Commit
·
1b1c786
1
Parent(s):
e5b773d
End of training
Browse files- all_results.json +8 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +329 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"total_flos": 5.018412278085267e+19,
|
| 4 |
+
"train_loss": 0.24150397533025497,
|
| 5 |
+
"train_runtime": 25397.5173,
|
| 6 |
+
"train_samples_per_second": 18.876,
|
| 7 |
+
"train_steps_per_second": 0.295
|
| 8 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262227057
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d8ba89b74d9af18fd22ea5414a67a0ad277aad9f2dc5ea50943c38d22fc983c
|
| 3 |
size 1262227057
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"total_flos": 5.018412278085267e+19,
|
| 4 |
+
"train_loss": 0.24150397533025497,
|
| 5 |
+
"train_runtime": 25397.5173,
|
| 6 |
+
"train_samples_per_second": 18.876,
|
| 7 |
+
"train_steps_per_second": 0.295
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,329 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 11.999599519423308,
|
| 5 |
+
"global_step": 7488,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.48,
|
| 12 |
+
"learning_rate": 9.836092715231788e-05,
|
| 13 |
+
"loss": 7.1097,
|
| 14 |
+
"step": 300
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.48,
|
| 18 |
+
"eval_cer": 0.9965361196010141,
|
| 19 |
+
"eval_loss": 3.3989405632019043,
|
| 20 |
+
"eval_runtime": 296.7149,
|
| 21 |
+
"eval_samples_per_second": 12.952,
|
| 22 |
+
"eval_steps_per_second": 1.621,
|
| 23 |
+
"eval_wer": 1.0,
|
| 24 |
+
"step": 300
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.96,
|
| 28 |
+
"learning_rate": 9.339403973509933e-05,
|
| 29 |
+
"loss": 3.0235,
|
| 30 |
+
"step": 600
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.96,
|
| 34 |
+
"eval_cer": 0.31626033596322883,
|
| 35 |
+
"eval_loss": 1.3183486461639404,
|
| 36 |
+
"eval_runtime": 296.8438,
|
| 37 |
+
"eval_samples_per_second": 12.946,
|
| 38 |
+
"eval_steps_per_second": 1.62,
|
| 39 |
+
"eval_wer": 0.7976831955321887,
|
| 40 |
+
"step": 600
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.44,
|
| 44 |
+
"learning_rate": 8.84271523178808e-05,
|
| 45 |
+
"loss": 1.1419,
|
| 46 |
+
"step": 900
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.44,
|
| 50 |
+
"eval_cer": 0.19131423697490937,
|
| 51 |
+
"eval_loss": 0.6415870785713196,
|
| 52 |
+
"eval_runtime": 297.4006,
|
| 53 |
+
"eval_samples_per_second": 12.922,
|
| 54 |
+
"eval_steps_per_second": 1.617,
|
| 55 |
+
"eval_wer": 0.5543314644734832,
|
| 56 |
+
"step": 900
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.92,
|
| 60 |
+
"learning_rate": 8.346026490066225e-05,
|
| 61 |
+
"loss": 0.8242,
|
| 62 |
+
"step": 1200
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.92,
|
| 66 |
+
"eval_cer": 0.16081218755182788,
|
| 67 |
+
"eval_loss": 0.5063228011131287,
|
| 68 |
+
"eval_runtime": 296.5772,
|
| 69 |
+
"eval_samples_per_second": 12.958,
|
| 70 |
+
"eval_steps_per_second": 1.622,
|
| 71 |
+
"eval_wer": 0.4804206025436855,
|
| 72 |
+
"step": 1200
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.56,
|
| 76 |
+
"learning_rate": 7.683774834437087e-05,
|
| 77 |
+
"loss": 0.6876,
|
| 78 |
+
"step": 1600
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.56,
|
| 82 |
+
"eval_cer": 0.1387068495747151,
|
| 83 |
+
"eval_loss": 0.44008567929267883,
|
| 84 |
+
"eval_runtime": 296.3457,
|
| 85 |
+
"eval_samples_per_second": 12.968,
|
| 86 |
+
"eval_steps_per_second": 1.623,
|
| 87 |
+
"eval_wer": 0.42799799297541397,
|
| 88 |
+
"step": 1600
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 3.21,
|
| 92 |
+
"learning_rate": 7.021523178807947e-05,
|
| 93 |
+
"loss": 0.5868,
|
| 94 |
+
"step": 2000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 3.21,
|
