End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_cer": 0.0058143100249896515,
|
| 4 |
+
"eval_loss": 0.06832413375377655,
|
| 5 |
+
"eval_runtime": 87.0708,
|
| 6 |
+
"eval_samples": 1965,
|
| 7 |
+
"eval_samples_per_second": 22.568,
|
| 8 |
+
"eval_steps_per_second": 2.825,
|
| 9 |
+
"eval_wer": 0.009874807524938073,
|
| 10 |
+
"train_loss": 0.15939651108792915,
|
| 11 |
+
"train_runtime": 19958.548,
|
| 12 |
+
"train_samples": 11135,
|
| 13 |
+
"train_samples_per_second": 16.737,
|
| 14 |
+
"train_steps_per_second": 0.523
|
| 15 |
+
}
|
runs/Jan09_00-52-51_bookbot-pt-2/events.out.tfevents.1673247932.bookbot-pt-2.11380.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34526f92343e7f064272517b81b1d32ffc5ab2b0ff0918f8c85980959b85b030
|
| 3 |
+
size 405
|
test_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_cer": 0.0058143100249896515,
|
| 4 |
+
"eval_loss": 0.06832413375377655,
|
| 5 |
+
"eval_runtime": 87.0708,
|
| 6 |
+
"eval_samples": 1965,
|
| 7 |
+
"eval_samples_per_second": 22.568,
|
| 8 |
+
"eval_steps_per_second": 2.825,
|
| 9 |
+
"eval_wer": 0.009874807524938073
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"train_loss": 0.15939651108792915,
|
| 4 |
+
"train_runtime": 19958.548,
|
| 5 |
+
"train_samples": 11135,
|
| 6 |
+
"train_samples_per_second": 16.737,
|
| 7 |
+
"train_steps_per_second": 0.523
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,445 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"global_step": 10440,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"eval_cer": 1.0,
|
| 13 |
+
"eval_loss": 2.2818498611450195,
|
| 14 |
+
"eval_runtime": 85.7153,
|
| 15 |
+
"eval_samples_per_second": 22.925,
|
| 16 |
+
"eval_steps_per_second": 2.87,
|
| 17 |
+
"eval_wer": 1.0,
|
| 18 |
+
"step": 348
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.44,
|
| 22 |
+
"learning_rate": 5e-05,
|
| 23 |
+
"loss": 2.6692,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.0,
|
| 28 |
+
"eval_cer": 0.029949254143222897,
|
| 29 |
+
"eval_loss": 0.2044876664876938,
|
| 30 |
+
"eval_runtime": 85.7259,
|
| 31 |
+
"eval_samples_per_second": 22.922,
|
| 32 |
+
"eval_steps_per_second": 2.87,
|
| 33 |
+
"eval_wer": 0.052662850639351944,
|
| 34 |
+
"step": 696
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.87,
|
| 38 |
+
"learning_rate": 0.0001,
|
| 39 |
+
"loss": 0.2225,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 3.0,
|
| 44 |
+
"eval_cer": 0.018884051083140417,
|
| 45 |
+
"eval_loss": 0.11616221815347672,
|
| 46 |
+
"eval_runtime": 85.4863,
|
| 47 |
+
"eval_samples_per_second": 22.986,
|
| 48 |
+
"eval_steps_per_second": 2.878,
|
| 49 |
+
"eval_wer": 0.031858806989355296,
|
| 50 |
+
"step": 1044
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_cer": 0.014687169423704908,
|
| 55 |
+
"eval_loss": 0.09268919378519058,
|
| 56 |
+
"eval_runtime": 85.2634,
|
| 57 |
+
"eval_samples_per_second": 23.046,
|
| 58 |
+
"eval_steps_per_second": 2.885,
|
| 59 |
+
"eval_wer": 0.023540536921737965,
|
| 60 |
+
"step": 1392
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.31,
|
| 64 |
+
"learning_rate": 9.470338983050848e-05,
