.552,
|
| 182 |
+
"eval_steps_per_second": 1.35,
|
| 183 |
+
"eval_wer": 0.2638395124428644,
|
| 184 |
+
"step": 16500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 39.13,
|
| 188 |
+
"learning_rate": 6.578755555555556e-05,
|
| 189 |
+
"loss": 0.9443,
|
| 190 |
+
"step": 18000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 39.13,
|
| 194 |
+
"eval_cer": 0.11871272037732679,
|
| 195 |
+
"eval_loss": 0.2498483508825302,
|
| 196 |
+
"eval_runtime": 206.2679,
|
| 197 |
+
"eval_samples_per_second": 21.671,
|
| 198 |
+
"eval_steps_per_second": 1.357,
|
| 199 |
+
"eval_wer": 0.2540629761300152,
|
| 200 |
+
"step": 18000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 42.39,
|
| 204 |
+
"learning_rate": 6.445422222222224e-05,
|
| 205 |
+
"loss": 0.9258,
|
| 206 |
+
"step": 19500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 42.39,
|
| 210 |
+
"eval_cer": 0.11627641827586621,
|
| 211 |
+
"eval_loss": 0.24673064053058624,
|
| 212 |
+
"eval_runtime": 211.3131,
|
| 213 |
+
"eval_samples_per_second": 21.153,
|
| 214 |
+
"eval_steps_per_second": 1.325,
|
| 215 |
+
"eval_wer": 0.24822244794311835,
|
| 216 |
+
"step": 19500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 45.65,
|
| 220 |
+
"learning_rate": 6.312088888888889e-05,
|
| 221 |
+
"loss": 0.9062,
|
| 222 |
+
"step": 21000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 45.65,
|
| 226 |
+
"eval_cer": 0.11399013477671231,
|
| 227 |
+
"eval_loss": 0.23752263188362122,
|
| 228 |
+
"eval_runtime": 210.4679,
|
| 229 |
+
"eval_samples_per_second": 21.238,
|
| 230 |
+
"eval_steps_per_second": 1.33,
|
| 231 |
+
"eval_wer": 0.24243633461510555,
|
| 232 |
+
"step": 21000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 48.91,
|
| 236 |
+
"learning_rate": 6.178844444444445e-05,
|
| 237 |
+
"loss": 0.8797,
|
| 238 |
+
"step": 22500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 48.91,
|
| 242 |
+
"eval_cer": 0.11319803655653302,
|
| 243 |
+
"eval_loss": 0.23775836825370789,
|
| 244 |
+
"eval_runtime": 200.5119,
|
| 245 |
+
"eval_samples_per_second": 22.293,
|
| 246 |
+
"eval_steps_per_second": 1.396,
|
| 247 |
+
"eval_wer": 0.24310745120800986,
|
| 248 |
+
"step": 22500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 52.17,
|
| 252 |
+
"learning_rate": 6.0456000000000005e-05,
|
| 253 |
+
"loss": 0.8691,
|
| 254 |
+
"step": 24000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 52.17,
|
| 258 |
+
"eval_cer": 0.11229792494269289,
|
| 259 |
+
"eval_loss": 0.23617656528949738,
|
| 260 |
+
"eval_runtime": 200.8473,
|
| 261 |
+
"eval_samples_per_second": 22.256,
|
| 262 |
+
"eval_steps_per_second": 1.394,
|
| 263 |
+
"eval_wer": 0.23674091271856634,
|
| 264 |
+
"step": 24000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 55.43,
|
| 268 |
+
"learning_rate": 5.912266666666667e-05,
|
| 269 |
+
"loss": 0.8554,
|
| 270 |
+
"step": 25500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 55.43,
|
| 274 |
+
"eval_cer": 0.11031767939224464,
|
| 275 |
+
"eval_loss": 0.23144488036632538,
|
| 276 |
+
"eval_runtime": 200.5411,
|
| 277 |
+
"eval_samples_per_second": 22.29,
|
| 278 |
+
"eval_steps_per_second": 1.396,
|
| 279 |
+
"eval_wer": 0.23295001088297176,
|
| 280 |
+
"step": 25500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 58.69,
