Model save
Browse files- README.md +38 -38
- trainer_state.json +425 -425
README.md
CHANGED
|
@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [jonatasgrosman/wav2vec2-large-xlsr-53-japanese](https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-japanese) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Wer: 0.
|
| 22 |
-
- Cer: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
@@ -51,41 +51,41 @@ The following hyperparameters were used during training:
|
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
|
| 54 |
-
| 14.
|
| 55 |
-
|
|
| 56 |
-
| 2.
|
| 57 |
-
| 1.
|
| 58 |
-
|
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
-
| 0.
|
| 80 |
-
| 0.
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
| 85 |
-
| 0.
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.
|
| 89 |
|
| 90 |
|
| 91 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [jonatasgrosman/wav2vec2-large-xlsr-53-japanese](https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-japanese) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.0032
|
| 21 |
+
- Wer: 0.1835
|
| 22 |
+
- Cer: 0.1557
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
|
| 54 |
+
| 14.5102 | 1.0 | 60 | 13.5418 | 1.0011 | 1.2966 |
|
| 55 |
+
| 6.2989 | 2.0 | 120 | 4.6126 | 1.0 | 0.8736 |
|
| 56 |
+
| 2.208 | 3.0 | 180 | 1.7112 | 0.9993 | 0.8160 |
|
| 57 |
+
| 1.098 | 4.0 | 240 | 0.8765 | 0.8410 | 0.5698 |
|
| 58 |
+
| 0.8201 | 5.0 | 300 | 0.6964 | 0.8221 | 0.5340 |
|
| 59 |
+
| 0.7499 | 6.0 | 360 | 0.6299 | 0.8217 | 0.5305 |
|
| 60 |
+
| 0.6753 | 7.0 | 420 | 0.5998 | 0.7691 | 0.4482 |
|
| 61 |
+
| 0.6003 | 8.0 | 480 | 0.5502 | 0.7394 | 0.4564 |
|
| 62 |
+
| 0.5732 | 9.0 | 540 | 0.5047 | 0.7098 | 0.3906 |
|
| 63 |
+
| 0.5404 | 10.0 | 600 | 0.4694 | 0.6679 | 0.3283 |
|
| 64 |
+
| 0.4889 | 11.0 | 660 | 0.3979 | 0.6379 | 0.3017 |
|
| 65 |
+
| 0.4401 | 12.0 | 720 | 0.3255 | 0.5849 | 0.2792 |
|
| 66 |
+
| 0.4295 | 13.0 | 780 | 0.2853 | 0.5044 | 0.2772 |
|
| 67 |
+
| 0.3216 | 14.0 | 840 | 0.2204 | 0.4511 | 0.2234 |
|
| 68 |
+
| 0.2583 | 15.0 | 900 | 0.1492 | 0.3929 | 0.2130 |
|
| 69 |
+
| 0.226 | 16.0 | 960 | 0.1007 | 0.2817 | 0.1690 |
|
| 70 |
+
| 0.2304 | 17.0 | 1020 | 0.0694 | 0.2439 | 0.1699 |
|
| 71 |
+
| 0.1487 | 18.0 | 1080 | 0.0471 | 0.2142 | 0.1758 |
|
| 72 |
+
| 0.1045 | 19.0 | 1140 | 0.0305 | 0.2168 | 0.1686 |
|
| 73 |
+
| 0.1104 | 20.0 | 1200 | 0.0256 | 0.2072 | 0.1625 |
|
| 74 |
+
| 0.094 | 21.0 | 1260 | 0.0226 | 0.2272 | 0.1760 |
|
| 75 |
+
| 0.0987 | 22.0 | 1320 | 0.0129 | 0.2013 | 0.1900 |
|
| 76 |
+
| 0.0753 | 23.0 | 1380 | 0.0110 | 0.2053 | 0.1786 |
|
| 77 |
+
| 0.0544 | 24.0 | 1440 | 0.0091 | 0.1909 | 0.1858 |
|
| 78 |
+
| 0.0684 | 25.0 | 1500 | 0.0083 | 0.1901 | 0.1728 |
|
| 79 |
+
| 0.0723 | 26.0 | 1560 | 0.0083 | 0.2027 | 0.1854 |
|
| 80 |
+
| 0.061 | 27.0 | 1620 | 0.0061 | 0.2020 | 0.1779 |
|
| 81 |
+
| 0.0635 | 28.0 | 1680 | 0.0059 | 0.1964 | 0.1818 |
|
| 82 |
+
| 0.0336 | 29.0 | 1740 | 0.0048 | 0.1887 | 0.1574 |
|
| 83 |
+
| 0.0455 | 30.0 | 1800 | 0.0036 | 0.1842 | 0.1694 |
|
| 84 |
+
| 0.0672 | 31.0 | 1860 | 0.0038 | 0.1838 | 0.1507 |
|
| 85 |
+
| 0.0315 | 32.0 | 1920 | 0.0033 | 0.1853 | 0.1555 |
|
| 86 |
+
| 0.0466 | 33.0 | 1980 | 0.0033 | 0.1827 | 0.1569 |
|
| 87 |
+
| 0.0491 | 34.0 | 2040 | 0.0035 | 0.1835 | 0.1556 |
|
| 88 |
+
| 0.0315 | 35.0 | 2100 | 0.0032 | 0.1835 | 0.1557 |
|
| 89 |
|
| 90 |
|
| 91 |
### Framework versions
|
trainer_state.json
CHANGED
|
@@ -11,1620 +11,1620 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.17,
|
| 13 |
"learning_rate": 5.000000000000001e-07,
|
| 14 |
-
"loss": 17.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.33,
|
| 19 |
"learning_rate": 1.0000000000000002e-06,
|
| 20 |
-
"loss": 17.
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.5,
|
| 25 |
"learning_rate": 1.5e-06,
|
| 26 |
-
"loss": 16.
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.67,
|
| 31 |
"learning_rate": 2.0000000000000003e-06,
|
| 32 |
-
"loss": 16.
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.83,
|
| 37 |
"learning_rate": 2.5e-06,
|
| 38 |
-
"loss":
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 1.0,
|
| 43 |
"learning_rate": 3e-06,
|
| 44 |
-
"loss": 14.
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 1.0,
|
| 49 |
-
"eval_cer": 1.
|
| 50 |
-
"eval_loss": 13.
|
| 51 |
-
"eval_runtime":
|
| 52 |
-
"eval_samples_per_second":
|
| 53 |
-
"eval_steps_per_second": 2.
|
| 54 |
-
"eval_wer":
|
| 55 |
"step": 60
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.17,
|
| 59 |
"learning_rate": 3.5000000000000004e-06,
|
| 60 |
-
"loss": 13.
|
| 61 |
"step": 70
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 1.33,
|
| 65 |
"learning_rate": 4.000000000000001e-06,
|
| 66 |
-
"loss": 11.
|
| 67 |
"step": 80
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 1.5,
|
| 71 |
"learning_rate": 4.5e-06,
|
| 72 |
-
"loss":
|
| 73 |
"step": 90
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 1.67,
|
| 77 |
"learning_rate": 5e-06,
|
| 78 |
-
"loss":
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.83,
|
| 83 |
"learning_rate": 5.500000000000001e-06,
|
| 84 |
-
"loss":
|
| 85 |
"step": 110
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"epoch": 2.0,
|
| 89 |
"learning_rate": 6e-06,
|
| 90 |
-
"loss":
|
| 91 |
"step": 120
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 2.0,
|
| 95 |
-
"eval_cer": 0.
|
| 96 |
-
"eval_loss":
|
| 97 |
-
"eval_runtime": 7.
|
| 98 |
-
"eval_samples_per_second":
|
| 99 |
-
"eval_steps_per_second":
|
| 100 |
-
"eval_wer": 0
|
| 101 |
"step": 120
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 2.17,
|
| 105 |
"learning_rate": 6.5000000000000004e-06,
|
| 106 |
-
"loss":
|
| 107 |
"step": 130
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 2.33,
|
| 111 |
"learning_rate": 7.000000000000001e-06,
|
| 112 |
-
"loss":
|
| 113 |
"step": 140
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 2.5,
|
| 117 |
"learning_rate": 7.5e-06,
|
| 118 |
-
"loss": 3.
|
| 119 |
"step": 150
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 2.67,
|
| 123 |
"learning_rate": 8.000000000000001e-06,
|
| 124 |
-
"loss": 2.
|
| 125 |
"step": 160
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 2.83,
|
| 129 |
"learning_rate": 8.500000000000002e-06,
|
| 130 |
-
"loss": 2.
|
| 131 |
"step": 170
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.0,
|
| 135 |
"learning_rate": 9e-06,
|
| 136 |
-
"loss": 2.
|
| 137 |
"step": 180
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 3.0,
|
| 141 |
-
"eval_cer": 0.
|
| 142 |
-
"eval_loss":
|
| 143 |
-
"eval_runtime":
|
| 144 |
-
"eval_samples_per_second": 68.
|
| 145 |
-
"eval_steps_per_second": 2.
|
| 146 |
-
"eval_wer":
|
| 147 |
"step": 180
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 3.17,
|
| 151 |
"learning_rate": 9.5e-06,
|
| 152 |
-
"loss":
|
| 153 |
"step": 190
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 3.33,
|
| 157 |
"learning_rate": 1e-05,
|
| 158 |
-
"loss":
|
| 159 |
"step": 200
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 3.5,
|
| 163 |
"learning_rate": 1.05e-05,
|
| 164 |
-
"loss":
|
| 165 |
"step": 210
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 3.67,
|
| 169 |
"learning_rate": 1.1000000000000001e-05,
|
| 170 |
-
"loss": 1.
|
| 171 |
"step": 220
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 3.83,
|
| 175 |
"learning_rate": 1.1500000000000002e-05,
|
| 176 |
-
"loss": 1.
|
| 177 |
"step": 230
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 4.0,
|
| 181 |
"learning_rate": 1.2e-05,
|
| 182 |
-
"loss": 1.
|
| 183 |
"step": 240
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 4.0,
|
| 187 |
-
"eval_cer": 0.
|
| 188 |
-
"eval_loss":
|
| 189 |
-
"eval_runtime":
|
| 190 |
-
"eval_samples_per_second": 68.
|
| 191 |
-
"eval_steps_per_second": 2.
|
| 192 |
-
"eval_wer": 0.
|
| 193 |
"step": 240
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 4.17,
|
| 197 |
"learning_rate": 1.25e-05,
|
| 198 |
-
"loss": 1.
|
| 199 |
"step": 250
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 4.33,
|
| 203 |
"learning_rate": 1.3000000000000001e-05,
|
| 204 |
-
"loss":
|
| 205 |
"step": 260
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 4.5,
|
| 209 |
"learning_rate": 1.3500000000000001e-05,
|
| 210 |
-
"loss":
|
| 211 |
"step": 270
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 4.67,
|
| 215 |
"learning_rate": 1.4000000000000001e-05,
|
| 216 |
-
"loss":
|
| 217 |
"step": 280
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 4.83,
|
| 221 |
"learning_rate": 1.45e-05,
|
| 222 |
-
"loss":
|
| 223 |
"step": 290
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 5.0,
|
| 227 |
"learning_rate": 1.5e-05,
|
| 228 |
-
"loss":
|
| 229 |
"step": 300
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 5.0,
|
| 233 |
-
"eval_cer": 0.
|
| 234 |
-
"eval_loss": 0.
|
| 235 |
-
"eval_runtime":
|
| 236 |
-
"eval_samples_per_second": 68.
|
| 237 |
-
"eval_steps_per_second": 2.
|
| 238 |
-
"eval_wer": 0.
|
| 239 |
"step": 300
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 5.17,
|
| 243 |
"learning_rate": 1.55e-05,
|
| 244 |
-
"loss":
|
| 245 |
"step": 310
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 5.33,
|
| 249 |
"learning_rate": 1.6000000000000003e-05,
|
| 250 |
-
"loss": 0.
|
| 251 |
"step": 320
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 5.5,
|
| 255 |
"learning_rate": 1.65e-05,
|
| 256 |
-
"loss": 0.
|
| 257 |
"step": 330
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 5.67,
|
| 261 |
"learning_rate": 1.7000000000000003e-05,
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 340
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 5.83,
|
| 267 |
"learning_rate": 1.75e-05,
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 350
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 6.0,
|
| 273 |
"learning_rate": 1.8e-05,
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 360
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 6.0,
|
| 279 |
-
"eval_cer": 0.
|
| 280 |
-
"eval_loss": 0.
|
| 281 |
-
"eval_runtime":
|
| 282 |
-
"eval_samples_per_second": 68.
