End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +535 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 49.64,
|
| 3 |
+
"eval_loss": 0.8213908076286316,
|
| 4 |
+
"eval_runtime": 57.4529,
|
| 5 |
+
"eval_samples": 842,
|
| 6 |
+
"eval_samples_per_second": 14.655,
|
| 7 |
+
"eval_steps_per_second": 0.922,
|
| 8 |
+
"eval_wer": 0.9639684823207993,
|
| 9 |
+
"train_loss": 1.3979950288625864,
|
| 10 |
+
"train_runtime": 89308.2068,
|
| 11 |
+
"train_samples": 3351,
|
| 12 |
+
"train_samples_per_second": 1.876,
|
| 13 |
+
"train_steps_per_second": 0.058
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 49.64,
|
| 3 |
+
"eval_loss": 0.8213908076286316,
|
| 4 |
+
"eval_runtime": 57.4529,
|
| 5 |
+
"eval_samples": 842,
|
| 6 |
+
"eval_samples_per_second": 14.655,
|
| 7 |
+
"eval_steps_per_second": 0.922,
|
| 8 |
+
"eval_wer": 0.9639684823207993
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 49.64,
|
| 3 |
+
"train_loss": 1.3979950288625864,
|
| 4 |
+
"train_runtime": 89308.2068,
|
| 5 |
+
"train_samples": 3351,
|
| 6 |
+
"train_samples_per_second": 1.876,
|
| 7 |
+
"train_steps_per_second": 0.058
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,535 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8213908076286316,
|
| 3 |
+
"best_model_checkpoint": "./wav2vec2-base-960h-paper/checkpoint-3352",
|
| 4 |
+
"epoch": 49.64200477326969,
|
| 5 |
+
"global_step": 5200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.99,
|
| 12 |
+
"eval_loss": 5.683954238891602,
|
| 13 |
+
"eval_runtime": 61.9414,
|
| 14 |
+
"eval_samples_per_second": 13.594,
|
| 15 |
+
"eval_steps_per_second": 0.856,
|
| 16 |
+
"eval_wer": 1.0,
|
| 17 |
+
"step": 104
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 2.0,
|
| 21 |
+
"eval_loss": 3.9771974086761475,
|
| 22 |
+
"eval_runtime": 58.5656,
|
| 23 |
+
"eval_samples_per_second": 14.377,
|
| 24 |
+
"eval_steps_per_second": 0.905,
|
| 25 |
+
"eval_wer": 1.0,
|
| 26 |
+
"step": 209
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 3.0,
|
| 30 |
+
"eval_loss": 3.4203968048095703,
|
| 31 |
+
"eval_runtime": 58.5182,
|
| 32 |
+
"eval_samples_per_second": 14.389,
|
| 33 |
+
"eval_steps_per_second": 0.906,
|
| 34 |
+
"eval_wer": 1.0,
|
| 35 |
+
"step": 314
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 4.0,
|
| 39 |
+
"eval_loss": 3.3691885471343994,
|
| 40 |
+
"eval_runtime": 58.4394,
|
| 41 |
+
"eval_samples_per_second": 14.408,
|
| 42 |
+
"eval_steps_per_second": 0.907,
|
| 43 |
+
"eval_wer": 1.0,
|
| 44 |
+
"step": 419
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 4.77,
|
| 48 |
+
"learning_rate": 4.916317991631799e-05,
|
| 49 |
+
"loss": 5.5612,
|
| 50 |
+
"step": 500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.99,
|
| 54 |
+
"eval_loss": 3.3944718837738037,
|
| 55 |
+
"eval_runtime": 58.5781,
|
| 56 |
+
"eval_samples_per_second": 14.374,
|
| 57 |
+
"eval_steps_per_second": 0.905,
|
| 58 |
+
"eval_wer": 1.0,
|
| 59 |
+
"step": 523
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 6.0,
|
| 63 |
+
"eval_loss": 3.342637777328491,
|
| 64 |
+