|
| 284 |
+
"learning_rate": 5.778933333333334e-05,
|
| 285 |
+
"loss": 0.8393,
|
| 286 |
+
"step": 27000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 58.69,
|
| 290 |
+
"eval_cer": 0.10892550676283859,
|
| 291 |
+
"eval_loss": 0.22345688939094543,
|
| 292 |
+
"eval_runtime": 199.926,
|
| 293 |
+
"eval_samples_per_second": 22.358,
|
| 294 |
+
"eval_steps_per_second": 1.401,
|
| 295 |
+
"eval_wer": 0.22779873757527389,
|
| 296 |
+
"step": 27000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 61.96,
|
| 300 |
+
"learning_rate": 5.645688888888889e-05,
|
| 301 |
+
"loss": 0.8268,
|
| 302 |
+
"step": 28500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 61.96,
|
| 306 |
+
"eval_cer": 0.10809140333401342,
|
| 307 |
+
"eval_loss": 0.21886605024337769,
|
| 308 |
+
"eval_runtime": 200.5151,
|
| 309 |
+
"eval_samples_per_second": 22.293,
|
| 310 |
+
"eval_steps_per_second": 1.396,
|
| 311 |
+
"eval_wer": 0.2251324094899514,
|
| 312 |
+
"step": 28500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 65.22,
|
| 316 |
+
"learning_rate": 5.512355555555556e-05,
|
| 317 |
+
"loss": 0.8116,
|
| 318 |
+
"step": 30000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 65.22,
|
| 322 |
+
"eval_cer": 0.10730530585792639,
|
| 323 |
+
"eval_loss": 0.22185355424880981,
|
| 324 |
+
"eval_runtime": 199.8759,
|
| 325 |
+
"eval_samples_per_second": 22.364,
|
| 326 |
+
"eval_steps_per_second": 1.401,
|
| 327 |
+
"eval_wer": 0.2221395922513241,
|
| 328 |
+
"step": 30000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 68.48,
|
| 332 |
+
"learning_rate": 5.3791111111111116e-05,
|
| 333 |
+
"loss": 0.7981,
|
| 334 |
+
"step": 31500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 68.48,
|
| 338 |
+
"eval_cer": 0.10677724037780685,
|
| 339 |
+
"eval_loss": 0.21808896958827972,
|
| 340 |
+
"eval_runtime": 201.4847,
|
| 341 |
+
"eval_samples_per_second": 22.185,
|
| 342 |
+
"eval_steps_per_second": 1.39,
|
| 343 |
+
"eval_wer": 0.221831241384314,
|
| 344 |
+
"step": 31500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 71.74,
|
| 348 |
+
"learning_rate": 5.245777777777778e-05,
|
| 349 |
+
"loss": 0.7839,
|
| 350 |
+
"step": 33000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 71.74,
|
| 354 |
+
"eval_cer": 0.10549308114206161,
|
| 355 |
+
"eval_loss": 0.21510420739650726,
|
| 356 |
+
"eval_runtime": 200.7972,
|
| 357 |
+
"eval_samples_per_second": 22.261,
|
| 358 |
+
"eval_steps_per_second": 1.394,
|
| 359 |
+
"eval_wer": 0.2175143292461728,
|
| 360 |
+
"step": 33000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 75.0,
|
| 364 |
+
"learning_rate": 5.112533333333334e-05,
|
| 365 |
+
"loss": 0.7714,
|
| 366 |
+
"step": 34500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 75.0,
|
| 370 |
+
"eval_cer": 0.10464097548095964,
|
| 371 |
+
"eval_loss": 0.21727131307125092,
|
| 372 |
+
"eval_runtime": 200.0575,
|
| 373 |
+
"eval_samples_per_second": 22.344,
|
| 374 |
+
"eval_steps_per_second": 1.4,
|
| 375 |
+
"eval_wer": 0.21642603206849018,
|
| 376 |
+
"step": 34500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 78.26,
|
| 380 |
+
"learning_rate": 4.9792e-05,
|
| 381 |
+
"loss": 0.7582,
|
| 382 |
+
"step": 36000
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 78.26,
|
| 386 |
+