|
| 283 |
-
"eval_steps_per_second": 2.
|
| 284 |
-
"eval_wer": 0.
|
| 285 |
"step": 360
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 6.17,
|
| 289 |
"learning_rate": 1.85e-05,
|
| 290 |
-
"loss": 0.
|
| 291 |
"step": 370
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 6.33,
|
| 295 |
"learning_rate": 1.9e-05,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 380
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 6.5,
|
| 301 |
"learning_rate": 1.9500000000000003e-05,
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 390
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 6.67,
|
| 307 |
"learning_rate": 2e-05,
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 400
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 6.83,
|
| 313 |
"learning_rate": 2.05e-05,
|
| 314 |
-
"loss": 0.
|
| 315 |
"step": 410
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 7.0,
|
| 319 |
"learning_rate": 2.1e-05,
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 420
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 7.0,
|
| 325 |
-
"eval_cer": 0.
|
| 326 |
-
"eval_loss": 0.
|
| 327 |
-
"eval_runtime":
|
| 328 |
-
"eval_samples_per_second": 68.
|
| 329 |
-
"eval_steps_per_second": 2.
|
| 330 |
-
"eval_wer": 0.
|
| 331 |
"step": 420
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 7.17,
|
| 335 |
"learning_rate": 2.15e-05,
|
| 336 |
-
"loss": 0.
|
| 337 |
"step": 430
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 7.33,
|
| 341 |
"learning_rate": 2.2000000000000003e-05,
|
| 342 |
-
"loss": 0.
|
| 343 |
"step": 440
|
| 344 |
},
|
| 345 |
{
|
| 346 |
"epoch": 7.5,
|
| 347 |
"learning_rate": 2.25e-05,
|
| 348 |
-
"loss": 0.
|
| 349 |
"step": 450
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 7.67,
|
| 353 |
"learning_rate": 2.3000000000000003e-05,
|
| 354 |
-
"loss": 0.
|
| 355 |
"step": 460
|
| 356 |
},
|
| 357 |
{
|
| 358 |
"epoch": 7.83,
|
| 359 |
"learning_rate": 2.35e-05,
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 470
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 8.0,
|
| 365 |
"learning_rate": 2.4e-05,
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 480
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 8.0,
|
| 371 |
-
"eval_cer": 0.
|
| 372 |
-
"eval_loss": 0.
|
| 373 |
-
"eval_runtime":
|
| 374 |
-
"eval_samples_per_second": 68.
|
| 375 |
-
"eval_steps_per_second": 2.
|
| 376 |
-
"eval_wer": 0.
|
| 377 |
"step": 480
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 8.17,
|
| 381 |
"learning_rate": 2.45e-05,
|
| 382 |
-
"loss": 0.
|
| 383 |
"step": 490
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 8.33,
|
| 387 |
"learning_rate": 2.5e-05,
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 500
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 8.5,
|
| 393 |
"learning_rate": 2.5500000000000003e-05,
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 510
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 8.67,
|
| 399 |
"learning_rate": 2.6000000000000002e-05,
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 520
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 8.83,
|
| 405 |
"learning_rate": 2.6500000000000004e-05,
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 530
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 9.0,
|
| 411 |
"learning_rate": 2.7000000000000002e-05,
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 540
|
| 414 |
},
|
| 415 |
{
|
| 416 |
"epoch": 9.0,
|
| 417 |
-
"eval_cer": 0.
|
| 418 |
-
"eval_loss": 0.
|
| 419 |
-
"eval_runtime": 7.
|
| 420 |
-
"eval_samples_per_second":
|
| 421 |
-
"eval_steps_per_second": 2.
|
| 422 |
-
"eval_wer": 0.
|
| 423 |
"step": 540
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 9.17,
|
| 427 |
"learning_rate": 2.7500000000000004e-05,
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 550
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 9.33,
|
| 433 |
"learning_rate": 2.8000000000000003e-05,
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 560
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 9.5,
|
| 439 |
"learning_rate": 2.8499999999999998e-05,
|
| 440 |
-
"loss": 0.
|
| 441 |
"step": 570
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 9.67,
|
| 445 |
"learning_rate": 2.9e-05,
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 580
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 9.83,
|
| 451 |
"learning_rate": 2.95e-05,
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 590
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 10.0,
|
| 457 |
"learning_rate": 3e-05,
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 600
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 10.0,
|
| 463 |
-
"eval_cer": 0.
|
| 464 |
-
"eval_loss": 0.
|
| 465 |
-
"eval_runtime": 7.
|
| 466 |
-
"eval_samples_per_second": 68.
|
| 467 |
-
"eval_steps_per_second": 2.
|
| 468 |
-
"eval_wer": 0.
|
| 469 |
"step": 600
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 10.17,
|
| 473 |
"learning_rate": 3.05e-05,
|
| 474 |
-
"loss": 0.
|
| 475 |
"step": 610
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 10.33,
|
| 479 |
"learning_rate": 3.1e-05,
|
| 480 |
-
"loss": 0.
|
| 481 |
"step": 620
|
| 482 |
},
|
| 483 |
{
|
| 484 |
"epoch": 10.5,
|
| 485 |
"learning_rate": 3.15e-05,
|
| 486 |
-
"loss": 0.
|
| 487 |
"step": 630
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 10.67,
|
| 491 |
"learning_rate": 3.2000000000000005e-05,
|
| 492 |
-
"loss": 0.
|
| 493 |
"step": 640
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 10.83,
|
| 497 |
"learning_rate": 3.2500000000000004e-05,
|
| 498 |
-
"loss": 0.
|
| 499 |
"step": 650
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 11.0,
|
| 503 |
"learning_rate": 3.3e-05,
|
| 504 |
-
"loss": 0.
|
| 505 |
"step": 660
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 11.0,
|
| 509 |
-
"eval_cer": 0.
|
| 510 |
-
"eval_loss": 0.
|
| 511 |
-
"eval_runtime": 7.
|
| 512 |
-
"eval_samples_per_second": 68.
|
| 513 |
-
"eval_steps_per_second": 2.
|
| 514 |
-
"eval_wer": 0.
|
| 515 |
"step": 660
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 11.17,
|
| 519 |
"learning_rate": 3.35e-05,
|
| 520 |
-
"loss": 0.
|
| 521 |
"step": 670
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 11.33,
|
| 525 |
"learning_rate": 3.4000000000000007e-05,
|
| 526 |
-
"loss": 0.
|
| 527 |
"step": 680
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 11.5,
|
| 531 |
"learning_rate": 3.45e-05,
|
| 532 |
-
"loss": 0.
|
| 533 |
"step": 690
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 11.67,
|
| 537 |
"learning_rate": 3.5e-05,
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 700
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 11.83,
|
| 543 |
"learning_rate": 3.55e-05,
|
| 544 |
-
"loss": 0.
|
| 545 |
"step": 710
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 12.0,
|
| 549 |
"learning_rate": 3.6e-05,
|
| 550 |
-
"loss": 0.
|
| 551 |
"step": 720
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 12.0,
|
| 555 |
-
"eval_cer": 0.
|
| 556 |
-
"eval_loss": 0.
|
| 557 |
-
"eval_runtime": 7.
|
| 558 |
-
"eval_samples_per_second": 68.
|
| 559 |
-
"eval_steps_per_second": 2.
|
| 560 |
-
"eval_wer": 0.
|
| 561 |
"step": 720
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 12.17,
|
| 565 |
"learning_rate": 3.65e-05,
|
| 566 |
-
"loss": 0.
|
| 567 |
"step": 730
|
| 568 |
},
|
| 569 |
{
|
| 570 |
"epoch": 12.33,
|
| 571 |
"learning_rate": 3.7e-05,
|
| 572 |
-
"loss": 0.
|
| 573 |
"step": 740
|
| 574 |
},
|
| 575 |
{
|
| 576 |
"epoch": 12.5,
|
| 577 |
"learning_rate": 3.7500000000000003e-05,
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 750
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 12.67,
|
| 583 |
"learning_rate": 3.8e-05,
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 760
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 12.83,
|
| 589 |
"learning_rate": 3.85e-05,
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 770
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 13.0,
|
| 595 |
"learning_rate": 3.9000000000000006e-05,
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 780
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 13.0,
|
| 601 |
-
"eval_cer": 0.
|
| 602 |
-
"eval_loss": 0.
|
| 603 |
-
"eval_runtime":
|
| 604 |
-
"eval_samples_per_second":
|
| 605 |
-
"eval_steps_per_second": 2.
|
| 606 |
-
"eval_wer": 0.
|
| 607 |
"step": 780
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 13.17,
|
| 611 |
"learning_rate": 3.9500000000000005e-05,
|
| 612 |
-
"loss": 0.
|
| 613 |
"step": 790
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"epoch": 13.33,
|
| 617 |
"learning_rate": 4e-05,
|
| 618 |
-
"loss": 0.
|
| 619 |
"step": 800
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 13.5,
|
| 623 |
"learning_rate": 4.05e-05,
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 810
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 13.67,
|
| 629 |
"learning_rate": 4.1e-05,
|
| 630 |
-
"loss": 0.
|
| 631 |
"step": 820
|
| 632 |
},
|
| 633 |
{
|
| 634 |
"epoch": 13.83,
|
| 635 |
"learning_rate": 4.15e-05,
|
| 636 |
-
"loss": 0.
|
| 637 |
"step": 830
|
| 638 |
},
|
| 639 |
{
|
| 640 |
"epoch": 14.0,
|
| 641 |
"learning_rate": 4.2e-05,
|
| 642 |
-
"loss": 0.
|
| 643 |
"step": 840
|
| 644 |
},
|
| 645 |
{
|
| 646 |
"epoch": 14.0,
|
| 647 |
-
"eval_cer": 0.
|
| 648 |
-
"eval_loss": 0.
|
| 649 |
-
"eval_runtime":
|
| 650 |
-
"eval_samples_per_second": 68.
|
| 651 |
-
"eval_steps_per_second": 2.
|
| 652 |
-
"eval_wer": 0.
|
| 653 |
"step": 840
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 14.17,
|
| 657 |
"learning_rate": 4.25e-05,
|
| 658 |
-
"loss": 0.
|
| 659 |
"step": 850
|
| 660 |
},
|
| 661 |
{
|
| 662 |
"epoch": 14.33,
|
| 663 |
"learning_rate": 4.3e-05,
|
| 664 |
-
"loss": 0.
|
| 665 |
"step": 860
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 14.5,
|
| 669 |
"learning_rate": 4.35e-05,
|
| 670 |
-
"loss": 0.
|
| 671 |
"step": 870
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 14.67,
|
| 675 |
"learning_rate": 4.4000000000000006e-05,
|
| 676 |
-
"loss": 0.
|
| 677 |
"step": 880
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 14.83,
|
| 681 |
"learning_rate": 4.4500000000000004e-05,
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 890
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 15.0,
|
| 687 |
"learning_rate": 4.5e-05,
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 900
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 15.0,
|
| 693 |
-
"eval_cer": 0.
|
| 694 |
-
"eval_loss": 0.
|
| 695 |
-
"eval_runtime":
|
| 696 |
-
"eval_samples_per_second":
|
| 697 |
-
"eval_steps_per_second": 2.
|
| 698 |
-
"eval_wer": 0.
|
| 699 |
"step": 900
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 15.17,
|
| 703 |
"learning_rate": 4.55e-05,
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 910
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 15.33,
|
| 709 |
"learning_rate": 4.600000000000001e-05,
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 920
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 15.5,
|
| 715 |
"learning_rate": 4.6500000000000005e-05,
|
| 716 |
-
"loss": 0.
|
| 717 |
"step": 930
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 15.67,
|
| 721 |
"learning_rate": 4.7e-05,
|
| 722 |
-
"loss": 0.
|
| 723 |
"step": 940
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 15.83,
|
| 727 |
"learning_rate": 4.75e-05,
|
| 728 |
-
"loss": 0.
|
| 729 |
"step": 950
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 16.0,
|
| 733 |
"learning_rate": 4.8e-05,
|
| 734 |
-
"loss": 0.
|
| 735 |
"step": 960
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 16.0,
|
| 739 |
-
"eval_cer": 0.
|
| 740 |
-
"eval_loss": 0.
|
| 741 |
-
"eval_runtime": 7.
|
| 742 |
-
"eval_samples_per_second":
|
| 743 |
-
"eval_steps_per_second": 2.
|
| 744 |
-
"eval_wer": 0.
|
| 745 |
"step": 960
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 16.17,
|
| 749 |
"learning_rate": 4.85e-05,
|
| 750 |
-
"loss": 0.