"eval_runtime": 58.3305,
|
| 65 |
+
"eval_samples_per_second": 14.435,
|
| 66 |
+
"eval_steps_per_second": 0.909,
|
| 67 |
+
"eval_wer": 1.0,
|
| 68 |
+
"step": 628
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 7.0,
|
| 72 |
+
"eval_loss": 3.3333232402801514,
|
| 73 |
+
"eval_runtime": 58.4524,
|
| 74 |
+
"eval_samples_per_second": 14.405,
|
| 75 |
+
"eval_steps_per_second": 0.907,
|
| 76 |
+
"eval_wer": 1.0,
|
| 77 |
+
"step": 733
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 8.0,
|
| 81 |
+
"eval_loss": 3.329634428024292,
|
| 82 |
+
"eval_runtime": 58.6528,
|
| 83 |
+
"eval_samples_per_second": 14.356,
|
| 84 |
+
"eval_steps_per_second": 0.904,
|
| 85 |
+
"eval_wer": 1.0000659917906212,
|
| 86 |
+
"step": 838
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 8.99,
|
| 90 |
+
"eval_loss": 3.1853039264678955,
|
| 91 |
+
"eval_runtime": 58.7103,
|
| 92 |
+
"eval_samples_per_second": 14.342,
|
| 93 |
+
"eval_steps_per_second": 0.903,
|
| 94 |
+
"eval_wer": 0.9999340082093787,
|
| 95 |
+
"step": 942
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 9.55,
|
| 99 |
+
"learning_rate": 4.393305439330544e-05,
|
| 100 |
+
"loss": 3.2743,
|
| 101 |
+
"step": 1000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 10.0,
|
| 105 |
+
"eval_loss": 2.1381497383117676,
|
| 106 |
+
"eval_runtime": 58.7525,
|
| 107 |
+
"eval_samples_per_second": 14.331,
|
| 108 |
+
"eval_steps_per_second": 0.902,
|
| 109 |
+
"eval_wer": 1.0245489461111037,
|
| 110 |
+
"step": 1047
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 11.0,
|
| 114 |
+
"eval_loss": 1.6965396404266357,
|
| 115 |
+
"eval_runtime": 58.7454,
|
| 116 |
+
"eval_samples_per_second": 14.333,
|
| 117 |
+
"eval_steps_per_second": 0.902,
|
| 118 |
+
"eval_wer": 1.0141750366254438,
|
| 119 |
+
"step": 1152
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 12.0,
|
| 123 |
+
"eval_loss": 1.4229621887207031,
|
| 124 |
+
"eval_runtime": 58.8247,
|
| 125 |
+
"eval_samples_per_second": 14.314,
|
| 126 |
+
"eval_steps_per_second": 0.901,
|
| 127 |
+
"eval_wer": 1.0011350587986854,
|
| 128 |
+
"step": 1257
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 12.99,
|
| 132 |
+
"eval_loss": 1.2678765058517456,
|
| 133 |
+
"eval_runtime": 58.7761,
|
| 134 |
+
"eval_samples_per_second": 14.326,
|
| 135 |
+
"eval_steps_per_second": 0.902,
|
| 136 |
+
"eval_wer": 0.9873295762007206,
|
| 137 |
+
"step": 1361
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 14.0,
|
| 141 |
+
"eval_loss": 1.1569700241088867,
|
| 142 |
+
"eval_runtime": 58.6837,
|
| 143 |
+
"eval_samples_per_second": 14.348,
|
| 144 |
+
"eval_steps_per_second": 0.903,
|
| 145 |
+
"eval_wer": 0.9835944408515581,
|
| 146 |
+
"step": 1466
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 14.32,
|
| 150 |
+
"learning_rate": 3.870292887029289e-05,
|
| 151 |
+
"loss": 1.5432,
|
| 152 |
+
"step": 1500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 15.0,
|
| 156 |
+
"eval_loss": 1.0858139991760254,
|
| 157 |
+
"eval_runtime": 58.71,
|
| 158 |
+
"eval_samples_per_second": 14.342,
|
| 159 |
+
"eval_steps_per_second": 0.903,