"eval_cer": 0.10387288023714941,
|
| 387 |
+
"eval_loss": 0.21982263028621674,
|
| 388 |
+
"eval_runtime": 201.7359,
|
| 389 |
+
"eval_samples_per_second": 22.158,
|
| 390 |
+
"eval_steps_per_second": 1.388,
|
| 391 |
+
"eval_wer": 0.2148298628745556,
|
| 392 |
+
"step": 36000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 81.52,
|
| 396 |
+
"learning_rate": 4.845955555555556e-05,
|
| 397 |
+
"loss": 0.7486,
|
| 398 |
+
"step": 37500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 81.52,
|
| 402 |
+
"eval_cer": 0.10329080806019947,
|
| 403 |
+
"eval_loss": 0.21195770800113678,
|
| 404 |
+
"eval_runtime": 199.1645,
|
| 405 |
+
"eval_samples_per_second": 22.444,
|
| 406 |
+
"eval_steps_per_second": 1.406,
|
| 407 |
+
"eval_wer": 0.21250816222883262,
|
| 408 |
+
"step": 37500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 84.78,
|
| 412 |
+
"learning_rate": 4.7127111111111114e-05,
|
| 413 |
+
"loss": 0.7326,
|
| 414 |
+
"step": 39000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 84.78,
|
| 418 |
+
"eval_cer": 0.10266072993051138,
|
| 419 |
+
"eval_loss": 0.21248655021190643,
|
| 420 |
+
"eval_runtime": 199.3582,
|
| 421 |
+
"eval_samples_per_second": 22.422,
|
| 422 |
+
"eval_steps_per_second": 1.405,
|
| 423 |
+
"eval_wer": 0.2124537473699485,
|
| 424 |
+
"step": 39000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 88.04,
|
| 428 |
+
"learning_rate": 4.5793777777777784e-05,
|
| 429 |
+
"loss": 0.7263,
|
| 430 |
+
"step": 40500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 88.04,
|
| 434 |
+
"eval_cer": 0.1012565558129208,
|
| 435 |
+
"eval_loss": 0.20737077295780182,
|
| 436 |
+
"eval_runtime": 206.4127,
|
| 437 |
+
"eval_samples_per_second": 21.656,
|
| 438 |
+
"eval_steps_per_second": 1.357,
|
| 439 |
+
"eval_wer": 0.2079735906551549,
|
| 440 |
+
"step": 40500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 91.3,
|
| 444 |
+
"learning_rate": 4.446044444444445e-05,
|
| 445 |
+
"loss": 0.7142,
|
| 446 |
+
"step": 42000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 91.3,
|
| 450 |
+
"eval_cer": 0.10114254167516772,
|
| 451 |
+
"eval_loss": 0.21454070508480072,
|
| 452 |
+
"eval_runtime": 204.8009,
|
| 453 |
+
"eval_samples_per_second": 21.826,
|
| 454 |
+
"eval_steps_per_second": 1.367,
|
| 455 |
+
"eval_wer": 0.2069397083363564,
|
| 456 |
+
"step": 42000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 94.56,
|
| 460 |
+
"learning_rate": 4.312711111111111e-05,
|
| 461 |
+
"loss": 0.7044,
|
| 462 |
+
"step": 43500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 94.56,
|
| 466 |
+
"eval_cer": 0.10105253051378371,
|
| 467 |
+
"eval_loss": 0.20794141292572021,
|
| 468 |
+
"eval_runtime": 205.8959,
|
| 469 |
+
"eval_samples_per_second": 21.71,
|
| 470 |
+
"eval_steps_per_second": 1.36,
|
| 471 |
+
"eval_wer": 0.2077922077922078,
|
| 472 |
+
"step": 43500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 97.83,
|
| 476 |
+
"learning_rate": 4.179555555555556e-05,
|
| 477 |
+
"loss": 0.6975,
|
| 478 |
+
"step": 45000
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 97.83,
|
| 482 |
+
"eval_cer": 0.0996543571402854,
|
| 483 |
+
"eval_loss": 0.20573750138282776,
|
| 484 |
+
"eval_runtime": 205.1157,
|
| 485 |
+
"eval_samples_per_second": 21.793,
|
| 486 |
+