|
| 751 |
"step": 970
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 16.33,
|
| 755 |
"learning_rate": 4.9e-05,
|
| 756 |
-
"loss": 0.
|
| 757 |
"step": 980
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 16.5,
|
| 761 |
"learning_rate": 4.9500000000000004e-05,
|
| 762 |
-
"loss": 0.
|
| 763 |
"step": 990
|
| 764 |
},
|
| 765 |
{
|
| 766 |
"epoch": 16.67,
|
| 767 |
"learning_rate": 5e-05,
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 1000
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 16.83,
|
| 773 |
"learning_rate": 4.9545454545454553e-05,
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 1010
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 17.0,
|
| 779 |
"learning_rate": 4.909090909090909e-05,
|
| 780 |
-
"loss": 0.
|
| 781 |
"step": 1020
|
| 782 |
},
|
| 783 |
{
|
| 784 |
"epoch": 17.0,
|
| 785 |
-
"eval_cer": 0.
|
| 786 |
-
"eval_loss": 0.
|
| 787 |
-
"eval_runtime": 7.
|
| 788 |
-
"eval_samples_per_second":
|
| 789 |
-
"eval_steps_per_second": 2.
|
| 790 |
-
"eval_wer": 0.
|
| 791 |
"step": 1020
|
| 792 |
},
|
| 793 |
{
|
| 794 |
"epoch": 17.17,
|
| 795 |
"learning_rate": 4.863636363636364e-05,
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 1030
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 17.33,
|
| 801 |
"learning_rate": 4.8181818181818186e-05,
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 1040
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 17.5,
|
| 807 |
"learning_rate": 4.772727272727273e-05,
|
| 808 |
-
"loss": 0.
|
| 809 |
"step": 1050
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 17.67,
|
| 813 |
"learning_rate": 4.7272727272727275e-05,
|
| 814 |
-
"loss": 0.
|
| 815 |
"step": 1060
|
| 816 |
},
|
| 817 |
{
|
| 818 |
"epoch": 17.83,
|
| 819 |
"learning_rate": 4.681818181818182e-05,
|
| 820 |
-
"loss": 0.
|
| 821 |
"step": 1070
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 18.0,
|
| 825 |
"learning_rate": 4.636363636363636e-05,
|
| 826 |
-
"loss": 0.
|
| 827 |
"step": 1080
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 18.0,
|
| 831 |
-
"eval_cer": 0.
|
| 832 |
-
"eval_loss": 0.
|
| 833 |
-
"eval_runtime": 7.
|
| 834 |
-
"eval_samples_per_second":
|
| 835 |
-
"eval_steps_per_second": 2.
|
| 836 |
-
"eval_wer": 0.
|
| 837 |
"step": 1080
|
| 838 |
},
|
| 839 |
{
|
| 840 |
"epoch": 18.17,
|
| 841 |
"learning_rate": 4.5909090909090914e-05,
|
| 842 |
-
"loss": 0.
|
| 843 |
"step": 1090
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 18.33,
|
| 847 |
"learning_rate": 4.545454545454546e-05,
|
| 848 |
-
"loss": 0.
|
| 849 |
"step": 1100
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 18.5,
|
| 853 |
"learning_rate": 4.5e-05,
|
| 854 |
-
"loss": 0.
|
| 855 |
"step": 1110
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 18.67,
|
| 859 |
"learning_rate": 4.454545454545455e-05,
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 1120
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 18.83,
|
| 865 |
"learning_rate": 4.409090909090909e-05,
|
| 866 |
-
"loss": 0.
|
| 867 |
"step": 1130
|
| 868 |
},
|
| 869 |
{
|
| 870 |
"epoch": 19.0,
|
| 871 |
"learning_rate": 4.3636363636363636e-05,
|
| 872 |
-
"loss": 0.
|
| 873 |
"step": 1140
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 19.0,
|
| 877 |
-
"eval_cer": 0.
|
| 878 |
-
"eval_loss": 0.
|
| 879 |
-
"eval_runtime":
|
| 880 |
-
"eval_samples_per_second": 68.
|
| 881 |
-
"eval_steps_per_second": 2.
|
| 882 |
-
"eval_wer": 0.
|
| 883 |
"step": 1140
|
| 884 |
},
|
| 885 |
{
|
| 886 |
"epoch": 19.17,
|
| 887 |
"learning_rate": 4.318181818181819e-05,
|
| 888 |
-
"loss": 0.
|
| 889 |
"step": 1150
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 19.33,
|
| 893 |
"learning_rate": 4.2727272727272724e-05,
|
| 894 |
-
"loss": 0.
|
| 895 |
"step": 1160
|
| 896 |
},
|
| 897 |
{
|
| 898 |
"epoch": 19.5,
|
| 899 |
"learning_rate": 4.2272727272727275e-05,
|
| 900 |
-
"loss": 0.
|
| 901 |
"step": 1170
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 19.67,
|
| 905 |
"learning_rate": 4.181818181818182e-05,
|
| 906 |
-
"loss": 0.
|
| 907 |
"step": 1180
|
| 908 |
},
|
| 909 |
{
|
| 910 |
"epoch": 19.83,
|
| 911 |
"learning_rate": 4.1363636363636364e-05,
|
| 912 |
-
"loss": 0.
|
| 913 |
"step": 1190
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 20.0,
|
| 917 |
"learning_rate": 4.0909090909090915e-05,
|
| 918 |
-
"loss": 0.
|
| 919 |
"step": 1200
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 20.0,
|
| 923 |
-
"eval_cer": 0.
|
| 924 |
-
"eval_loss": 0.
|
| 925 |
-
"eval_runtime": 7.
|
| 926 |
-
"eval_samples_per_second":
|
| 927 |
-
"eval_steps_per_second": 2.
|
| 928 |
-
"eval_wer": 0.
|
| 929 |
"step": 1200
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 20.17,
|
| 933 |
"learning_rate": 4.045454545454546e-05,
|
| 934 |
-
"loss": 0.
|
| 935 |
"step": 1210
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 20.33,
|
| 939 |
"learning_rate": 4e-05,
|
| 940 |
-
"loss": 0.
|
| 941 |
"step": 1220
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 20.5,
|
| 945 |
"learning_rate": 3.954545454545455e-05,
|
| 946 |
-
"loss": 0.
|
| 947 |
"step": 1230
|
| 948 |
},
|
| 949 |
{
|
| 950 |
"epoch": 20.67,
|
| 951 |
"learning_rate": 3.909090909090909e-05,
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 1240
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 20.83,
|
| 957 |
"learning_rate": 3.8636363636363636e-05,
|
| 958 |
-
"loss": 0.
|
| 959 |
"step": 1250
|
| 960 |
},
|
| 961 |
{
|
| 962 |
"epoch": 21.0,
|
| 963 |
"learning_rate": 3.818181818181819e-05,
|
| 964 |
-
"loss": 0.
|
| 965 |
"step": 1260
|
| 966 |
},
|
| 967 |
{
|
| 968 |
"epoch": 21.0,
|
| 969 |
-
"eval_cer": 0.
|
| 970 |
-
"eval_loss": 0.
|
| 971 |
-
"eval_runtime": 7.
|
| 972 |
-
"eval_samples_per_second":
|
| 973 |
-
"eval_steps_per_second": 2.
|
| 974 |
-
"eval_wer": 0.
|
| 975 |
"step": 1260
|
| 976 |
},
|
| 977 |
{
|
| 978 |
"epoch": 21.17,
|
| 979 |
"learning_rate": 3.7727272727272725e-05,
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 1270
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 21.33,
|
| 985 |
"learning_rate": 3.7272727272727276e-05,
|
| 986 |
-
"loss": 0.
|
| 987 |
"step": 1280
|
| 988 |
},
|
| 989 |
{
|
| 990 |
"epoch": 21.5,
|
| 991 |
"learning_rate": 3.681818181818182e-05,
|
| 992 |
-
"loss": 0.
|
| 993 |
"step": 1290
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 21.67,
|
| 997 |
"learning_rate": 3.6363636363636364e-05,
|
| 998 |
-
"loss": 0.
|
| 999 |
"step": 1300
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
"epoch": 21.83,
|
| 1003 |
"learning_rate": 3.590909090909091e-05,
|
| 1004 |
-
"loss": 0.
|
| 1005 |
"step": 1310
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
"epoch": 22.0,
|
| 1009 |
"learning_rate": 3.545454545454546e-05,
|
| 1010 |
-
"loss": 0.
|
| 1011 |
"step": 1320
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
"epoch": 22.0,
|
| 1015 |
-
"eval_cer": 0.
|
| 1016 |
-
"eval_loss": 0.
|
| 1017 |
-
"eval_runtime": 7.
|
| 1018 |
-
"eval_samples_per_second":
|
| 1019 |
-
"eval_steps_per_second": 2.
|
| 1020 |
-
"eval_wer": 0.
|
| 1021 |
"step": 1320
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 22.17,
|
| 1025 |
"learning_rate": 3.5e-05,
|
| 1026 |
-
"loss": 0.
|
| 1027 |
"step": 1330
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
"epoch": 22.33,
|
| 1031 |
"learning_rate": 3.454545454545455e-05,
|
| 1032 |
-
"loss": 0.
|
| 1033 |
"step": 1340
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 22.5,
|
| 1037 |
"learning_rate": 3.409090909090909e-05,
|
| 1038 |
-
"loss": 0.
|
| 1039 |
"step": 1350
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
"epoch": 22.67,
|
| 1043 |
"learning_rate": 3.3636363636363636e-05,
|
| 1044 |
-
"loss": 0.
|
| 1045 |
"step": 1360
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
"epoch": 22.83,
|
| 1049 |
"learning_rate": 3.318181818181819e-05,
|
| 1050 |
-
"loss": 0.
|
| 1051 |
"step": 1370
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 23.0,
|
| 1055 |
"learning_rate": 3.272727272727273e-05,
|
| 1056 |
-
"loss": 0.
|
| 1057 |
"step": 1380
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 23.0,
|
| 1061 |
-
"eval_cer": 0.
|
| 1062 |
-
"eval_loss": 0.
|
| 1063 |
-
"eval_runtime":
|
| 1064 |
-
"eval_samples_per_second":
|
| 1065 |
-
"eval_steps_per_second": 2.
|
| 1066 |
-
"eval_wer": 0.
|
| 1067 |
"step": 1380
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
"epoch": 23.17,
|
| 1071 |
"learning_rate": 3.2272727272727276e-05,
|
| 1072 |
-
"loss": 0.
|
| 1073 |
"step": 1390
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 23.33,
|
| 1077 |
"learning_rate": 3.181818181818182e-05,
|
| 1078 |
-
"loss": 0.
|
| 1079 |
"step": 1400
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 23.5,
|
| 1083 |
"learning_rate": 3.1363636363636365e-05,
|
| 1084 |
-
"loss": 0.
|
| 1085 |
"step": 1410
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 23.67,
|
| 1089 |
"learning_rate": 3.090909090909091e-05,
|
| 1090 |
-
"loss": 0.
|
| 1091 |
"step": 1420
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
"epoch": 23.83,
|
| 1095 |
"learning_rate": 3.0454545454545456e-05,
|
| 1096 |
-
"loss": 0.
|
| 1097 |
"step": 1430
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
"epoch": 24.0,
|
| 1101 |
"learning_rate": 3e-05,
|
| 1102 |
-
"loss": 0.
|
| 1103 |
"step": 1440
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
"epoch": 24.0,
|
| 1107 |
-
"eval_cer": 0.
|
| 1108 |
-
"eval_loss": 0.
|
| 1109 |
-
"eval_runtime":
|
| 1110 |
-
"eval_samples_per_second":
|
| 1111 |
-
"eval_steps_per_second": 2.
|
| 1112 |
-
"eval_wer": 0.
|
| 1113 |
"step": 1440
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 24.17,
|
| 1117 |
"learning_rate": 2.954545454545455e-05,
|
| 1118 |
-
"loss": 0.
|
| 1119 |
"step": 1450
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 24.33,
|
| 1123 |
"learning_rate": 2.909090909090909e-05,
|
| 1124 |
-
"loss": 0.
|
| 1125 |
"step": 1460
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
"epoch": 24.5,
|
| 1129 |
"learning_rate": 2.863636363636364e-05,
|
| 1130 |
-
"loss": 0.
|
| 1131 |
"step": 1470
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
"epoch": 24.67,
|
| 1135 |
"learning_rate": 2.818181818181818e-05,
|
| 1136 |
-
"loss": 0.