|
| 65 |
+
"loss": 0.0868,
|
| 66 |
+
"step": 1500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 5.0,
|
| 70 |
+
"eval_cer": 0.014284728716635749,
|
| 71 |
+
"eval_loss": 0.07971413433551788,
|
| 72 |
+
"eval_runtime": 85.4679,
|
| 73 |
+
"eval_samples_per_second": 22.991,
|
| 74 |
+
"eval_steps_per_second": 2.878,
|
| 75 |
+
"eval_wer": 0.021841735288210484,
|
| 76 |
+
"step": 1740
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 5.75,
|
| 80 |
+
"learning_rate": 8.940677966101694e-05,
|
| 81 |
+
"loss": 0.0598,
|
| 82 |
+
"step": 2000
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 6.0,
|
| 86 |
+
"eval_cer": 0.012847440477103041,
|
| 87 |
+
"eval_loss": 0.07152710855007172,
|
| 88 |
+
"eval_runtime": 85.5884,
|
| 89 |
+
"eval_samples_per_second": 22.959,
|
| 90 |
+
"eval_steps_per_second": 2.874,
|
| 91 |
+
"eval_wer": 0.01967429872129611,
|
| 92 |
+
"step": 2088
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 7.0,
|
| 96 |
+
"eval_cer": 0.010252656108666656,
|
| 97 |
+
"eval_loss": 0.06518065184354782,
|
| 98 |
+
"eval_runtime": 85.121,
|
| 99 |
+
"eval_samples_per_second": 23.085,
|
| 100 |
+
"eval_steps_per_second": 2.89,
|
| 101 |
+
"eval_wer": 0.015975430139921,
|
| 102 |
+
"step": 2436
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 7.18,
|
| 106 |
+
"learning_rate": 8.411016949152542e-05,
|
| 107 |
+
"loss": 0.0447,
|
| 108 |
+
"step": 2500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 8.0,
|
| 112 |
+
"eval_cer": 0.00946693853772211,
|
| 113 |
+
"eval_loss": 0.057057663798332214,
|
| 114 |
+
"eval_runtime": 84.3173,
|
| 115 |
+
"eval_samples_per_second": 23.305,
|
| 116 |
+
"eval_steps_per_second": 2.918,
|
| 117 |
+
"eval_wer": 0.015230635335073977,
|
| 118 |
+
"step": 2784
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 8.62,
|
| 122 |
+
"learning_rate": 7.88135593220339e-05,
|
| 123 |
+
"loss": 0.0368,
|
| 124 |
+
"step": 3000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 9.0,
|
| 128 |
+
"eval_cer": 0.011180186119245098,
|
| 129 |
+
"eval_loss": 0.060811206698417664,
|
| 130 |
+
"eval_runtime": 84.3248,
|
| 131 |
+
"eval_samples_per_second": 23.303,
|
| 132 |
+
"eval_steps_per_second": 2.917,
|
| 133 |
+
"eval_wer": 0.01630180089710116,
|
| 134 |
+
"step": 3132
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.0,
|
| 138 |
+
"eval_cer": 0.008297944102902173,
|
| 139 |
+
"eval_loss": 0.058583296835422516,
|
| 140 |
+
"eval_runtime": 84.6721,
|
| 141 |
+
"eval_samples_per_second": 23.207,
|
| 142 |
+
"eval_steps_per_second": 2.905,
|
| 143 |
+
"eval_wer": 0.013657360915846555,
|
| 144 |
+
"step": 3480
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 10.06,
|
| 148 |
+
"learning_rate": 7.351694915254238e-05,
|
| 149 |
+
"loss": 0.0303,
|
| 150 |
+
"step": 3500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 11.0,
|
| 154 |
+
"eval_cer": 0.008535575758504913,
|
| 155 |
+
"eval_loss": 0.06412886828184128,
|
| 156 |
+
"eval_runtime": 87.3267,
|
| 157 |
+
"eval_samples_per_second": 22.502,
|
| 158 |
+
"eval_steps_per_second": 2.817,
|
| 159 |
+
"eval_wer": 0.014125995849233446,
|
| 160 |