| 98 |
+
"eval_cer": 0.12487975928163575,
|
| 99 |
+
"eval_loss": 0.39397352933883667,
|
| 100 |
+
"eval_runtime": 295.9984,
|
| 101 |
+
"eval_samples_per_second": 12.983,
|
| 102 |
+
"eval_steps_per_second": 1.625,
|
| 103 |
+
"eval_wer": 0.3906716987717882,
|
| 104 |
+
"step": 2000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.85,
|
| 108 |
+
"learning_rate": 6.359271523178809e-05,
|
| 109 |
+
"loss": 0.5285,
|
| 110 |
+
"step": 2400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.85,
|
| 114 |
+
"eval_cer": 0.1200464377946786,
|
| 115 |
+
"eval_loss": 0.36609867215156555,
|
| 116 |
+
"eval_runtime": 296.3002,
|
| 117 |
+
"eval_samples_per_second": 12.97,
|
| 118 |
+
"eval_steps_per_second": 1.623,
|
| 119 |
+
"eval_wer": 0.3762952943999651,
|
| 120 |
+
"step": 2400
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 4.49,
|
| 124 |
+
"learning_rate": 6.433863885839737e-05,
|
| 125 |
+
"loss": 0.5,
|
| 126 |
+
"step": 2800
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 4.49,
|
| 130 |
+
"eval_cer": 0.11358305494349279,
|
| 131 |
+
"eval_loss": 0.35275039076805115,
|
| 132 |
+
"eval_runtime": 294.0242,
|
| 133 |
+
"eval_samples_per_second": 13.07,
|
| 134 |
+
"eval_steps_per_second": 1.636,
|
| 135 |
+
"eval_wer": 0.36098082418900934,
|
| 136 |
+
"step": 2800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 5.13,
|
| 140 |
+
"learning_rate": 5.885016465422613e-05,
|
| 141 |
+
"loss": 0.4538,
|
| 142 |
+
"step": 3200
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 5.13,
|
| 146 |
+
"eval_cer": 0.10862653114412302,
|
| 147 |
+
"eval_loss": 0.34030598402023315,
|
| 148 |
+
"eval_runtime": 295.0902,
|
| 149 |
+
"eval_samples_per_second": 13.023,
|
| 150 |
+
"eval_steps_per_second": 1.63,
|
| 151 |
+
"eval_wer": 0.34854599794934443,
|
| 152 |
+
"step": 3200
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.77,
|
| 156 |
+
"learning_rate": 5.336169045005489e-05,
|
| 157 |
+
"loss": 0.4165,
|
| 158 |
+
"step": 3600
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 5.77,
|
| 162 |
+
"eval_cer": 0.1061861776482574,
|
| 163 |
+
"eval_loss": 0.33348363637924194,
|
| 164 |
+
"eval_runtime": 294.4649,
|
| 165 |
+
"eval_samples_per_second": 13.051,
|
| 166 |
+
"eval_steps_per_second": 1.633,
|
| 167 |
+
"eval_wer": 0.3438774842383124,
|
| 168 |
+
"step": 3600
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.41,
|
| 172 |
+
"learning_rate": 4.787321624588365e-05,
|
| 173 |
+
"loss": 0.3989,
|
| 174 |
+
"step": 4000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 6.41,
|
| 178 |
+
"eval_cer": 0.1035941905371147,
|
| 179 |
+
"eval_loss": 0.3264155387878418,
|
| 180 |
+
"eval_runtime": 294.7693,
|
| 181 |
+
"eval_samples_per_second": 13.037,
|
| 182 |
+
"eval_steps_per_second": 1.632,
|
| 183 |
+
"eval_wer": 0.3339950697004734,
|
| 184 |
+
"step": 4000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 7.05,
|
| 188 |
+
"learning_rate": 4.2384742041712404e-05,
|
| 189 |
+
"loss": 0.3679,
|
| 190 |
+
"step": 4400
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 7.05,
|
| 194 |
+
"eval_cer": 0.10127703935366171,
|
| 195 |
+
"eval_loss": 0.3256165385246277,
|
| 196 |
+
"eval_runtime": 293.9503,
|
| 197 |
+
"eval_samples_per_second": 13.074,
|
| 198 |
+
"eval_steps_per_second": 1.636,
|
| 199 |
+
"eval_wer": 0.3287375379044045,
|
| 200 |
+
"step": 4400
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 7.69,
|
| 204 |
+
"learning_rate": 3.689626783754117e-05,
|
| 205 |
+
"loss": 0.3517,
|
| 206 |
+
"step": 4800
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 7.69,
|
| 210 |
+
"eval_cer": 0.10020612694576729,
|
| 211 |
+
"eval_loss": 0.3212122619152069,
|
| 212 |
+
"eval_runtime": 293.3063,
|
| 213 |
+
"eval_samples_per_second": 13.102,