|
| 1137 |
"step": 1480
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
"epoch": 24.83,
|
| 1141 |
"learning_rate": 2.772727272727273e-05,
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 1490
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 25.0,
|
| 1147 |
"learning_rate": 2.7272727272727273e-05,
|
| 1148 |
-
"loss": 0.
|
| 1149 |
"step": 1500
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
"epoch": 25.0,
|
| 1153 |
-
"eval_cer": 0.
|
| 1154 |
-
"eval_loss": 0.
|
| 1155 |
-
"eval_runtime": 7.
|
| 1156 |
-
"eval_samples_per_second":
|
| 1157 |
-
"eval_steps_per_second": 2.
|
| 1158 |
-
"eval_wer": 0.
|
| 1159 |
"step": 1500
|
| 1160 |
},
|
| 1161 |
{
|
| 1162 |
"epoch": 25.17,
|
| 1163 |
"learning_rate": 2.681818181818182e-05,
|
| 1164 |
-
"loss": 0.
|
| 1165 |
"step": 1510
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
"epoch": 25.33,
|
| 1169 |
"learning_rate": 2.636363636363636e-05,
|
| 1170 |
-
"loss": 0.
|
| 1171 |
"step": 1520
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 25.5,
|
| 1175 |
"learning_rate": 2.590909090909091e-05,
|
| 1176 |
-
"loss": 0.
|
| 1177 |
"step": 1530
|
| 1178 |
},
|
| 1179 |
{
|
| 1180 |
"epoch": 25.67,
|
| 1181 |
"learning_rate": 2.5454545454545454e-05,
|
| 1182 |
-
"loss": 0.
|
| 1183 |
"step": 1540
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
"epoch": 25.83,
|
| 1187 |
"learning_rate": 2.5e-05,
|
| 1188 |
-
"loss": 0.
|
| 1189 |
"step": 1550
|
| 1190 |
},
|
| 1191 |
{
|
| 1192 |
"epoch": 26.0,
|
| 1193 |
"learning_rate": 2.4545454545454545e-05,
|
| 1194 |
-
"loss": 0.
|
| 1195 |
"step": 1560
|
| 1196 |
},
|
| 1197 |
{
|
| 1198 |
"epoch": 26.0,
|
| 1199 |
-
"eval_cer": 0.
|
| 1200 |
-
"eval_loss": 0.
|
| 1201 |
-
"eval_runtime": 7.
|
| 1202 |
-
"eval_samples_per_second":
|
| 1203 |
-
"eval_steps_per_second": 2.
|
| 1204 |
-
"eval_wer": 0.
|
| 1205 |
"step": 1560
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
"epoch": 26.17,
|
| 1209 |
"learning_rate": 2.4090909090909093e-05,
|
| 1210 |
-
"loss": 0.
|
| 1211 |
"step": 1570
|
| 1212 |
},
|
| 1213 |
{
|
| 1214 |
"epoch": 26.33,
|
| 1215 |
"learning_rate": 2.3636363636363637e-05,
|
| 1216 |
-
"loss": 0.
|
| 1217 |
"step": 1580
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 26.5,
|
| 1221 |
"learning_rate": 2.318181818181818e-05,
|
| 1222 |
-
"loss": 0.
|
| 1223 |
"step": 1590
|
| 1224 |
},
|
| 1225 |
{
|
| 1226 |
"epoch": 26.67,
|
| 1227 |
"learning_rate": 2.272727272727273e-05,
|
| 1228 |
-
"loss": 0.
|
| 1229 |
"step": 1600
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
"epoch": 26.83,
|
| 1233 |
"learning_rate": 2.2272727272727274e-05,
|
| 1234 |
-
"loss": 0.
|
| 1235 |
"step": 1610
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
"epoch": 27.0,
|
| 1239 |
"learning_rate": 2.1818181818181818e-05,
|
| 1240 |
-
"loss": 0.
|
| 1241 |
"step": 1620
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 27.0,
|
| 1245 |
-
"eval_cer": 0.
|
| 1246 |
-
"eval_loss": 0.
|
| 1247 |
-
"eval_runtime": 7.
|
| 1248 |
-
"eval_samples_per_second":
|
| 1249 |
-
"eval_steps_per_second": 2.
|
| 1250 |
-
"eval_wer": 0.
|
| 1251 |
"step": 1620
|
| 1252 |
},
|
| 1253 |
{
|
| 1254 |
"epoch": 27.17,
|
| 1255 |
"learning_rate": 2.1363636363636362e-05,
|
| 1256 |
-
"loss": 0.
|
| 1257 |
"step": 1630
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
"epoch": 27.33,
|
| 1261 |
"learning_rate": 2.090909090909091e-05,
|
| 1262 |
-
"loss": 0.
|
| 1263 |
"step": 1640
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
"epoch": 27.5,
|
| 1267 |
"learning_rate": 2.0454545454545457e-05,
|
| 1268 |
-
"loss": 0.
|
| 1269 |
"step": 1650
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 27.67,
|
| 1273 |
"learning_rate": 2e-05,
|
| 1274 |
-
"loss": 0.
|
| 1275 |
"step": 1660
|
| 1276 |
},
|
| 1277 |
{
|
| 1278 |
"epoch": 27.83,
|
| 1279 |
"learning_rate": 1.9545454545454546e-05,
|
| 1280 |
-
"loss": 0.
|
| 1281 |
"step": 1670
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
"epoch": 28.0,
|
| 1285 |
"learning_rate": 1.9090909090909094e-05,
|
| 1286 |
-
"loss": 0.
|
| 1287 |
"step": 1680
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
"epoch": 28.0,
|
| 1291 |
-
"eval_cer": 0.
|
| 1292 |
-
"eval_loss": 0.
|
| 1293 |
-
"eval_runtime": 7.
|
| 1294 |
-
"eval_samples_per_second":
|
| 1295 |
-
"eval_steps_per_second": 2.
|
| 1296 |
-
"eval_wer": 0.
|
| 1297 |
"step": 1680
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
"epoch": 28.17,
|
| 1301 |
"learning_rate": 1.8636363636363638e-05,
|
| 1302 |
-
"loss": 0.
|
| 1303 |
"step": 1690
|
| 1304 |
},
|
| 1305 |
{
|
| 1306 |
"epoch": 28.33,
|
| 1307 |
"learning_rate": 1.8181818181818182e-05,
|
| 1308 |
-
"loss": 0.
|
| 1309 |
"step": 1700
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
"epoch": 28.5,
|
| 1313 |
"learning_rate": 1.772727272727273e-05,
|
| 1314 |
-
"loss": 0.
|
| 1315 |
"step": 1710
|
| 1316 |
},
|
| 1317 |
{
|
| 1318 |
"epoch": 28.67,
|
| 1319 |
"learning_rate": 1.7272727272727274e-05,
|
| 1320 |
-
"loss": 0.
|
| 1321 |
"step": 1720
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
"epoch": 28.83,
|
| 1325 |
"learning_rate": 1.6818181818181818e-05,
|
| 1326 |
-
"loss": 0.
|
| 1327 |
"step": 1730
|
| 1328 |
},
|
| 1329 |
{
|
| 1330 |
"epoch": 29.0,
|
| 1331 |
"learning_rate": 1.6363636363636366e-05,
|
| 1332 |
-
"loss": 0.
|
| 1333 |
"step": 1740
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
"epoch": 29.0,
|
| 1337 |
-
"eval_cer": 0.
|
| 1338 |
-
"eval_loss": 0.
|
| 1339 |
-
"eval_runtime": 7.
|
| 1340 |
-
"eval_samples_per_second":
|
| 1341 |
-
"eval_steps_per_second": 2.
|
| 1342 |
-
"eval_wer": 0.
|
| 1343 |
"step": 1740
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
"epoch": 29.17,
|
| 1347 |
"learning_rate": 1.590909090909091e-05,
|
| 1348 |
-
"loss": 0.
|
| 1349 |
"step": 1750
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
"epoch": 29.33,
|
| 1353 |
"learning_rate": 1.5454545454545454e-05,
|
| 1354 |
-
"loss": 0.
|
| 1355 |
"step": 1760
|
| 1356 |
},
|
| 1357 |
{
|
| 1358 |
"epoch": 29.5,
|
| 1359 |
"learning_rate": 1.5e-05,
|
| 1360 |
-
"loss": 0.
|
| 1361 |
"step": 1770
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
"epoch": 29.67,
|
| 1365 |
"learning_rate": 1.4545454545454545e-05,
|
| 1366 |
-
"loss": 0.
|
| 1367 |
"step": 1780
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
"epoch": 29.83,
|
| 1371 |
"learning_rate": 1.409090909090909e-05,
|
| 1372 |
-
"loss": 0.
|
| 1373 |
"step": 1790
|
| 1374 |
},
|
| 1375 |
{
|
| 1376 |
"epoch": 30.0,
|
| 1377 |
"learning_rate": 1.3636363636363637e-05,
|
| 1378 |
-
"loss": 0.
|
| 1379 |
"step": 1800
|
| 1380 |
},
|
| 1381 |
{
|
| 1382 |
"epoch": 30.0,
|
| 1383 |
-
"eval_cer": 0.
|
| 1384 |
-
"eval_loss": 0.
|
| 1385 |
-
"eval_runtime": 7.
|
| 1386 |
-
"eval_samples_per_second":
|
| 1387 |
-
"eval_steps_per_second": 2.
|
| 1388 |
-
"eval_wer": 0.
|
| 1389 |
"step": 1800
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 30.17,
|
| 1393 |
"learning_rate": 1.318181818181818e-05,
|
| 1394 |
-
"loss": 0.
|
| 1395 |
"step": 1810
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
"epoch": 30.33,
|
| 1399 |
"learning_rate": 1.2727272727272727e-05,
|
| 1400 |
-
"loss": 0.
|
| 1401 |
"step": 1820
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
"epoch": 30.5,
|
| 1405 |
"learning_rate": 1.2272727272727273e-05,
|
| 1406 |
-
"loss": 0.
|
| 1407 |
"step": 1830
|
| 1408 |
},
|
| 1409 |
{
|
| 1410 |
"epoch": 30.67,
|
| 1411 |
"learning_rate": 1.1818181818181819e-05,
|
| 1412 |
-
"loss": 0.
|
| 1413 |
"step": 1840
|
| 1414 |
},
|
| 1415 |
{
|
| 1416 |
"epoch": 30.83,
|
| 1417 |
"learning_rate": 1.1363636363636365e-05,
|
| 1418 |
-
"loss": 0.
|
| 1419 |
"step": 1850
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
"epoch": 31.0,
|
| 1423 |
"learning_rate": 1.0909090909090909e-05,
|
| 1424 |
-
"loss": 0.
|
| 1425 |
"step": 1860
|
| 1426 |
},
|
| 1427 |
{
|
| 1428 |
"epoch": 31.0,
|
| 1429 |
-
"eval_cer": 0.
|
| 1430 |
-
"eval_loss": 0.
|
| 1431 |
-
"eval_runtime": 7.
|
| 1432 |
-
"eval_samples_per_second":
|
| 1433 |
-
"eval_steps_per_second": 2.
|
| 1434 |
-
"eval_wer": 0.
|
| 1435 |
"step": 1860
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
"epoch": 31.17,
|
| 1439 |
"learning_rate": 1.0454545454545455e-05,
|
| 1440 |
-
"loss": 0.
|
| 1441 |
"step": 1870
|
| 1442 |
},
|
| 1443 |
{
|
| 1444 |
"epoch": 31.33,
|
| 1445 |
"learning_rate": 1e-05,
|
| 1446 |
-
"loss": 0.
|
| 1447 |
"step": 1880
|
| 1448 |
},
|
| 1449 |
{
|
| 1450 |
"epoch": 31.5,
|
| 1451 |
"learning_rate": 9.545454545454547e-06,
|
| 1452 |
-
"loss": 0.
|
| 1453 |
"step": 1890
|
| 1454 |
},
|
| 1455 |
{
|
| 1456 |
"epoch": 31.67,
|
| 1457 |
"learning_rate": 9.090909090909091e-06,
|
| 1458 |
-
"loss": 0.
|
| 1459 |
"step": 1900
|
| 1460 |
},
|
| 1461 |
{
|
| 1462 |
"epoch": 31.83,
|
| 1463 |
"learning_rate": 8.636363636363637e-06,
|
| 1464 |
-
"loss": 0.
|
| 1465 |
"step": 1910
|
| 1466 |
},
|
| 1467 |
{
|
| 1468 |
"epoch": 32.0,
|
| 1469 |
"learning_rate": 8.181818181818183e-06,
|
| 1470 |
-
"loss": 0.
|
| 1471 |
"step": 1920
|
| 1472 |
},
|
| 1473 |
{
|
| 1474 |
"epoch": 32.0,
|
| 1475 |
-
"eval_cer": 0.