+
"step": 3828
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 11.49,
|
| 164 |
+
"learning_rate": 6.822033898305085e-05,
|
| 165 |
+
"loss": 0.0273,
|
| 166 |
+
"step": 4000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 12.0,
|
| 170 |
+
"eval_cer": 0.007933831082220552,
|
| 171 |
+
"eval_loss": 0.06564020365476608,
|
| 172 |
+
"eval_runtime": 84.2667,
|
| 173 |
+
"eval_samples_per_second": 23.319,
|
| 174 |
+
"eval_steps_per_second": 2.919,
|
| 175 |
+
"eval_wer": 0.013071567249112941,
|
| 176 |
+
"step": 4176
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 12.93,
|
| 180 |
+
"learning_rate": 6.29343220338983e-05,
|
| 181 |
+
"loss": 0.0232,
|
| 182 |
+
"step": 4500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.0,
|
| 186 |
+
"eval_cer": 0.008225121498765848,
|
| 187 |
+
"eval_loss": 0.06898853182792664,
|
| 188 |
+
"eval_runtime": 84.1359,
|
| 189 |
+
"eval_samples_per_second": 23.355,
|
| 190 |
+
"eval_steps_per_second": 2.924,
|
| 191 |
+
"eval_wer": 0.0132640423110397,
|
| 192 |
+
"step": 4524
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 14.0,
|
| 196 |
+
"eval_cer": 0.00787250678400049,
|
| 197 |
+
"eval_loss": 0.05983823910355568,
|
| 198 |
+
"eval_runtime": 84.1873,
|
| 199 |
+
"eval_samples_per_second": 23.341,
|
| 200 |
+
"eval_steps_per_second": 2.922,
|
| 201 |
+
"eval_wer": 0.012803775858606146,
|
| 202 |
+
"step": 4872
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 14.37,
|
| 206 |
+
"learning_rate": 5.763771186440679e-05,
|
| 207 |
+
"loss": 0.0189,
|
| 208 |
+
"step": 5000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 15.0,
|
| 212 |
+
"eval_cer": 0.007420240084627531,
|
| 213 |
+
"eval_loss": 0.06711488217115402,
|
| 214 |
+
"eval_runtime": 84.8039,
|
| 215 |
+
"eval_samples_per_second": 23.171,
|
| 216 |
+
"eval_steps_per_second": 2.901,
|
| 217 |
+
"eval_wer": 0.012100823458525808,
|
| 218 |
+
"step": 5220
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 15.8,
|
| 222 |
+
"learning_rate": 5.2341101694915265e-05,
|
| 223 |
+
"loss": 0.017,
|
| 224 |
+
"step": 5500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 16.0,
|
| 228 |
+
"eval_cer": 0.006906649087034511,
|
| 229 |
+
"eval_loss": 0.06541039049625397,
|
| 230 |
+
"eval_runtime": 84.2563,
|
| 231 |
+
"eval_samples_per_second": 23.322,
|
| 232 |
+
"eval_steps_per_second": 2.92,
|
| 233 |
+
"eval_wer": 0.011364397134632121,
|
| 234 |
+
"step": 5568
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 17.0,
|
| 238 |
+
"eval_cer": 0.007335919174574946,
|
| 239 |
+
"eval_loss": 0.07511687278747559,
|
| 240 |
+
"eval_runtime": 84.0992,
|
| 241 |
+
"eval_samples_per_second": 23.365,
|
| 242 |
+
"eval_steps_per_second": 2.925,
|
| 243 |
+
"eval_wer": 0.011807926625159,
|
| 244 |
+
"step": 5916
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 17.24,
|
| 248 |
+
"learning_rate": 4.705508474576271e-05,
|
| 249 |
+
"loss": 0.0146,
|
| 250 |
+
"step": 6000
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 18.0,
|
| 254 |
+
"eval_cer": 0.006753338341484355,
|
| 255 |
+
"eval_loss": 0.06527850776910782,
|
| 256 |
+
"eval_runtime": 83.7988,