"eval_steps_per_second": 1.365,
|
| 487 |
+
"eval_wer": 0.2029855619241094,
|
| 488 |
+
"step": 45000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 101.09,
|
| 492 |
+
"learning_rate": 4.0462222222222225e-05,
|
| 493 |
+
"loss": 0.683,
|
| 494 |
+
"step": 46500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 101.09,
|
| 498 |
+
"eval_cer": 0.09906628421924318,
|
| 499 |
+
"eval_loss": 0.20465880632400513,
|
| 500 |
+
"eval_runtime": 205.9334,
|
| 501 |
+
"eval_samples_per_second": 21.706,
|
| 502 |
+
"eval_steps_per_second": 1.36,
|
| 503 |
+
"eval_wer": 0.20209678589566857,
|
| 504 |
+
"step": 46500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 104.35,
|
| 508 |
+
"learning_rate": 3.913066666666667e-05,
|
| 509 |
+
"loss": 0.6763,
|
| 510 |
+
"step": 48000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 104.35,
|
| 514 |
+
"eval_cer": 0.0991922998451808,
|
| 515 |
+
"eval_loss": 0.20821230113506317,
|
| 516 |
+
"eval_runtime": 207.1439,
|
| 517 |
+
"eval_samples_per_second": 21.579,
|
| 518 |
+
"eval_steps_per_second": 1.352,
|
| 519 |
+
"eval_wer": 0.20189726474642675,
|
| 520 |
+
"step": 48000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 107.61,
|
| 524 |
+
"learning_rate": 3.779733333333333e-05,
|
| 525 |
+
"loss": 0.6687,
|
| 526 |
+
"step": 49500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 107.61,
|
| 530 |
+
"eval_cer": 0.09791414135352784,
|
| 531 |
+
"eval_loss": 0.20512840151786804,
|
| 532 |
+
"eval_runtime": 206.0617,
|
| 533 |
+
"eval_samples_per_second": 21.693,
|
| 534 |
+
"eval_steps_per_second": 1.359,
|
| 535 |
+
"eval_wer": 0.19807008633824277,
|
| 536 |
+
"step": 49500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 110.87,
|
| 540 |
+
"learning_rate": 3.6464e-05,
|
| 541 |
+
"loss": 0.6573,
|
| 542 |
+
"step": 51000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 110.87,
|
| 546 |
+
"eval_cer": 0.09767411158983715,
|
| 547 |
+
"eval_loss": 0.20361328125,
|
| 548 |
+
"eval_runtime": 205.3176,
|
| 549 |
+
"eval_samples_per_second": 21.771,
|
| 550 |
+
"eval_steps_per_second": 1.364,
|
| 551 |
+
"eval_wer": 0.197616629180875,
|
| 552 |
+
"step": 51000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 114.13,
|
| 556 |
+
"learning_rate": 3.513155555555556e-05,
|
| 557 |
+
"loss": 0.6497,
|
| 558 |
+
"step": 52500
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 114.13,
|
| 562 |
+
"eval_cer": 0.09658197616504446,
|
| 563 |
+
"eval_loss": 0.2067527323961258,
|
| 564 |
+
"eval_runtime": 206.5547,
|
| 565 |
+
"eval_samples_per_second": 21.641,
|
| 566 |
+
"eval_steps_per_second": 1.356,
|
| 567 |
+
"eval_wer": 0.19573024740622505,
|
| 568 |
+
"step": 52500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 117.39,
|
| 572 |
+
"learning_rate": 3.379822222222222e-05,
|
| 573 |
+
"loss": 0.6406,
|
| 574 |
+
"step": 54000
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 117.39,
|
| 578 |
+
"eval_cer": 0.09563985934255848,
|
| 579 |
+
"eval_loss": 0.20097151398658752,
|
| 580 |
+
"eval_runtime": 205.4638,
|
| 581 |
+
"eval_samples_per_second": 21.756,
|
| 582 |
+
"eval_steps_per_second": 1.363,
|
| 583 |
+
"eval_wer": 0.1919212072843358,
|
| 584 |
+
"step": 54000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 120.65,
|
| 588 |
+