|
| 1476 |
-
"eval_loss": 0.
|
| 1477 |
-
"eval_runtime": 7.
|
| 1478 |
-
"eval_samples_per_second":
|
| 1479 |
-
"eval_steps_per_second": 2.
|
| 1480 |
-
"eval_wer": 0.
|
| 1481 |
"step": 1920
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
"epoch": 32.17,
|
| 1485 |
"learning_rate": 7.727272727272727e-06,
|
| 1486 |
-
"loss": 0.
|
| 1487 |
"step": 1930
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
"epoch": 32.33,
|
| 1491 |
"learning_rate": 7.272727272727272e-06,
|
| 1492 |
-
"loss": 0.
|
| 1493 |
"step": 1940
|
| 1494 |
},
|
| 1495 |
{
|
| 1496 |
"epoch": 32.5,
|
| 1497 |
"learning_rate": 6.818181818181818e-06,
|
| 1498 |
-
"loss": 0.
|
| 1499 |
"step": 1950
|
| 1500 |
},
|
| 1501 |
{
|
| 1502 |
"epoch": 32.67,
|
| 1503 |
"learning_rate": 6.363636363636363e-06,
|
| 1504 |
-
"loss": 0.
|
| 1505 |
"step": 1960
|
| 1506 |
},
|
| 1507 |
{
|
| 1508 |
"epoch": 32.83,
|
| 1509 |
"learning_rate": 5.909090909090909e-06,
|
| 1510 |
-
"loss": 0.
|
| 1511 |
"step": 1970
|
| 1512 |
},
|
| 1513 |
{
|
| 1514 |
"epoch": 33.0,
|
| 1515 |
"learning_rate": 5.4545454545454545e-06,
|
| 1516 |
-
"loss": 0.
|
| 1517 |
"step": 1980
|
| 1518 |
},
|
| 1519 |
{
|
| 1520 |
"epoch": 33.0,
|
| 1521 |
-
"eval_cer": 0.
|
| 1522 |
-
"eval_loss": 0.
|
| 1523 |
-
"eval_runtime": 7.
|
| 1524 |
-
"eval_samples_per_second":
|
| 1525 |
-
"eval_steps_per_second": 2.
|
| 1526 |
-
"eval_wer": 0.
|
| 1527 |
"step": 1980
|
| 1528 |
},
|
| 1529 |
{
|
| 1530 |
"epoch": 33.17,
|
| 1531 |
"learning_rate": 5e-06,
|
| 1532 |
-
"loss": 0.
|
| 1533 |
"step": 1990
|
| 1534 |
},
|
| 1535 |
{
|
| 1536 |
"epoch": 33.33,
|
| 1537 |
"learning_rate": 4.5454545454545455e-06,
|
| 1538 |
-
"loss": 0.
|
| 1539 |
"step": 2000
|
| 1540 |
},
|
| 1541 |
{
|
| 1542 |
"epoch": 33.5,
|
| 1543 |
"learning_rate": 4.0909090909090915e-06,
|
| 1544 |
-
"loss": 0.
|
| 1545 |
"step": 2010
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
"epoch": 33.67,
|
| 1549 |
"learning_rate": 3.636363636363636e-06,
|
| 1550 |
-
"loss": 0.
|
| 1551 |
"step": 2020
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
"epoch": 33.83,
|
| 1555 |
"learning_rate": 3.1818181818181817e-06,
|
| 1556 |
-
"loss": 0.
|
| 1557 |
"step": 2030
|
| 1558 |
},
|
| 1559 |
{
|
| 1560 |
"epoch": 34.0,
|
| 1561 |
"learning_rate": 2.7272727272727272e-06,
|
| 1562 |
-
"loss": 0.
|
| 1563 |
"step": 2040
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
"epoch": 34.0,
|
| 1567 |
-
"eval_cer": 0.
|
| 1568 |
-
"eval_loss": 0.
|
| 1569 |
-
"eval_runtime": 7.
|
| 1570 |
-
"eval_samples_per_second":
|
| 1571 |
-
"eval_steps_per_second": 2.
|
| 1572 |
-
"eval_wer": 0.
|
| 1573 |
"step": 2040
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
"epoch": 34.17,
|
| 1577 |
"learning_rate": 2.2727272727272728e-06,
|
| 1578 |
-
"loss": 0.
|
| 1579 |
"step": 2050
|
| 1580 |
},
|
| 1581 |
{
|
| 1582 |
"epoch": 34.33,
|
| 1583 |
"learning_rate": 1.818181818181818e-06,
|
| 1584 |
-
"loss": 0.
|
| 1585 |
"step": 2060
|
| 1586 |
},
|
| 1587 |
{
|
| 1588 |
"epoch": 34.5,
|
| 1589 |
"learning_rate": 1.3636363636363636e-06,
|
| 1590 |
-
"loss": 0.
|
| 1591 |
"step": 2070
|
| 1592 |
},
|
| 1593 |
{
|
| 1594 |
"epoch": 34.67,
|
| 1595 |
"learning_rate": 9.09090909090909e-07,
|
| 1596 |
-
"loss": 0.
|
| 1597 |
"step": 2080
|
| 1598 |
},
|
| 1599 |
{
|
| 1600 |
"epoch": 34.83,
|
| 1601 |
"learning_rate": 4.545454545454545e-07,
|
| 1602 |
-
"loss": 0.
|
| 1603 |
"step": 2090
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
"epoch": 35.0,
|
| 1607 |
"learning_rate": 0.0,
|
| 1608 |
-
"loss": 0.
|
| 1609 |
"step": 2100
|
| 1610 |
},
|
| 1611 |
{
|
| 1612 |
"epoch": 35.0,
|
| 1613 |
-
"eval_cer": 0.
|
| 1614 |
-
"eval_loss": 0.
|
| 1615 |
-
"eval_runtime": 7.
|
| 1616 |
-
"eval_samples_per_second":
|
| 1617 |
-
"eval_steps_per_second": 2.
|
| 1618 |
-
"eval_wer": 0.
|
| 1619 |
"step": 2100
|
| 1620 |
},
|
| 1621 |
{
|
| 1622 |
"epoch": 35.0,
|
| 1623 |
"step": 2100,
|
| 1624 |
-
"total_flos": 5.
|
| 1625 |
-
"train_loss": 1.
|
| 1626 |
-
"train_runtime":
|
| 1627 |
-
"train_samples_per_second": 15.
|
| 1628 |
"train_steps_per_second": 0.497
|
| 1629 |
}
|
| 1630 |
],
|
|
@@ -1632,7 +1632,7 @@
|
|
| 1632 |
"max_steps": 2100,
|
| 1633 |
"num_train_epochs": 35,
|
| 1634 |
"save_steps": 500,
|
| 1635 |
-
"total_flos": 5.
|
| 1636 |
"trial_name": null,
|
| 1637 |
"trial_params": null
|
| 1638 |
}
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.17,
|
| 13 |
"learning_rate": 5.000000000000001e-07,
|
| 14 |
+
"loss": 17.2251,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.33,
|
| 19 |
"learning_rate": 1.0000000000000002e-06,
|
| 20 |
+
"loss": 17.329,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.5,
|
| 25 |
"learning_rate": 1.5e-06,
|
| 26 |
+
"loss": 16.6944,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.67,
|
| 31 |
"learning_rate": 2.0000000000000003e-06,
|
| 32 |
+
"loss": 16.5026,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.83,
|
| 37 |
"learning_rate": 2.5e-06,
|
| 38 |
+
"loss": 15.637,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 1.0,
|
| 43 |
"learning_rate": 3e-06,
|
| 44 |
+
"loss": 14.5102,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 1.0,
|
| 49 |
+
"eval_cer": 1.2966024915062289,
|
| 50 |
+
"eval_loss": 13.54184341430664,
|
| 51 |
+
"eval_runtime": 7.0367,
|
| 52 |
+
"eval_samples_per_second": 68.214,
|
| 53 |
+
"eval_steps_per_second": 2.132,
|
| 54 |
+
"eval_wer": 1.0011119347664936,
|
| 55 |
"step": 60
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.17,
|
| 59 |
"learning_rate": 3.5000000000000004e-06,
|
| 60 |
+
"loss": 13.6535,
|
| 61 |
"step": 70
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 1.33,
|
| 65 |
"learning_rate": 4.000000000000001e-06,
|
| 66 |
+
"loss": 11.923,
|
| 67 |
"step": 80
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 1.5,
|
| 71 |
"learning_rate": 4.5e-06,
|
| 72 |
+
"loss": 10.5358,
|
| 73 |
"step": 90
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 1.67,
|
| 77 |
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 8.2827,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.83,
|
| 83 |
"learning_rate": 5.500000000000001e-06,
|
| 84 |
+
"loss": 6.8578,
|
| 85 |
"step": 110
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"epoch": 2.0,
|
| 89 |
"learning_rate": 6e-06,
|
| 90 |
+
"loss": 6.2989,
|
| 91 |
"step": 120
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 2.0,
|
| 95 |
+
"eval_cer": 0.8736126840317101,
|
| 96 |
+
"eval_loss": 4.612649917602539,
|
| 97 |
+
"eval_runtime": 7.5054,
|
| 98 |
+
"eval_samples_per_second": 63.954,
|
| 99 |
+
"eval_steps_per_second": 1.999,
|
| 100 |
+
"eval_wer": 1.0,
|
| 101 |
"step": 120
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 2.17,
|
| 105 |
"learning_rate": 6.5000000000000004e-06,
|
| 106 |
+
"loss": 4.6838,
|
| 107 |
"step": 130
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 2.33,
|
| 111 |
"learning_rate": 7.000000000000001e-06,
|
| 112 |
+
"loss": 4.0318,
|
| 113 |
"step": 140
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 2.5,
|
| 117 |
"learning_rate": 7.5e-06,
|
| 118 |
+
"loss": 3.2615,
|
| 119 |
"step": 150
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 2.67,
|
| 123 |
"learning_rate": 8.000000000000001e-06,
|
| 124 |
+
"loss": 2.8846,
|
| 125 |
"step": 160
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 2.83,
|
| 129 |
"learning_rate": 8.500000000000002e-06,
|
| 130 |
+
"loss": 2.5007,
|
| 131 |
"step": 170
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.0,
|
| 135 |
"learning_rate": 9e-06,
|
| 136 |
+
"loss": 2.208,
|
| 137 |
"step": 180
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 3.0,
|
| 141 |
+
"eval_cer": 0.8159682899207248,
|
| 142 |
+
"eval_loss": 1.7111728191375732,
|
| 143 |
+
"eval_runtime": 7.0161,
|
| 144 |
+
"eval_samples_per_second": 68.414,
|
| 145 |
+
"eval_steps_per_second": 2.138,
|
| 146 |
+
"eval_wer": 0.9992587101556709,
|
| 147 |
"step": 180
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 3.17,
|
| 151 |
"learning_rate": 9.5e-06,
|
| 152 |
+
"loss": 1.8824,
|
| 153 |
"step": 190
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 3.33,
|
| 157 |
"learning_rate": 1e-05,
|
| 158 |
+
"loss": 1.6231,
|
| 159 |
"step": 200
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 3.5,
|
| 163 |
"learning_rate": 1.05e-05,
|
| 164 |
+
"loss": 1.5416,
|
| 165 |
"step": 210
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 3.67,
|
| 169 |
"learning_rate": 1.1000000000000001e-05,
|
| 170 |
+
"loss": 1.3219,
|
| 171 |
"step": 220
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 3.83,
|
| 175 |
"learning_rate": 1.1500000000000002e-05,
|
| 176 |
+
"loss": 1.1937,
|
| 177 |
"step": 230
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 4.0,
|
| 181 |
"learning_rate": 1.2e-05,
|
| 182 |
+
"loss": 1.098,
|
| 183 |
"step": 240
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 4.0,
|
| 187 |
+
"eval_cer": 0.569762174405436,
|
| 188 |
+
"eval_loss": 0.8765236139297485,
|
| 189 |
+
"eval_runtime": 7.0276,
|
| 190 |
+
"eval_samples_per_second": 68.302,
|
| 191 |
+
"eval_steps_per_second": 2.134,
|
| 192 |
+
"eval_wer": 0.8409933283914011,
|
| 193 |
"step": 240
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 4.17,
|
| 197 |
"learning_rate": 1.25e-05,
|
| 198 |
+
"loss": 1.1276,
|
| 199 |
"step": 250
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 4.33,
|
| 203 |
"learning_rate": 1.3000000000000001e-05,
|
| 204 |
+
"loss": 0.9733,
|
| 205 |
"step": 260
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 4.5,
|
| 209 |
"learning_rate": 1.3500000000000001e-05,
|
| 210 |
+
"loss": 0.9508,
|
| 211 |
"step": 270
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 4.67,
|
| 215 |
"learning_rate": 1.4000000000000001e-05,
|
| 216 |
+
"loss": 0.8988,
|
| 217 |
"step": 280
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 4.83,
|
| 221 |