|
| 160 |
+
"eval_wer": 0.9784206844668524,
|
| 161 |
+
"step": 1571
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 16.0,
|
| 165 |
+
"eval_loss": 1.0303007364273071,
|
| 166 |
+
"eval_runtime": 58.7899,
|
| 167 |
+
"eval_samples_per_second": 14.322,
|
| 168 |
+
"eval_steps_per_second": 0.902,
|
| 169 |
+
"eval_wer": 0.9768632782081909,
|
| 170 |
+
"step": 1676
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 16.99,
|
| 174 |
+
"eval_loss": 0.985497236251831,
|
| 175 |
+
"eval_runtime": 58.9329,
|
| 176 |
+
"eval_samples_per_second": 14.287,
|
| 177 |
+
"eval_steps_per_second": 0.899,
|
| 178 |
+
"eval_wer": 0.97459316061082,
|
| 179 |
+
"step": 1780
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 18.0,
|
| 183 |
+
"eval_loss": 0.9558756947517395,
|
| 184 |
+
"eval_runtime": 58.7163,
|
| 185 |
+
"eval_samples_per_second": 14.34,
|
| 186 |
+
"eval_steps_per_second": 0.903,
|
| 187 |
+
"eval_wer": 0.9709240170522787,
|
| 188 |
+
"step": 1885
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 19.0,
|
| 192 |
+
"eval_loss": 0.93276047706604,
|
| 193 |
+
"eval_runtime": 58.6741,
|
| 194 |
+
"eval_samples_per_second": 14.35,
|
| 195 |
+
"eval_steps_per_second": 0.903,
|
| 196 |
+
"eval_wer": 0.9727981839059221,
|
| 197 |
+
"step": 1990
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 19.09,
|
| 201 |
+
"learning_rate": 3.3472803347280334e-05,
|
| 202 |
+
"loss": 0.902,
|
| 203 |
+
"step": 2000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 20.0,
|
| 207 |
+
"eval_loss": 0.916550874710083,
|
| 208 |
+
"eval_runtime": 58.8598,
|
| 209 |
+
"eval_samples_per_second": 14.305,
|
| 210 |
+
"eval_steps_per_second": 0.9,
|
| 211 |
+
"eval_wer": 0.9738144574814893,
|
| 212 |
+
"step": 2095
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 20.99,
|
| 216 |
+
"eval_loss": 0.8990672826766968,
|
| 217 |
+
"eval_runtime": 58.8033,
|
| 218 |
+
"eval_samples_per_second": 14.319,
|
| 219 |
+
"eval_steps_per_second": 0.901,
|
| 220 |
+
"eval_wer": 0.9697625615373447,
|
| 221 |
+
"step": 2199
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 22.0,
|
| 225 |
+
"eval_loss": 0.8716745376586914,
|
| 226 |
+
"eval_runtime": 58.795,
|
| 227 |
+
"eval_samples_per_second": 14.321,
|
| 228 |
+
"eval_steps_per_second": 0.901,
|
| 229 |
+
"eval_wer": 0.9680731716974409,
|
| 230 |
+
"step": 2304
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 23.0,
|
| 234 |
+
"eval_loss": 0.8665304780006409,
|
| 235 |
+
"eval_runtime": 58.9314,
|
| 236 |
+
"eval_samples_per_second": 14.288,
|
| 237 |
+
"eval_steps_per_second": 0.899,
|
| 238 |
+
"eval_wer": 0.9669249145406311,
|
| 239 |
+
"step": 2409
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 23.87,
|
| 243 |
+
"learning_rate": 2.824267782426778e-05,
|
| 244 |
+
"loss": 0.7003,
|
| 245 |
+
"step": 2500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 24.0,
|
| 249 |
+
"eval_loss": 0.8588829636573792,
|
| 250 |
+
"eval_runtime": 58.6586,
|
| 251 |
+
"eval_samples_per_second": 14.354,
|
| 252 |
+
"eval_steps_per_second": 0.904,
|
| 253 |
+
"eval_wer": 0.9670173030475009,