"learning_rate": 3.2466666666666665e-05,
|
| 589 |
+
"loss": 0.6313,
|
| 590 |
+
"step": 55500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 120.65,
|
| 594 |
+
"eval_cer": 0.09662398137369034,
|
| 595 |
+
"eval_loss": 0.20263761281967163,
|
| 596 |
+
"eval_runtime": 210.159,
|
| 597 |
+
"eval_samples_per_second": 21.27,
|
| 598 |
+
"eval_steps_per_second": 1.332,
|
| 599 |
+
"eval_wer": 0.19520423710367846,
|
| 600 |
+
"step": 55500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 123.91,
|
| 604 |
+
"learning_rate": 3.1133333333333336e-05,
|
| 605 |
+
"loss": 0.6261,
|
| 606 |
+
"step": 57000
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 123.91,
|
| 610 |
+
"eval_cer": 0.09623993375178522,
|
| 611 |
+
"eval_loss": 0.20339860022068024,
|
| 612 |
+
"eval_runtime": 208.9616,
|
| 613 |
+
"eval_samples_per_second": 21.391,
|
| 614 |
+
"eval_steps_per_second": 1.34,
|
| 615 |
+
"eval_wer": 0.1936080679097439,
|
| 616 |
+
"step": 57000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 127.17,
|
| 620 |
+
"learning_rate": 2.9800000000000003e-05,
|
| 621 |
+
"loss": 0.6179,
|
| 622 |
+
"step": 58500
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 127.17,
|
| 626 |
+
"eval_cer": 0.09569986678348115,
|
| 627 |
+
"eval_loss": 0.20739464461803436,
|
| 628 |
+
"eval_runtime": 215.3433,
|
| 629 |
+
"eval_samples_per_second": 20.758,
|
| 630 |
+
"eval_steps_per_second": 1.3,
|
| 631 |
+
"eval_wer": 0.1929550896031343,
|
| 632 |
+
"step": 58500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 130.43,
|
| 636 |
+
"learning_rate": 2.846755555555556e-05,
|
| 637 |
+
"loss": 0.6082,
|
| 638 |
+
"step": 60000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 130.43,
|
| 642 |
+
"eval_cer": 0.09512979609471574,
|
| 643 |
+
"eval_loss": 0.20194286108016968,
|
| 644 |
+
"eval_runtime": 213.4283,
|
| 645 |
+
"eval_samples_per_second": 20.944,
|
| 646 |
+
"eval_steps_per_second": 1.312,
|
| 647 |
+
"eval_wer": 0.1902887615178118,
|
| 648 |
+
"step": 60000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 133.69,
|
| 652 |
+
"learning_rate": 2.7134222222222226e-05,
|
| 653 |
+
"loss": 0.6039,
|
| 654 |
+
"step": 61500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 133.69,
|
| 658 |
+
"eval_cer": 0.09520780576791522,
|
| 659 |
+
"eval_loss": 0.20227235555648804,
|
| 660 |
+
"eval_runtime": 209.2948,
|
| 661 |
+
"eval_samples_per_second": 21.357,
|
| 662 |
+
"eval_steps_per_second": 1.338,
|
| 663 |
+
"eval_wer": 0.1912682289777262,
|
| 664 |
+
"step": 61500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 136.96,
|
| 668 |
+
"learning_rate": 2.580177777777778e-05,
|
| 669 |
+
"loss": 0.5948,
|
| 670 |
+
"step": 63000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 136.96,
|
| 674 |
+
"eval_cer": 0.09544183478751365,
|
| 675 |
+
"eval_loss": 0.20598873496055603,
|
| 676 |
+
"eval_runtime": 207.4517,
|
| 677 |
+
"eval_samples_per_second": 21.547,
|
| 678 |
+
"eval_steps_per_second": 1.35,
|
| 679 |
+
"eval_wer": 0.19108684611477908,
|
| 680 |
+
"step": 63000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 140.22,
|
| 684 |
+
"learning_rate": 2.4468444444444447e-05,
|
| 685 |
+
"loss": 0.5867,
|
| 686 |
+
"step": 64500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 140.22,
|
| 690 |
+