"learning_rate": 1.45e-05,
|
| 222 |
+
"loss": 0.8499,
|
| 223 |
"step": 290
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 5.0,
|
| 227 |
"learning_rate": 1.5e-05,
|
| 228 |
+
"loss": 0.8201,
|
| 229 |
"step": 300
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 5.0,
|
| 233 |
+
"eval_cer": 0.5339750849377124,
|
| 234 |
+
"eval_loss": 0.6964037418365479,
|
| 235 |
+
"eval_runtime": 7.0359,
|
| 236 |
+
"eval_samples_per_second": 68.221,
|
| 237 |
+
"eval_steps_per_second": 2.132,
|
| 238 |
+
"eval_wer": 0.8220904373610082,
|
| 239 |
"step": 300
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 5.17,
|
| 243 |
"learning_rate": 1.55e-05,
|
| 244 |
+
"loss": 0.7975,
|
| 245 |
"step": 310
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 5.33,
|
| 249 |
"learning_rate": 1.6000000000000003e-05,
|
| 250 |
+
"loss": 0.7687,
|
| 251 |
"step": 320
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 5.5,
|
| 255 |
"learning_rate": 1.65e-05,
|
| 256 |
+
"loss": 0.7861,
|
| 257 |
"step": 330
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 5.67,
|
| 261 |
"learning_rate": 1.7000000000000003e-05,
|
| 262 |
+
"loss": 0.7617,
|
| 263 |
"step": 340
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 5.83,
|
| 267 |
"learning_rate": 1.75e-05,
|
| 268 |
+
"loss": 0.7434,
|
| 269 |
"step": 350
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 6.0,
|
| 273 |
"learning_rate": 1.8e-05,
|
| 274 |
+
"loss": 0.7499,
|
| 275 |
"step": 360
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 6.0,
|
| 279 |
+
"eval_cer": 0.5304643261608154,
|
| 280 |
+
"eval_loss": 0.6299313902854919,
|
| 281 |
+
"eval_runtime": 7.057,
|
| 282 |
+
"eval_samples_per_second": 68.018,
|
| 283 |
+
"eval_steps_per_second": 2.126,
|
| 284 |
+
"eval_wer": 0.8217197924388436,
|
| 285 |
"step": 360
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 6.17,
|
| 289 |
"learning_rate": 1.85e-05,
|
| 290 |
+
"loss": 0.7481,
|
| 291 |
"step": 370
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 6.33,
|
| 295 |
"learning_rate": 1.9e-05,
|
| 296 |
+
"loss": 0.6974,
|
| 297 |
"step": 380
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 6.5,
|
| 301 |
"learning_rate": 1.9500000000000003e-05,
|
| 302 |
+
"loss": 0.7176,
|
| 303 |
"step": 390
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 6.67,
|
| 307 |
"learning_rate": 2e-05,
|
| 308 |
+
"loss": 0.6873,
|
| 309 |
"step": 400
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 6.83,
|
| 313 |
"learning_rate": 2.05e-05,
|
| 314 |
+
"loss": 0.6699,
|
| 315 |
"step": 410
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 7.0,
|
| 319 |
"learning_rate": 2.1e-05,
|
| 320 |
+
"loss": 0.6753,
|
| 321 |
"step": 420
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 7.0,
|
| 325 |
+
"eval_cer": 0.4482446206115515,
|
| 326 |
+
"eval_loss": 0.5997506976127625,
|
| 327 |
+
"eval_runtime": 7.0307,
|
| 328 |
+
"eval_samples_per_second": 68.272,
|
| 329 |
+
"eval_steps_per_second": 2.134,
|
| 330 |
+
"eval_wer": 0.7690882134914752,
|
| 331 |
"step": 420
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 7.17,
|
| 335 |
"learning_rate": 2.15e-05,
|
| 336 |
+
"loss": 0.6553,
|
| 337 |
"step": 430
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 7.33,
|
| 341 |
"learning_rate": 2.2000000000000003e-05,
|
| 342 |
+
"loss": 0.6329,
|
| 343 |
"step": 440
|
| 344 |
},
|
| 345 |
{
|
| 346 |
"epoch": 7.5,
|
| 347 |
"learning_rate": 2.25e-05,
|
| 348 |
+
"loss": 0.6416,
|
| 349 |
"step": 450
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 7.67,
|
| 353 |
"learning_rate": 2.3000000000000003e-05,
|
| 354 |
+
"loss": 0.6469,
|
| 355 |
"step": 460
|
| 356 |
},
|
| 357 |
{
|
| 358 |
"epoch": 7.83,
|
| 359 |
"learning_rate": 2.35e-05,
|
| 360 |
+
"loss": 0.6278,
|
| 361 |
"step": 470
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 8.0,
|
| 365 |
"learning_rate": 2.4e-05,
|
| 366 |
+
"loss": 0.6003,
|
| 367 |
"step": 480
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 8.0,
|
| 371 |
+
"eval_cer": 0.4563986409966025,
|
| 372 |
+
"eval_loss": 0.5502179861068726,
|
| 373 |
+
"eval_runtime": 7.0474,
|
| 374 |
+
"eval_samples_per_second": 68.11,
|
| 375 |
+
"eval_steps_per_second": 2.128,
|
| 376 |
+
"eval_wer": 0.7394366197183099,
|
| 377 |
"step": 480
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 8.17,
|
| 381 |
"learning_rate": 2.45e-05,
|
| 382 |
+
"loss": 0.6474,
|
| 383 |
"step": 490
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 8.33,
|
| 387 |
"learning_rate": 2.5e-05,
|
| 388 |
+
"loss": 0.5815,
|
| 389 |
"step": 500
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 8.5,
|
| 393 |
"learning_rate": 2.5500000000000003e-05,
|
| 394 |
+
"loss": 0.5954,
|
| 395 |
"step": 510
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 8.67,
|
| 399 |
"learning_rate": 2.6000000000000002e-05,
|
| 400 |
+
"loss": 0.5791,
|
| 401 |
"step": 520
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 8.83,
|
| 405 |
"learning_rate": 2.6500000000000004e-05,
|
| 406 |
+
"loss": 0.5805,
|
| 407 |
"step": 530
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 9.0,
|
| 411 |
"learning_rate": 2.7000000000000002e-05,
|
| 412 |
+
"loss": 0.5732,
|
| 413 |
"step": 540
|
| 414 |
},
|
| 415 |
{
|
| 416 |
"epoch": 9.0,
|
| 417 |
+
"eval_cer": 0.39060022650056625,
|
| 418 |
+
"eval_loss": 0.5047246217727661,
|
| 419 |
+
"eval_runtime": 7.0721,
|
| 420 |
+
"eval_samples_per_second": 67.872,
|
| 421 |
+
"eval_steps_per_second": 2.121,
|
| 422 |
+
"eval_wer": 0.7097850259451446,
|
| 423 |
"step": 540
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 9.17,
|
| 427 |
"learning_rate": 2.7500000000000004e-05,
|
| 428 |
+
"loss": 0.5427,
|
| 429 |
"step": 550
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 9.33,
|
| 433 |
"learning_rate": 2.8000000000000003e-05,
|
| 434 |
+
"loss": 0.5792,
|
| 435 |
"step": 560
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 9.5,
|
| 439 |
"learning_rate": 2.8499999999999998e-05,
|
| 440 |
+
"loss": 0.5728,
|
| 441 |
"step": 570
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 9.67,
|
| 445 |
"learning_rate": 2.9e-05,
|
| 446 |
+
"loss": 0.5513,
|
| 447 |
"step": 580
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 9.83,
|
| 451 |
"learning_rate": 2.95e-05,
|
| 452 |
+
"loss": 0.5349,
|
| 453 |
"step": 590
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 10.0,
|
| 457 |
"learning_rate": 3e-05,
|
| 458 |
+
"loss": 0.5404,
|
| 459 |
"step": 600
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 10.0,
|
| 463 |
+
"eval_cer": 0.32831257078142695,
|
| 464 |
+
"eval_loss": 0.4693681299686432,
|
| 465 |
+
"eval_runtime": 7.0538,
|
| 466 |
+
"eval_samples_per_second": 68.048,
|
| 467 |
+
"eval_steps_per_second": 2.127,
|
| 468 |
+
"eval_wer": 0.6679021497405485,
|
| 469 |
"step": 600
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 10.17,
|
| 473 |
"learning_rate": 3.05e-05,
|
| 474 |
+
"loss": 0.5152,
|
| 475 |
"step": 610
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 10.33,
|
| 479 |
"learning_rate": 3.1e-05,
|
| 480 |
+
"loss": 0.5459,
|
| 481 |
"step": 620
|
| 482 |
},
|
| 483 |
{
|
| 484 |
"epoch": 10.5,
|
| 485 |
"learning_rate": 3.15e-05,
|
| 486 |
+
"loss": 0.5283,
|
| 487 |
"step": 630
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 10.67,
|
| 491 |
"learning_rate": 3.2000000000000005e-05,
|
| 492 |
+
"loss": 0.5099,
|
| 493 |
"step": 640
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 10.83,
|
| 497 |
"learning_rate": 3.2500000000000004e-05,
|
| 498 |
+
"loss": 0.5053,
|
| 499 |
"step": 650
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 11.0,
|
| 503 |
"learning_rate": 3.3e-05,
|
| 504 |
+
"loss": 0.4889,
|
| 505 |
"step": 660
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 11.0,
|
| 509 |
+
"eval_cer": 0.30169875424688564,
|
| 510 |
+
"eval_loss": 0.3979022204875946,
|
| 511 |
+
"eval_runtime": 7.0584,
|
| 512 |
+
"eval_samples_per_second": 68.004,
|
| 513 |
+
"eval_steps_per_second": 2.125,
|
| 514 |
+
"eval_wer": 0.6378799110452187,
|
| 515 |
"step": 660
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 11.17,
|
| 519 |
"learning_rate": 3.35e-05,
|
| 520 |
+
"loss": 0.4605,
|
| 521 |
"step": 670
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 11.33,
|
| 525 |
"learning_rate": 3.4000000000000007e-05,
|
| 526 |
+
"loss": 0.4497,
|
| 527 |
"step": 680
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 11.5,
|
| 531 |
"learning_rate": 3.45e-05,
|
| 532 |
+
"loss": 0.4669,
|
| 533 |
"step": 690
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 11.67,
|
| 537 |
"learning_rate": 3.5e-05,
|
| 538 |
+
"loss": 0.4863,
|
| 539 |
"step": 700
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 11.83,
|
| 543 |
"learning_rate": 3.55e-05,
|
| 544 |
+
"loss": 0.4751,
|
| 545 |
"step": 710
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 12.0,
|
| 549 |
"learning_rate": 3.6e-05,
|
| 550 |
+
"loss": 0.4401,
|
| 551 |
"step": 720
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 12.0,
|
| 555 |
+
"eval_cer": 0.27916194790486976,
|
| 556 |
+
"eval_loss": 0.3254798352718353,
|
| 557 |
+
"eval_runtime": 7.0501,
|
| 558 |
+
"eval_samples_per_second": 68.085,
|
| 559 |
+
"eval_steps_per_second": 2.128,
|
| 560 |
+
"eval_wer": 0.5848776871756857,
|
| 561 |
"step": 720
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 12.17,
|
| 565 |
"learning_rate": 3.65e-05,
|
| 566 |
+
"loss": 0.4469,
|
| 567 |
"step": 730
|
| 568 |
},
|
| 569 |
{
|
| 570 |
"epoch": 12.33,
|
| 571 |
"learning_rate": 3.7e-05,
|
| 572 |
+
"loss": 0.4009,
|
| 573 |
"step": 740
|
| 574 |
},
|
| 575 |
{
|
| 576 |
"epoch": 12.5,
|
| 577 |
"learning_rate": 3.7500000000000003e-05,
|
| 578 |
+
"loss": 0.3882,
|
| 579 |
"step": 750
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 12.67,
|
| 583 |
"learning_rate": 3.8e-05,
|
| 584 |
+
"loss": 0.3879,
|
| 585 |
"step": 760