|
| 254 |
+
"step": 2514
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 24.99,
|
| 258 |
+
"eval_loss": 0.8419524431228638,
|
| 259 |
+
"eval_runtime": 58.7876,
|
| 260 |
+
"eval_samples_per_second": 14.323,
|
| 261 |
+
"eval_steps_per_second": 0.902,
|
| 262 |
+
"eval_wer": 0.9659086409650639,
|
| 263 |
+
"step": 2618
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 26.0,
|
| 267 |
+
"eval_loss": 0.8473154306411743,
|
| 268 |
+
"eval_runtime": 59.0243,
|
| 269 |
+
"eval_samples_per_second": 14.265,
|
| 270 |
+
"eval_steps_per_second": 0.898,
|
| 271 |
+
"eval_wer": 0.9660538229044306,
|
| 272 |
+
"step": 2723
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 27.0,
|
| 276 |
+
"eval_loss": 0.8542926907539368,
|
| 277 |
+
"eval_runtime": 58.7124,
|
| 278 |
+
"eval_samples_per_second": 14.341,
|
| 279 |
+
"eval_steps_per_second": 0.903,
|
| 280 |
+
"eval_wer": 0.9666345506618976,
|
| 281 |
+
"step": 2828
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 28.0,
|
| 285 |
+
"eval_loss": 0.8315206170082092,
|
| 286 |
+
"eval_runtime": 58.7261,
|
| 287 |
+
"eval_samples_per_second": 14.338,
|
| 288 |
+
"eval_steps_per_second": 0.902,
|
| 289 |
+
"eval_wer": 0.9622922908390196,
|
| 290 |
+
"step": 2933
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 28.64,
|
| 294 |
+
"learning_rate": 2.301255230125523e-05,
|
| 295 |
+
"loss": 0.5914,
|
| 296 |
+
"step": 3000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 28.99,
|
| 300 |
+
"eval_loss": 0.8280870914459229,
|
| 301 |
+
"eval_runtime": 58.9202,
|
| 302 |
+
"eval_samples_per_second": 14.291,
|
| 303 |
+
"eval_steps_per_second": 0.9,
|
| 304 |
+
"eval_wer": 0.9625694563596289,
|
| 305 |
+
"step": 3037
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 30.0,
|
| 309 |
+
"eval_loss": 0.8315181732177734,
|
| 310 |
+
"eval_runtime": 58.6719,
|
| 311 |
+
"eval_samples_per_second": 14.351,
|
| 312 |
+
"eval_steps_per_second": 0.903,
|
| 313 |
+
"eval_wer": 0.9625430596433804,
|
| 314 |
+
"step": 3142
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 31.0,
|
| 318 |
+
"eval_loss": 0.826146125793457,
|
| 319 |
+
"eval_runtime": 58.8509,
|
| 320 |
+
"eval_samples_per_second": 14.307,
|
| 321 |
+
"eval_steps_per_second": 0.901,
|
| 322 |
+
"eval_wer": 0.9619623318859134,
|
| 323 |
+
"step": 3247
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 32.0,
|
| 327 |
+
"eval_loss": 0.8213908076286316,
|
| 328 |
+
"eval_runtime": 58.6666,
|
| 329 |
+
"eval_samples_per_second": 14.352,
|
| 330 |
+
"eval_steps_per_second": 0.903,
|
| 331 |
+
"eval_wer": 0.9639684823207993,
|
| 332 |
+
"step": 3352
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 32.99,
|
| 336 |
+
"eval_loss": 0.8309536576271057,
|
| 337 |
+
"eval_runtime": 58.499,
|
| 338 |
+
"eval_samples_per_second": 14.393,
|
| 339 |
+
"eval_steps_per_second": 0.906,
|
| 340 |
+
"eval_wer": 0.9633745562052081,
|
| 341 |
+
"step": 3456
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 33.41,
|
| 345 |
+
"learning_rate": 1.7782426778242676e-05,
|
| 346 |
+
"loss": 0.5157,
|
| 347 |
+
"step": 3500