|
| 257 |
+
"eval_samples_per_second": 23.449,
|
| 258 |
+
"eval_steps_per_second": 2.936,
|
| 259 |
+
"eval_wer": 0.011171922072705363,
|
| 260 |
+
"step": 6264
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 18.68,
|
| 264 |
+
"learning_rate": 4.175847457627119e-05,
|
| 265 |
+
"loss": 0.0127,
|
| 266 |
+
"step": 6500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 19.0,
|
| 270 |
+
"eval_cer": 0.006921980161589526,
|
| 271 |
+
"eval_loss": 0.06817645579576492,
|
| 272 |
+
"eval_runtime": 84.0515,
|
| 273 |
+
"eval_samples_per_second": 23.379,
|
| 274 |
+
"eval_steps_per_second": 2.927,
|
| 275 |
+
"eval_wer": 0.01123886992033206,
|
| 276 |
+
"step": 6612
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 20.0,
|
| 280 |
+
"eval_cer": 0.006814662639704417,
|
| 281 |
+
"eval_loss": 0.06784532964229584,
|
| 282 |
+
"eval_runtime": 83.9653,
|
| 283 |
+
"eval_samples_per_second": 23.403,
|
| 284 |
+
"eval_steps_per_second": 2.93,
|
| 285 |
+
"eval_wer": 0.01137276561558546,
|
| 286 |
+
"step": 6960
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 20.11,
|
| 290 |
+
"learning_rate": 3.6461864406779664e-05,
|
| 291 |
+
"loss": 0.0114,
|
| 292 |
+
"step": 7000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 21.0,
|
| 296 |
+
"eval_cer": 0.006584696521379184,
|
| 297 |
+
"eval_loss": 0.06555593758821487,
|
| 298 |
+
"eval_runtime": 83.8204,
|
| 299 |
+
"eval_samples_per_second": 23.443,
|
| 300 |
+
"eval_steps_per_second": 2.935,
|
| 301 |
+
"eval_wer": 0.011113342706032,
|
| 302 |
+
"step": 7308
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 21.55,
|
| 306 |
+
"learning_rate": 3.117584745762712e-05,
|
| 307 |
+
"loss": 0.0101,
|
| 308 |
+
"step": 7500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 22.0,
|
| 312 |
+
"eval_cer": 0.006596194827295445,
|
| 313 |
+
"eval_loss": 0.06685744225978851,
|
| 314 |
+
"eval_runtime": 84.0101,
|
| 315 |
+
"eval_samples_per_second": 23.39,
|
| 316 |
+
"eval_steps_per_second": 2.928,
|
| 317 |
+
"eval_wer": 0.010920867644105242,
|
| 318 |
+
"step": 7656
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 22.99,
|
| 322 |
+
"learning_rate": 2.5879237288135593e-05,
|
| 323 |
+
"loss": 0.0092,
|
| 324 |
+
"step": 8000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 23.0,
|
| 328 |
+
"eval_cer": 0.006477378999494075,
|
| 329 |
+
"eval_loss": 0.06765928864479065,
|
| 330 |
+
"eval_runtime": 84.2885,
|
| 331 |
+
"eval_samples_per_second": 23.313,
|
| 332 |
+
"eval_steps_per_second": 2.919,
|
| 333 |
+
"eval_wer": 0.010778603467898508,
|
| 334 |
+
"step": 8004
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 24.0,
|
| 338 |
+
"eval_cer": 0.006331733791221427,
|
| 339 |
+
"eval_loss": 0.0652570053935051,
|
| 340 |
+
"eval_runtime": 84.1568,
|
| 341 |
+
"eval_samples_per_second": 23.349,
|
| 342 |
+
"eval_steps_per_second": 2.923,
|
| 343 |
+
"eval_wer": 0.010402021824998326,
|
| 344 |
+
"step": 8352
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 24.43,
|
| 348 |
+
"learning_rate": 2.058262711864407e-05,
|
| 349 |
+
"loss": 0.0088,
|
| 350 |
+