"eval_cer": 0.09470374326416475,
|
| 691 |
+
"eval_loss": 0.20384302735328674,
|
| 692 |
+
"eval_runtime": 209.3468,
|
| 693 |
+
"eval_samples_per_second": 21.352,
|
| 694 |
+
"eval_steps_per_second": 1.337,
|
| 695 |
+
"eval_wer": 0.18905535804977147,
|
| 696 |
+
"step": 64500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 143.48,
|
| 700 |
+
"learning_rate": 2.3136000000000003e-05,
|
| 701 |
+
"loss": 0.5804,
|
| 702 |
+
"step": 66000
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 143.48,
|
| 706 |
+
"eval_cer": 0.09508779088606988,
|
| 707 |
+
"eval_loss": 0.2093733698129654,
|
| 708 |
+
"eval_runtime": 212.8045,
|
| 709 |
+
"eval_samples_per_second": 21.005,
|
| 710 |
+
"eval_steps_per_second": 1.316,
|
| 711 |
+
"eval_wer": 0.19039759123558006,
|
| 712 |
+
"step": 66000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 146.74,
|
| 716 |
+
"learning_rate": 2.1803555555555557e-05,
|
| 717 |
+
"loss": 0.5757,
|
| 718 |
+
"step": 67500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 146.74,
|
| 722 |
+
"eval_cer": 0.09395365025263133,
|
| 723 |
+
"eval_loss": 0.20662076771259308,
|
| 724 |
+
"eval_runtime": 210.0316,
|
| 725 |
+
"eval_samples_per_second": 21.283,
|
| 726 |
+
"eval_steps_per_second": 1.333,
|
| 727 |
+
"eval_wer": 0.18718711456141623,
|
| 728 |
+
"step": 67500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 150.0,
|
| 732 |
+
"learning_rate": 2.0470222222222224e-05,
|
| 733 |
+
"loss": 0.5684,
|
| 734 |
+
"step": 69000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 150.0,
|
| 738 |
+
"eval_cer": 0.09401965843764627,
|
| 739 |
+
"eval_loss": 0.20758455991744995,
|
| 740 |
+
"eval_runtime": 210.1895,
|
| 741 |
+
"eval_samples_per_second": 21.267,
|
| 742 |
+
"eval_steps_per_second": 1.332,
|
| 743 |
+
"eval_wer": 0.1878038162954364,
|
| 744 |
+
"step": 69000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 153.26,
|
| 748 |
+
"learning_rate": 1.913777777777778e-05,
|
| 749 |
+
"loss": 0.5622,
|
| 750 |
+
"step": 70500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 153.26,
|
| 754 |
+
"eval_cer": 0.09470974400825702,
|
| 755 |
+
"eval_loss": 0.21354267001152039,
|
| 756 |
+
"eval_runtime": 210.2099,
|
| 757 |
+
"eval_samples_per_second": 21.264,
|
| 758 |
+
"eval_steps_per_second": 1.332,
|
| 759 |
+
"eval_wer": 0.1891641877675397,
|
| 760 |
+
"step": 70500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 156.52,
|
| 764 |
+
"learning_rate": 1.7805333333333334e-05,
|
| 765 |
+
"loss": 0.5572,
|
| 766 |
+
"step": 72000
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 156.52,
|
| 770 |
+
"eval_cer": 0.09412167108721481,
|
| 771 |
+
"eval_loss": 0.21093089878559113,
|
| 772 |
+
"eval_runtime": 208.6793,
|
| 773 |
+
"eval_samples_per_second": 21.42,
|
| 774 |
+
"eval_steps_per_second": 1.342,
|
| 775 |
+
"eval_wer": 0.18854748603351956,
|
| 776 |
+
"step": 72000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 159.78,
|
| 780 |
+
"learning_rate": 1.647288888888889e-05,
|
| 781 |
+
"loss": 0.5513,
|
| 782 |
+
"step": 73500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 159.78,
|
| 786 |
+
"eval_cer": 0.09389964355580092,
|
| 787 |
+
"eval_loss": 0.20995952188968658,
|
| 788 |
+
"eval_runtime": 211.9593,
|
| 789 |
+
"eval_samples_per_second": 21.089,
|
| 790 |