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 34.0,
|
| 351 |
+
"eval_loss": 0.8252176642417908,
|
| 352 |
+
"eval_runtime": 58.5446,
|
| 353 |
+
"eval_samples_per_second": 14.382,
|
| 354 |
+
"eval_steps_per_second": 0.905,
|
| 355 |
+
"eval_wer": 0.9634669447120778,
|
| 356 |
+
"step": 3561
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 35.0,
|
| 360 |
+
"eval_loss": 0.8372853398323059,
|
| 361 |
+
"eval_runtime": 58.5403,
|
| 362 |
+
"eval_samples_per_second": 14.383,
|
| 363 |
+
"eval_steps_per_second": 0.905,
|
| 364 |
+
"eval_wer": 0.9638101020233083,
|
| 365 |
+
"step": 3666
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 36.0,
|
| 369 |
+
"eval_loss": 0.8421617746353149,
|
| 370 |
+
"eval_runtime": 58.6533,
|
| 371 |
+
"eval_samples_per_second": 14.356,
|
| 372 |
+
"eval_steps_per_second": 0.904,
|
| 373 |
+
"eval_wer": 0.9629258120289836,
|
| 374 |
+
"step": 3771
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 36.99,
|
| 378 |
+
"eval_loss": 0.8294224143028259,
|
| 379 |
+
"eval_runtime": 58.6527,
|
| 380 |
+
"eval_samples_per_second": 14.356,
|
| 381 |
+
"eval_steps_per_second": 0.904,
|
| 382 |
+
"eval_wer": 0.9631897791914685,
|
| 383 |
+
"step": 3875
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 38.0,
|
| 387 |
+
"eval_loss": 0.8332000374794006,
|
| 388 |
+
"eval_runtime": 58.5856,
|
| 389 |
+
"eval_samples_per_second": 14.372,
|
| 390 |
+
"eval_steps_per_second": 0.905,
|
| 391 |
+
"eval_wer": 0.9576332704211596,
|
| 392 |
+
"step": 3980
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 38.19,
|
| 396 |
+
"learning_rate": 1.2552301255230125e-05,
|
| 397 |
+
"loss": 0.4655,
|
| 398 |
+
"step": 4000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 39.0,
|
| 402 |
+
"eval_loss": 0.8329895734786987,
|
| 403 |
+
"eval_runtime": 58.7603,
|
| 404 |
+
"eval_samples_per_second": 14.329,
|
| 405 |
+
"eval_steps_per_second": 0.902,
|
| 406 |
+
"eval_wer": 0.959507437274803,
|
| 407 |
+
"step": 4085
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 40.0,
|
| 411 |
+
"eval_loss": 0.8296785950660706,
|
| 412 |
+
"eval_runtime": 58.6153,
|
| 413 |
+
"eval_samples_per_second": 14.365,
|
| 414 |
+
"eval_steps_per_second": 0.904,
|
| 415 |
+
"eval_wer": 0.9625034645690076,
|
| 416 |
+
"step": 4190
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 40.99,
|
| 420 |
+
"eval_loss": 0.8365075588226318,
|
| 421 |
+
"eval_runtime": 58.6008,
|
| 422 |
+
"eval_samples_per_second": 14.368,
|
| 423 |
+
"eval_steps_per_second": 0.904,
|
| 424 |
+
"eval_wer": 0.9620547203927832,
|
| 425 |
+
"step": 4294
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 42.0,
|
| 429 |
+
"eval_loss": 0.8360996842384338,
|
| 430 |
+
"eval_runtime": 58.6605,
|
| 431 |
+
"eval_samples_per_second": 14.354,
|
| 432 |
+
"eval_steps_per_second": 0.904,
|
| 433 |
+
"eval_wer": 0.9621339105415286,
|
| 434 |
+
"step": 4399
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 42.96,
|
| 438 |
+
"learning_rate": 7.3221757322175736e-06,
|
| 439 |
+
"loss": 0.4266,
|
| 440 |
+
"step": 4500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 43.0,