"step": 8500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 25.0,
|
| 354 |
+
"eval_cer": 0.006266576724362611,
|
| 355 |
+
"eval_loss": 0.0673212930560112,
|
| 356 |
+
"eval_runtime": 83.9435,
|
| 357 |
+
"eval_samples_per_second": 23.409,
|
| 358 |
+
"eval_steps_per_second": 2.931,
|
| 359 |
+
"eval_wer": 0.01020117828211823,
|
| 360 |
+
"step": 8700
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 25.86,
|
| 364 |
+
"learning_rate": 1.5286016949152543e-05,
|
| 365 |
+
"loss": 0.0074,
|
| 366 |
+
"step": 9000
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 26.0,
|
| 370 |
+
"eval_cer": 0.006350897634415196,
|
| 371 |
+
"eval_loss": 0.06691750884056091,
|
| 372 |
+
"eval_runtime": 84.013,
|
| 373 |
+
"eval_samples_per_second": 23.389,
|
| 374 |
+
"eval_steps_per_second": 2.928,
|
| 375 |
+
"eval_wer": 0.0104857066345317,
|
| 376 |
+
"step": 9048
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 27.0,
|
| 380 |
+
"eval_cer": 0.006113265978812455,
|
| 381 |
+
"eval_loss": 0.0707407295703888,
|
| 382 |
+
"eval_runtime": 84.4435,
|
| 383 |
+
"eval_samples_per_second": 23.27,
|
| 384 |
+
"eval_steps_per_second": 2.913,
|
| 385 |
+
"eval_wer": 0.01013423043449153,
|
| 386 |
+
"step": 9396
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 27.3,
|
| 390 |
+
"learning_rate": 9.989406779661017e-06,
|
| 391 |
+
"loss": 0.0066,
|
| 392 |
+
"step": 9500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 28.0,
|
| 396 |
+
"eval_cer": 0.0059829518450948225,
|
| 397 |
+
"eval_loss": 0.06726762652397156,
|
| 398 |
+
"eval_runtime": 84.279,
|
| 399 |
+
"eval_samples_per_second": 23.315,
|
| 400 |
+
"eval_steps_per_second": 2.919,
|
| 401 |
+
"eval_wer": 0.009966860815424784,
|
| 402 |
+
"step": 9744
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 28.74,
|
| 406 |
+
"learning_rate": 4.692796610169492e-06,
|
| 407 |
+
"loss": 0.0058,
|
| 408 |
+
"step": 10000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 29.0,
|
| 412 |
+
"eval_cer": 0.005867968785932206,
|
| 413 |
+
"eval_loss": 0.06885003298521042,
|
| 414 |
+
"eval_runtime": 84.0405,
|
| 415 |
+
"eval_samples_per_second": 23.382,
|
| 416 |
+
"eval_steps_per_second": 2.927,
|
| 417 |
+
"eval_wer": 0.010000334739238134,
|
| 418 |
+
"step": 10092
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 30.0,
|
| 422 |
+
"eval_cer": 0.0058143100249896515,
|
| 423 |
+
"eval_loss": 0.06832413375377655,
|
| 424 |
+
"eval_runtime": 84.1565,
|
| 425 |
+
"eval_samples_per_second": 23.349,
|
| 426 |
+
"eval_steps_per_second": 2.923,
|
| 427 |
+
"eval_wer": 0.009874807524938073,
|
| 428 |
+
"step": 10440
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 30.0,
|
| 432 |
+
"step": 10440,
|
| 433 |
+
"total_flos": 2.004174730615405e+19,
|
| 434 |
+
"train_loss": 0.15939651108792915,
|
| 435 |
+
"train_runtime": 19958.548,
|
| 436 |
+
"train_samples_per_second": 16.737,
|
| 437 |
+
"train_steps_per_second": 0.523
|
| 438 |
+
}
|
| 439 |
+
],
|
| 440 |
+
"max_steps": 10440,
|
| 441 |
+
"num_train_epochs": 30,
|
| 442 |
+
"total_flos": 2.004174730615405e+19,
|
| 443 |
+
"trial_name": null,
|
| 444 |
+
"trial_params": null
|
| 445 |
+
}
|