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 12.83,
|
| 589 |
"learning_rate": 3.85e-05,
|
| 590 |
+
"loss": 0.4118,
|
| 591 |
"step": 770
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 13.0,
|
| 595 |
"learning_rate": 3.9000000000000006e-05,
|
| 596 |
+
"loss": 0.4295,
|
| 597 |
"step": 780
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 13.0,
|
| 601 |
+
"eval_cer": 0.2772366930917327,
|
| 602 |
+
"eval_loss": 0.2853207290172577,
|
| 603 |
+
"eval_runtime": 7.0684,
|
| 604 |
+
"eval_samples_per_second": 67.908,
|
| 605 |
+
"eval_steps_per_second": 2.122,
|
| 606 |
+
"eval_wer": 0.5044477390659748,
|
| 607 |
"step": 780
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 13.17,
|
| 611 |
"learning_rate": 3.9500000000000005e-05,
|
| 612 |
+
"loss": 0.3879,
|
| 613 |
"step": 790
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"epoch": 13.33,
|
| 617 |
"learning_rate": 4e-05,
|
| 618 |
+
"loss": 0.3465,
|
| 619 |
"step": 800
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 13.5,
|
| 623 |
"learning_rate": 4.05e-05,
|
| 624 |
+
"loss": 0.3395,
|
| 625 |
"step": 810
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 13.67,
|
| 629 |
"learning_rate": 4.1e-05,
|
| 630 |
+
"loss": 0.3304,
|
| 631 |
"step": 820
|
| 632 |
},
|
| 633 |
{
|
| 634 |
"epoch": 13.83,
|
| 635 |
"learning_rate": 4.15e-05,
|
| 636 |
+
"loss": 0.3393,
|
| 637 |
"step": 830
|
| 638 |
},
|
| 639 |
{
|
| 640 |
"epoch": 14.0,
|
| 641 |
"learning_rate": 4.2e-05,
|
| 642 |
+
"loss": 0.3216,
|
| 643 |
"step": 840
|
| 644 |
},
|
| 645 |
{
|
| 646 |
"epoch": 14.0,
|
| 647 |
+
"eval_cer": 0.2234428086070215,
|
| 648 |
+
"eval_loss": 0.22039471566677094,
|
| 649 |
+
"eval_runtime": 7.0253,
|
| 650 |
+
"eval_samples_per_second": 68.325,
|
| 651 |
+
"eval_steps_per_second": 2.135,
|
| 652 |
+
"eval_wer": 0.45107487027427723,
|
| 653 |
"step": 840
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 14.17,
|
| 657 |
"learning_rate": 4.25e-05,
|
| 658 |
+
"loss": 0.3149,
|
| 659 |
"step": 850
|
| 660 |
},
|
| 661 |
{
|
| 662 |
"epoch": 14.33,
|
| 663 |
"learning_rate": 4.3e-05,
|
| 664 |
+
"loss": 0.3068,
|
| 665 |
"step": 860
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 14.5,
|
| 669 |
"learning_rate": 4.35e-05,
|
| 670 |
+
"loss": 0.2827,
|
| 671 |
"step": 870
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 14.67,
|
| 675 |
"learning_rate": 4.4000000000000006e-05,
|
| 676 |
+
"loss": 0.3114,
|
| 677 |
"step": 880
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 14.83,
|
| 681 |
"learning_rate": 4.4500000000000004e-05,
|
| 682 |
+
"loss": 0.3113,
|
| 683 |
"step": 890
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 15.0,
|
| 687 |
"learning_rate": 4.5e-05,
|
| 688 |
+
"loss": 0.2583,
|
| 689 |
"step": 900
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 15.0,
|
| 693 |
+
"eval_cer": 0.2130237825594564,
|
| 694 |
+
"eval_loss": 0.14919017255306244,
|
| 695 |
+
"eval_runtime": 7.0711,
|
| 696 |
+
"eval_samples_per_second": 67.882,
|
| 697 |
+
"eval_steps_per_second": 2.121,
|
| 698 |
+
"eval_wer": 0.39288361749444034,
|
| 699 |
"step": 900
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 15.17,
|
| 703 |
"learning_rate": 4.55e-05,
|
| 704 |
+
"loss": 0.275,
|
| 705 |
"step": 910
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 15.33,
|
| 709 |
"learning_rate": 4.600000000000001e-05,
|
| 710 |
+
"loss": 0.2363,
|
| 711 |
"step": 920
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 15.5,
|
| 715 |
"learning_rate": 4.6500000000000005e-05,
|
| 716 |
+
"loss": 0.246,
|
| 717 |
"step": 930
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 15.67,
|
| 721 |
"learning_rate": 4.7e-05,
|
| 722 |
+
"loss": 0.2328,
|
| 723 |
"step": 940
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 15.83,
|
| 727 |
"learning_rate": 4.75e-05,
|
| 728 |
+
"loss": 0.2271,
|
| 729 |
"step": 950
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 16.0,
|
| 733 |
"learning_rate": 4.8e-05,
|
| 734 |
+
"loss": 0.226,
|
| 735 |
"step": 960
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 16.0,
|
| 739 |
+
"eval_cer": 0.16896942242355606,
|
| 740 |
+
"eval_loss": 0.10074901580810547,
|
| 741 |
+
"eval_runtime": 7.0789,
|
| 742 |
+
"eval_samples_per_second": 67.807,
|
| 743 |
+
"eval_steps_per_second": 2.119,
|
| 744 |
+
"eval_wer": 0.28169014084507044,
|
| 745 |
"step": 960
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 16.17,
|
| 749 |
"learning_rate": 4.85e-05,
|
| 750 |
+
"loss": 0.1883,
|
| 751 |
"step": 970
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 16.33,
|
| 755 |
"learning_rate": 4.9e-05,
|
| 756 |
+
"loss": 0.1925,
|
| 757 |
"step": 980
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 16.5,
|
| 761 |
"learning_rate": 4.9500000000000004e-05,
|
| 762 |
+
"loss": 0.2024,
|
| 763 |
"step": 990
|
| 764 |
},
|
| 765 |
{
|
| 766 |
"epoch": 16.67,
|
| 767 |
"learning_rate": 5e-05,
|
| 768 |
+
"loss": 0.1988,
|
| 769 |
"step": 1000
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 16.83,
|
| 773 |
"learning_rate": 4.9545454545454553e-05,
|
| 774 |
+
"loss": 0.1694,
|
| 775 |
"step": 1010
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 17.0,
|
| 779 |
"learning_rate": 4.909090909090909e-05,
|
| 780 |
+
"loss": 0.2304,
|
| 781 |
"step": 1020
|
| 782 |
},
|
| 783 |
{
|
| 784 |
"epoch": 17.0,
|
| 785 |
+
"eval_cer": 0.16987542468856173,
|
| 786 |
+
"eval_loss": 0.06942722201347351,
|
| 787 |
+
"eval_runtime": 7.0727,
|
| 788 |
+
"eval_samples_per_second": 67.866,
|
| 789 |
+
"eval_steps_per_second": 2.121,
|
| 790 |
+
"eval_wer": 0.24388435878428466,
|
| 791 |
"step": 1020
|
| 792 |
},
|
| 793 |
{
|
| 794 |
"epoch": 17.17,
|
| 795 |
"learning_rate": 4.863636363636364e-05,
|
| 796 |
+
"loss": 0.2023,
|
| 797 |
"step": 1030
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 17.33,
|
| 801 |
"learning_rate": 4.8181818181818186e-05,
|
| 802 |
+
"loss": 0.1851,
|
| 803 |
"step": 1040
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 17.5,
|
| 807 |
"learning_rate": 4.772727272727273e-05,
|
| 808 |
+
"loss": 0.1517,
|
| 809 |
"step": 1050
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 17.67,
|
| 813 |
"learning_rate": 4.7272727272727275e-05,
|
| 814 |
+
"loss": 0.2016,
|
| 815 |
"step": 1060
|
| 816 |
},
|
| 817 |
{
|
| 818 |
"epoch": 17.83,
|
| 819 |
"learning_rate": 4.681818181818182e-05,
|
| 820 |
+
"loss": 0.1894,
|
| 821 |
"step": 1070
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 18.0,
|
| 825 |
"learning_rate": 4.636363636363636e-05,
|
| 826 |
+
"loss": 0.1487,
|
| 827 |
"step": 1080
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 18.0,
|
| 831 |
+
"eval_cer": 0.17576443941109854,
|
| 832 |
+
"eval_loss": 0.047105852514505386,
|
| 833 |
+
"eval_runtime": 7.0679,
|
| 834 |
+
"eval_samples_per_second": 67.913,
|
| 835 |
+
"eval_steps_per_second": 2.122,
|
| 836 |
+
"eval_wer": 0.21423276501111935,
|
| 837 |
"step": 1080
|
| 838 |
},
|
| 839 |
{
|
| 840 |
"epoch": 18.17,
|
| 841 |
"learning_rate": 4.5909090909090914e-05,
|
| 842 |
+
"loss": 0.2165,
|
| 843 |
"step": 1090
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 18.33,
|
| 847 |
"learning_rate": 4.545454545454546e-05,
|
| 848 |
+
"loss": 0.1735,
|
| 849 |
"step": 1100
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 18.5,
|
| 853 |
"learning_rate": 4.5e-05,
|
| 854 |
+
"loss": 0.1464,
|
| 855 |
"step": 1110
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 18.67,
|
| 859 |
"learning_rate": 4.454545454545455e-05,
|
| 860 |
+
"loss": 0.162,
|
| 861 |
"step": 1120
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 18.83,
|
| 865 |
"learning_rate": 4.409090909090909e-05,
|
| 866 |
+
"loss": 0.1301,
|
| 867 |
"step": 1130
|
| 868 |
},
|
| 869 |
{
|
| 870 |
"epoch": 19.0,
|
| 871 |
"learning_rate": 4.3636363636363636e-05,
|
| 872 |
+
"loss": 0.1045,
|
| 873 |
"step": 1140
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 19.0,
|
| 877 |
+
"eval_cer": 0.16862967157417894,
|
| 878 |
+
"eval_loss": 0.030526038259267807,
|
| 879 |
+
"eval_runtime": 7.0581,
|
| 880 |
+
"eval_samples_per_second": 68.007,
|
| 881 |
+
"eval_steps_per_second": 2.125,
|
| 882 |
+
"eval_wer": 0.2168272794662713,
|
| 883 |
"step": 1140
|
| 884 |
},
|
| 885 |
{
|
| 886 |
"epoch": 19.17,
|
| 887 |
"learning_rate": 4.318181818181819e-05,
|
| 888 |
+
"loss": 0.1027,
|
| 889 |
"step": 1150
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 19.33,
|
| 893 |
"learning_rate": 4.2727272727272724e-05,
|
| 894 |
+
"loss": 0.1608,
|
| 895 |
"step": 1160
|
| 896 |
},
|
| 897 |
{
|
| 898 |
"epoch": 19.5,
|
| 899 |
"learning_rate": 4.2272727272727275e-05,
|
| 900 |
+
"loss": 0.116,
|
| 901 |
"step": 1170
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 19.67,
|
| 905 |
"learning_rate": 4.181818181818182e-05,
|
| 906 |
+
"loss": 0.119,
|
| 907 |
"step": 1180
|
| 908 |
},
|
| 909 |
{
|
| 910 |
"epoch": 19.83,
|
| 911 |
"learning_rate": 4.1363636363636364e-05,
|
| 912 |
+
"loss": 0.1125,
|
| 913 |
"step": 1190
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 20.0,
|
| 917 |
"learning_rate": 4.0909090909090915e-05,
|
| 918 |
+
"loss": 0.1104,
|
| 919 |
"step": 1200
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 20.0,
|
| 923 |
+
"eval_cer": 0.1625141562853907,
|
| 924 |
+
"eval_loss": 0.02558966353535652,
|
| 925 |
+
"eval_runtime": 7.0898,
|
| 926 |
+
"eval_samples_per_second": 67.703,
|
| 927 |
+
"eval_steps_per_second": 2.116,
|
| 928 |
+
"eval_wer": 0.2071905114899926,
|
| 929 |
"step": 1200
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 20.17,
|
| 933 |
"learning_rate": 4.045454545454546e-05,
|
| 934 |
+
"loss": 0.0967,
|
| 935 |
"step": 1210
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 20.33,
|
| 939 |
"learning_rate": 4e-05,
|
| 940 |
+
"loss": 0.1141,
|
| 941 |
"step": 1220
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 20.5,
|
| 945 |
"learning_rate": 3.954545454545455e-05,