|
| 444 |
+
"eval_loss": 0.8416324853897095,
|
| 445 |
+
"eval_runtime": 58.5672,
|
| 446 |
+
"eval_samples_per_second": 14.377,
|
| 447 |
+
"eval_steps_per_second": 0.905,
|
| 448 |
+
"eval_wer": 0.9625034645690076,
|
| 449 |
+
"step": 4504
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 44.0,
|
| 453 |
+
"eval_loss": 0.8381497859954834,
|
| 454 |
+
"eval_runtime": 58.6851,
|
| 455 |
+
"eval_samples_per_second": 14.348,
|
| 456 |
+
"eval_steps_per_second": 0.903,
|
| 457 |
+
"eval_wer": 0.9633745562052081,
|
| 458 |
+
"step": 4609
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 44.99,
|
| 462 |
+
"eval_loss": 0.8447730541229248,
|
| 463 |
+
"eval_runtime": 58.6902,
|
| 464 |
+
"eval_samples_per_second": 14.347,
|
| 465 |
+
"eval_steps_per_second": 0.903,
|
| 466 |
+
"eval_wer": 0.9644964166457692,
|
| 467 |
+
"step": 4713
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 46.0,
|
| 471 |
+
"eval_loss": 0.8447101712226868,
|
| 472 |
+
"eval_runtime": 58.8118,
|
| 473 |
+
"eval_samples_per_second": 14.317,
|
| 474 |
+
"eval_steps_per_second": 0.901,
|
| 475 |
+
"eval_wer": 0.9624902662108834,
|
| 476 |
+
"step": 4818
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 47.0,
|
| 480 |
+
"eval_loss": 0.8463994264602661,
|
| 481 |
+
"eval_runtime": 58.6459,
|
| 482 |
+
"eval_samples_per_second": 14.357,
|
| 483 |
+
"eval_steps_per_second": 0.904,
|
| 484 |
+
"eval_wer": 0.9641268626182903,
|
| 485 |
+
"step": 4923
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 47.73,
|
| 489 |
+
"learning_rate": 2.092050209205021e-06,
|
| 490 |
+
"loss": 0.4019,
|
| 491 |
+
"step": 5000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 48.0,
|
| 495 |
+
"eval_loss": 0.8449317216873169,
|
| 496 |
+
"eval_runtime": 58.893,
|
| 497 |
+
"eval_samples_per_second": 14.297,
|
| 498 |
+
"eval_steps_per_second": 0.9,
|
| 499 |
+
"eval_wer": 0.9627542333733684,
|
| 500 |
+
"step": 5028
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 48.99,
|
| 504 |
+
"eval_loss": 0.8487057089805603,
|
| 505 |
+
"eval_runtime": 58.7287,
|
| 506 |
+
"eval_samples_per_second": 14.337,
|
| 507 |
+
"eval_steps_per_second": 0.902,
|
| 508 |
+
"eval_wer": 0.9625562580015046,
|
| 509 |
+
"step": 5132
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 49.64,
|
| 513 |
+
"eval_loss": 0.846527636051178,
|
| 514 |
+
"eval_runtime": 58.7826,
|
| 515 |
+
"eval_samples_per_second": 14.324,
|
| 516 |
+
"eval_steps_per_second": 0.902,
|
| 517 |
+
"eval_wer": 0.9629126136708593,
|
| 518 |
+
"step": 5200
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 49.64,
|
| 522 |
+
"step": 5200,
|
| 523 |
+
"total_flos": 4.180377449644987e+19,
|
| 524 |
+
"train_loss": 1.3979950288625864,
|
| 525 |
+
"train_runtime": 89308.2068,
|
| 526 |
+
"train_samples_per_second": 1.876,
|
| 527 |
+
"train_steps_per_second": 0.058
|
| 528 |
+
}
|
| 529 |
+
],
|
| 530 |
+
"max_steps": 5200,
|
| 531 |
+
"num_train_epochs": 50,
|
| 532 |
+
"total_flos": 4.180377449644987e+19,
|
| 533 |
+
"trial_name": null,
|
| 534 |
+
"trial_params": null
|
| 535 |
+
}
|