+
"eval_steps_per_second": 1.321,
|
| 791 |
+
"eval_wer": 0.1872959442791845,
|
| 792 |
+
"step": 73500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 163.04,
|
| 796 |
+
"learning_rate": 1.5139555555555556e-05,
|
| 797 |
+
"loss": 0.5463,
|
| 798 |
+
"step": 75000
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 163.04,
|
| 802 |
+
"eval_cer": 0.09368961751257156,
|
| 803 |
+
"eval_loss": 0.2099747508764267,
|
| 804 |
+
"eval_runtime": 206.1662,
|
| 805 |
+
"eval_samples_per_second": 21.682,
|
| 806 |
+
"eval_steps_per_second": 1.358,
|
| 807 |
+
"eval_wer": 0.18666110425886961,
|
| 808 |
+
"step": 75000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 166.3,
|
| 812 |
+
"learning_rate": 1.3807111111111111e-05,
|
| 813 |
+
"loss": 0.5402,
|
| 814 |
+
"step": 76500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 166.3,
|
| 818 |
+
"eval_cer": 0.0937376234653097,
|
| 819 |
+
"eval_loss": 0.2125527560710907,
|
| 820 |
+
"eval_runtime": 206.4719,
|
| 821 |
+
"eval_samples_per_second": 21.649,
|
| 822 |
+
"eval_steps_per_second": 1.356,
|
| 823 |
+
"eval_wer": 0.18675179569034317,
|
| 824 |
+
"step": 76500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 169.56,
|
| 828 |
+
"learning_rate": 1.2473777777777778e-05,
|
| 829 |
+
"loss": 0.5361,
|
| 830 |
+
"step": 78000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 169.56,
|
| 834 |
+
"eval_cer": 0.09316155203245202,
|
| 835 |
+
"eval_loss": 0.20593136548995972,
|
| 836 |
+
"eval_runtime": 204.7532,
|
| 837 |
+
"eval_samples_per_second": 21.831,
|
| 838 |
+
"eval_steps_per_second": 1.368,
|
| 839 |
+
"eval_wer": 0.18530073278676631,
|
| 840 |
+
"step": 78000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 172.83,
|
| 844 |
+
"learning_rate": 1.1140444444444445e-05,
|
| 845 |
+
"loss": 0.5322,
|
| 846 |
+
"step": 79500
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 172.83,
|
| 850 |
+
"eval_cer": 0.09320955798519016,
|
| 851 |
+
"eval_loss": 0.20699192583560944,
|
| 852 |
+
"eval_runtime": 206.6944,
|
| 853 |
+
"eval_samples_per_second": 21.626,
|
| 854 |
+
"eval_steps_per_second": 1.355,
|
| 855 |
+
"eval_wer": 0.1852825945004716,
|
| 856 |
+
"step": 79500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 176.09,
|
| 860 |
+
"learning_rate": 9.807111111111112e-06,
|
| 861 |
+
"loss": 0.5274,
|
| 862 |
+
"step": 81000
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 176.09,
|
| 866 |
+
"eval_cer": 0.0925194724145794,
|
| 867 |
+
"eval_loss": 0.2075292468070984,
|
| 868 |
+
"eval_runtime": 205.8591,
|
| 869 |
+
"eval_samples_per_second": 21.714,
|
| 870 |
+
"eval_steps_per_second": 1.36,
|
| 871 |
+
"eval_wer": 0.18423057389537836,
|
| 872 |
+
"step": 81000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 179.35,
|
| 876 |
+
"learning_rate": 8.474666666666667e-06,
|
| 877 |
+
"loss": 0.5229,
|
| 878 |
+
"step": 82500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 179.35,
|
| 882 |
+
"eval_cer": 0.09307154087106802,
|
| 883 |
+
"eval_loss": 0.2078840136528015,
|
| 884 |
+
"eval_runtime": 207.0411,
|
| 885 |
+
"eval_samples_per_second": 21.59,
|
| 886 |
+
"eval_steps_per_second": 1.352,
|
| 887 |
+
"eval_wer": 0.18519190306899805,
|
| 888 |
+
"step": 82500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 182.61,