|
| 214 |
+
"eval_steps_per_second": 1.64,
|
| 215 |
+
"eval_wer": 0.32230196993826216,
|
| 216 |
+
"step": 4800
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 8.33,
|
| 220 |
+
"learning_rate": 3.142151481888035e-05,
|
| 221 |
+
"loss": 0.3357,
|
| 222 |
+
"step": 5200
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 8.33,
|
| 226 |
+
"eval_cer": 0.09861871253583529,
|
| 227 |
+
"eval_loss": 0.317330539226532,
|
| 228 |
+
"eval_runtime": 293.5785,
|
| 229 |
+
"eval_samples_per_second": 13.09,
|
| 230 |
+
"eval_steps_per_second": 1.638,
|
| 231 |
+
"eval_wer": 0.3196404808132813,
|
| 232 |
+
"step": 5200
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 8.97,
|
| 236 |
+
"learning_rate": 2.5933040614709113e-05,
|
| 237 |
+
"loss": 0.3225,
|
| 238 |
+
"step": 5600
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 8.97,
|
| 242 |
+
"eval_cer": 0.09852868007676452,
|
| 243 |
+
"eval_loss": 0.31416377425193787,
|
| 244 |
+
"eval_runtime": 294.148,
|
| 245 |
+
"eval_samples_per_second": 13.065,
|
| 246 |
+
"eval_steps_per_second": 1.635,
|
| 247 |
+
"eval_wer": 0.31769890268112305,
|
| 248 |
+
"step": 5600
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 9.62,
|
| 252 |
+
"learning_rate": 2.044456641053787e-05,
|
| 253 |
+
"loss": 0.3057,
|
| 254 |
+
"step": 6000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 9.62,
|
| 258 |
+
"eval_cer": 0.09754780012794086,
|
| 259 |
+
"eval_loss": 0.31993794441223145,
|
| 260 |
+
"eval_runtime": 293.1881,
|
| 261 |
+
"eval_samples_per_second": 13.108,
|
| 262 |
+
"eval_steps_per_second": 1.641,
|
| 263 |
+
"eval_wer": 0.3155828006719169,
|
| 264 |
+
"step": 6000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 10.26,
|
| 268 |
+
"learning_rate": 1.4956092206366632e-05,
|
| 269 |
+
"loss": 0.2972,
|
| 270 |
+
"step": 6400
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 10.26,
|
| 274 |
+
"eval_cer": 0.09668064539057503,
|
| 275 |
+
"eval_loss": 0.3139244616031647,
|
| 276 |
+
"eval_runtime": 292.3994,
|
| 277 |
+
"eval_samples_per_second": 13.143,
|
| 278 |
+
"eval_steps_per_second": 1.645,
|
| 279 |
+
"eval_wer": 0.3127686031545191,
|
| 280 |
+
"step": 6400
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 10.9,
|
| 284 |
+
"learning_rate": 9.46761800219539e-06,
|
| 285 |
+
"loss": 0.2881,
|
| 286 |
+
"step": 6800
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 10.9,
|
| 290 |
+
"eval_cer": 0.09570450399222878,
|
| 291 |
+
"eval_loss": 0.31835824251174927,
|
| 292 |
+
"eval_runtime": 293.4248,
|
| 293 |
+
"eval_samples_per_second": 13.097,
|
| 294 |
+
"eval_steps_per_second": 1.639,
|
| 295 |
+
"eval_wer": 0.31065250114531295,
|
| 296 |
+
"step": 6800
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 11.54,
|
| 300 |
+
"learning_rate": 3.979143798024149e-06,
|
| 301 |
+
"loss": 0.2791,
|
| 302 |
+
"step": 7200
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 11.54,
|
| 306 |
+
"eval_cer": 0.09584192195607363,
|
| 307 |
+
"eval_loss": 0.3184186816215515,
|
| 308 |
+
"eval_runtime": 293.2834,
|
| 309 |
+
"eval_samples_per_second": 13.103,
|
| 310 |
+
"eval_steps_per_second": 1.64,
|
| 311 |
+
"eval_wer": 0.31036889984511007,
|
| 312 |
+
"step": 7200
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 12.0,
|
| 316 |
+
"step": 7488,
|
| 317 |
+
"total_flos": 5.018412278085267e+19,
|
| 318 |
+
"train_loss": 0.24150397533025497,
|
| 319 |
+
"train_runtime": 25397.5173,
|
| 320 |
+
"train_samples_per_second": 18.876,
|
| 321 |
+
"train_steps_per_second": 0.295
|
| 322 |
+
}
|
| 323 |
+
],
|
| 324 |
+
"max_steps": 7488,
|
| 325 |
+
"num_train_epochs": 12,
|
| 326 |
+
"total_flos": 5.018412278085267e+19,
|
| 327 |
+
"trial_name": null,
|
| 328 |
+
"trial_params": null
|
| 329 |
+
}
|