|
| 892 |
+
"learning_rate": 7.141333333333333e-06,
|
| 893 |
+
"loss": 0.5167,
|
| 894 |
+
"step": 84000
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 182.61,
|
| 898 |
+
"eval_cer": 0.09254347539094848,
|
| 899 |
+
"eval_loss": 0.2052079439163208,
|
| 900 |
+
"eval_runtime": 205.153,
|
| 901 |
+
"eval_samples_per_second": 21.789,
|
| 902 |
+
"eval_steps_per_second": 1.365,
|
| 903 |
+
"eval_wer": 0.1834869041572952,
|
| 904 |
+
"step": 84000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 185.87,
|
| 908 |
+
"learning_rate": 5.80888888888889e-06,
|
| 909 |
+
"loss": 0.515,
|
| 910 |
+
"step": 85500
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 185.87,
|
| 914 |
+
"eval_cer": 0.0924474634854722,
|
| 915 |
+
"eval_loss": 0.20651493966579437,
|
| 916 |
+
"eval_runtime": 206.3346,
|
| 917 |
+
"eval_samples_per_second": 21.664,
|
| 918 |
+
"eval_steps_per_second": 1.357,
|
| 919 |
+
"eval_wer": 0.1834869041572952,
|
| 920 |
+
"step": 85500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 189.13,
|
| 924 |
+
"learning_rate": 4.476444444444445e-06,
|
| 925 |
+
"loss": 0.5116,
|
| 926 |
+
"step": 87000
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 189.13,
|
| 930 |
+
"eval_cer": 0.09247746720593354,
|
| 931 |
+
"eval_loss": 0.20912431180477142,
|
| 932 |
+
"eval_runtime": 206.6532,
|
| 933 |
+
"eval_samples_per_second": 21.63,
|
| 934 |
+
"eval_steps_per_second": 1.355,
|
| 935 |
+
"eval_wer": 0.18323296814916926,
|
| 936 |
+
"step": 87000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 192.39,
|
| 940 |
+
"learning_rate": 3.143111111111111e-06,
|
| 941 |
+
"loss": 0.5097,
|
| 942 |
+
"step": 88500
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 192.39,
|
| 946 |
+
"eval_cer": 0.0925194724145794,
|
| 947 |
+
"eval_loss": 0.2088840752840042,
|
| 948 |
+
"eval_runtime": 206.9892,
|
| 949 |
+
"eval_samples_per_second": 21.595,
|
| 950 |
+
"eval_steps_per_second": 1.353,
|
| 951 |
+
"eval_wer": 0.18355945730247406,
|
| 952 |
+
"step": 88500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 195.65,
|
| 956 |
+
"learning_rate": 1.8106666666666667e-06,
|
| 957 |
+
"loss": 0.5056,
|
| 958 |
+
"step": 90000
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 195.65,
|
| 962 |
+
"eval_cer": 0.09236345306818046,
|
| 963 |
+
"eval_loss": 0.2069932371377945,
|
| 964 |
+
"eval_runtime": 205.214,
|
| 965 |
+
"eval_samples_per_second": 21.782,
|
| 966 |
+
"eval_steps_per_second": 1.364,
|
| 967 |
+
"eval_wer": 0.1829608938547486,
|
| 968 |
+
"step": 90000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 198.91,
|
| 972 |
+
"learning_rate": 4.782222222222222e-07,
|
| 973 |
+
"loss": 0.5068,
|
| 974 |
+
"step": 91500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 198.91,
|
| 978 |
+
"eval_cer": 0.09240545827682632,
|
| 979 |
+
"eval_loss": 0.20784138143062592,
|
| 980 |
+
"eval_runtime": 205.9346,
|
| 981 |
+
"eval_samples_per_second": 21.706,
|
| 982 |
+
"eval_steps_per_second": 1.36,
|
| 983 |
+
"eval_wer": 0.183124138431401,
|
| 984 |
+
"step": 91500
|
| 985 |
+
}
|
| 986 |
+
],
|
| 987 |
+
"max_steps": 92000,
|
| 988 |
+
"num_train_epochs": 200,
|
| 989 |
+
"total_flos": 1.0091178676342205e+21,
|
| 990 |
+
"trial_name": null,
|
| 991 |
+
"trial_params": null
|
| 992 |
+
}
|