|
| 946 |
+
"loss": 0.1222,
|
| 947 |
"step": 1230
|
| 948 |
},
|
| 949 |
{
|
| 950 |
"epoch": 20.67,
|
| 951 |
"learning_rate": 3.909090909090909e-05,
|
| 952 |
+
"loss": 0.0991,
|
| 953 |
"step": 1240
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 20.83,
|
| 957 |
"learning_rate": 3.8636363636363636e-05,
|
| 958 |
+
"loss": 0.0998,
|
| 959 |
"step": 1250
|
| 960 |
},
|
| 961 |
{
|
| 962 |
"epoch": 21.0,
|
| 963 |
"learning_rate": 3.818181818181819e-05,
|
| 964 |
+
"loss": 0.094,
|
| 965 |
"step": 1260
|
| 966 |
},
|
| 967 |
{
|
| 968 |
"epoch": 21.0,
|
| 969 |
+
"eval_cer": 0.17599093997734994,
|
| 970 |
+
"eval_loss": 0.022575559094548225,
|
| 971 |
+
"eval_runtime": 7.0875,
|
| 972 |
+
"eval_samples_per_second": 67.725,
|
| 973 |
+
"eval_steps_per_second": 2.116,
|
| 974 |
+
"eval_wer": 0.22720533728687917,
|
| 975 |
"step": 1260
|
| 976 |
},
|
| 977 |
{
|
| 978 |
"epoch": 21.17,
|
| 979 |
"learning_rate": 3.7727272727272725e-05,
|
| 980 |
+
"loss": 0.0912,
|
| 981 |
"step": 1270
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 21.33,
|
| 985 |
"learning_rate": 3.7272727272727276e-05,
|
| 986 |
+
"loss": 0.0801,
|
| 987 |
"step": 1280
|
| 988 |
},
|
| 989 |
{
|
| 990 |
"epoch": 21.5,
|
| 991 |
"learning_rate": 3.681818181818182e-05,
|
| 992 |
+
"loss": 0.0872,
|
| 993 |
"step": 1290
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 21.67,
|
| 997 |
"learning_rate": 3.6363636363636364e-05,
|
| 998 |
+
"loss": 0.0801,
|
| 999 |
"step": 1300
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
"epoch": 21.83,
|
| 1003 |
"learning_rate": 3.590909090909091e-05,
|
| 1004 |
+
"loss": 0.0882,
|
| 1005 |
"step": 1310
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
"epoch": 22.0,
|
| 1009 |
"learning_rate": 3.545454545454546e-05,
|
| 1010 |
+
"loss": 0.0987,
|
| 1011 |
"step": 1320
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
"epoch": 22.0,
|
| 1015 |
+
"eval_cer": 0.19003397508493772,
|
| 1016 |
+
"eval_loss": 0.012922757305204868,
|
| 1017 |
+
"eval_runtime": 7.0893,
|
| 1018 |
+
"eval_samples_per_second": 67.708,
|
| 1019 |
+
"eval_steps_per_second": 2.116,
|
| 1020 |
+
"eval_wer": 0.20126019273535953,
|
| 1021 |
"step": 1320
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 22.17,
|
| 1025 |
"learning_rate": 3.5e-05,
|
| 1026 |
+
"loss": 0.0917,
|
| 1027 |
"step": 1330
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
"epoch": 22.33,
|
| 1031 |
"learning_rate": 3.454545454545455e-05,
|
| 1032 |
+
"loss": 0.0725,
|
| 1033 |
"step": 1340
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 22.5,
|
| 1037 |
"learning_rate": 3.409090909090909e-05,
|
| 1038 |
+
"loss": 0.102,
|
| 1039 |
"step": 1350
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
"epoch": 22.67,
|
| 1043 |
"learning_rate": 3.3636363636363636e-05,
|
| 1044 |
+
"loss": 0.0645,
|
| 1045 |
"step": 1360
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
"epoch": 22.83,
|
| 1049 |
"learning_rate": 3.318181818181819e-05,
|
| 1050 |
+
"loss": 0.069,
|
| 1051 |
"step": 1370
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 23.0,
|
| 1055 |
"learning_rate": 3.272727272727273e-05,
|
| 1056 |
+
"loss": 0.0753,
|
| 1057 |
"step": 1380
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 23.0,
|
| 1061 |
+
"eval_cer": 0.17859569648924123,
|
| 1062 |
+
"eval_loss": 0.011034397408366203,
|
| 1063 |
+
"eval_runtime": 7.0845,
|
| 1064 |
+
"eval_samples_per_second": 67.754,
|
| 1065 |
+
"eval_steps_per_second": 2.117,
|
| 1066 |
+
"eval_wer": 0.20533728687916974,
|
| 1067 |
"step": 1380
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
"epoch": 23.17,
|
| 1071 |
"learning_rate": 3.2272727272727276e-05,
|
| 1072 |
+
"loss": 0.067,
|
| 1073 |
"step": 1390
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 23.33,
|
| 1077 |
"learning_rate": 3.181818181818182e-05,
|
| 1078 |
+
"loss": 0.0618,
|
| 1079 |
"step": 1400
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 23.5,
|
| 1083 |
"learning_rate": 3.1363636363636365e-05,
|
| 1084 |
+
"loss": 0.0865,
|
| 1085 |
"step": 1410
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 23.67,
|
| 1089 |
"learning_rate": 3.090909090909091e-05,
|
| 1090 |
+
"loss": 0.0777,
|
| 1091 |
"step": 1420
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
"epoch": 23.83,
|
| 1095 |
"learning_rate": 3.0454545454545456e-05,
|
| 1096 |
+
"loss": 0.0495,
|
| 1097 |
"step": 1430
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
"epoch": 24.0,
|
| 1101 |
"learning_rate": 3e-05,
|
| 1102 |
+
"loss": 0.0544,
|
| 1103 |
"step": 1440
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
"epoch": 24.0,
|
| 1107 |
+
"eval_cer": 0.18584371460928653,
|
| 1108 |
+
"eval_loss": 0.009088627062737942,
|
| 1109 |
+
"eval_runtime": 7.0963,
|
| 1110 |
+
"eval_samples_per_second": 67.641,
|
| 1111 |
+
"eval_steps_per_second": 2.114,
|
| 1112 |
+
"eval_wer": 0.19088213491475167,
|
| 1113 |
"step": 1440
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 24.17,
|
| 1117 |
"learning_rate": 2.954545454545455e-05,
|
| 1118 |
+
"loss": 0.0697,
|
| 1119 |
"step": 1450
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 24.33,
|
| 1123 |
"learning_rate": 2.909090909090909e-05,
|
| 1124 |
+
"loss": 0.0652,
|
| 1125 |
"step": 1460
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
"epoch": 24.5,
|
| 1129 |
"learning_rate": 2.863636363636364e-05,
|
| 1130 |
+
"loss": 0.0608,
|
| 1131 |
"step": 1470
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
"epoch": 24.67,
|
| 1135 |
"learning_rate": 2.818181818181818e-05,
|
| 1136 |
+
"loss": 0.0617,
|
| 1137 |
"step": 1480
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
"epoch": 24.83,
|
| 1141 |
"learning_rate": 2.772727272727273e-05,
|
| 1142 |
+
"loss": 0.1019,
|
| 1143 |
"step": 1490
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 25.0,
|
| 1147 |
"learning_rate": 2.7272727272727273e-05,
|
| 1148 |
+
"loss": 0.0684,
|
| 1149 |
"step": 1500
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
"epoch": 25.0,
|
| 1153 |
+
"eval_cer": 0.17281993204983012,
|
| 1154 |
+
"eval_loss": 0.008332420140504837,
|
| 1155 |
+
"eval_runtime": 7.0855,
|
| 1156 |
+
"eval_samples_per_second": 67.744,
|
| 1157 |
+
"eval_steps_per_second": 2.117,
|
| 1158 |
+
"eval_wer": 0.19014084507042253,
|
| 1159 |
"step": 1500
|
| 1160 |
},
|
| 1161 |
{
|
| 1162 |
"epoch": 25.17,
|
| 1163 |
"learning_rate": 2.681818181818182e-05,
|
| 1164 |
+
"loss": 0.0579,
|
| 1165 |
"step": 1510
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
"epoch": 25.33,
|
| 1169 |
"learning_rate": 2.636363636363636e-05,
|
| 1170 |
+
"loss": 0.0733,
|
| 1171 |
"step": 1520
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 25.5,
|
| 1175 |
"learning_rate": 2.590909090909091e-05,
|
| 1176 |
+
"loss": 0.0514,
|
| 1177 |
"step": 1530
|
| 1178 |
},
|
| 1179 |
{
|
| 1180 |
"epoch": 25.67,
|
| 1181 |
"learning_rate": 2.5454545454545454e-05,
|
| 1182 |
+
"loss": 0.0597,
|
| 1183 |
"step": 1540
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
"epoch": 25.83,
|
| 1187 |
"learning_rate": 2.5e-05,
|
| 1188 |
+
"loss": 0.1024,
|
| 1189 |
"step": 1550
|
| 1190 |
},
|
| 1191 |
{
|
| 1192 |
"epoch": 26.0,
|
| 1193 |
"learning_rate": 2.4545454545454545e-05,
|
| 1194 |
+
"loss": 0.0723,
|
| 1195 |
"step": 1560
|
| 1196 |
},
|
| 1197 |
{
|
| 1198 |
"epoch": 26.0,
|
| 1199 |
+
"eval_cer": 0.18539071347678368,
|
| 1200 |
+
"eval_loss": 0.008348221890628338,
|
| 1201 |
+
"eval_runtime": 7.0631,
|
| 1202 |
+
"eval_samples_per_second": 67.959,
|
| 1203 |
+
"eval_steps_per_second": 2.124,
|
| 1204 |
+
"eval_wer": 0.20274277242401778,
|
| 1205 |
"step": 1560
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
"epoch": 26.17,
|
| 1209 |
"learning_rate": 2.4090909090909093e-05,
|
| 1210 |
+
"loss": 0.0613,
|
| 1211 |
"step": 1570
|
| 1212 |
},
|
| 1213 |
{
|
| 1214 |
"epoch": 26.33,
|
| 1215 |
"learning_rate": 2.3636363636363637e-05,
|
| 1216 |
+
"loss": 0.0478,
|
| 1217 |
"step": 1580
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 26.5,
|
| 1221 |
"learning_rate": 2.318181818181818e-05,
|
| 1222 |
+
"loss": 0.0807,
|
| 1223 |
"step": 1590
|
| 1224 |
},
|
| 1225 |
{
|
| 1226 |
"epoch": 26.67,
|
| 1227 |
"learning_rate": 2.272727272727273e-05,
|
| 1228 |
+
"loss": 0.0645,
|
| 1229 |
"step": 1600
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
"epoch": 26.83,
|
| 1233 |
"learning_rate": 2.2272727272727274e-05,
|
| 1234 |
+
"loss": 0.0622,
|
| 1235 |
"step": 1610
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
"epoch": 27.0,
|
| 1239 |
"learning_rate": 2.1818181818181818e-05,
|
| 1240 |
+
"loss": 0.061,
|
| 1241 |
"step": 1620
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 27.0,
|
| 1245 |
+
"eval_cer": 0.17791619479048698,
|
| 1246 |
+
"eval_loss": 0.006116455886512995,
|
| 1247 |
+
"eval_runtime": 7.0609,
|
| 1248 |
+
"eval_samples_per_second": 67.98,
|
| 1249 |
+
"eval_steps_per_second": 2.124,
|
| 1250 |
+
"eval_wer": 0.20200148257968867,
|
| 1251 |
"step": 1620
|
| 1252 |
},
|
| 1253 |
{
|
| 1254 |
"epoch": 27.17,
|
| 1255 |
"learning_rate": 2.1363636363636362e-05,
|
| 1256 |
+
"loss": 0.0518,
|
| 1257 |
"step": 1630
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
"epoch": 27.33,
|
| 1261 |
"learning_rate": 2.090909090909091e-05,
|
| 1262 |
+
"loss": 0.0557,
|
| 1263 |
"step": 1640
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
"epoch": 27.5,
|
| 1267 |
"learning_rate": 2.0454545454545457e-05,
|
| 1268 |
+
"loss": 0.0494,
|
| 1269 |
"step": 1650
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 27.67,
|
| 1273 |
"learning_rate": 2e-05,
|
| 1274 |
+
"loss": 0.0396,
|
| 1275 |
"step": 1660
|
| 1276 |
},
|
| 1277 |
{
|
| 1278 |
"epoch": 27.83,
|
| 1279 |
"learning_rate": 1.9545454545454546e-05,
|
| 1280 |
+
"loss": 0.0419,
|
| 1281 |
"step": 1670
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
"epoch": 28.0,
|
| 1285 |
"learning_rate": 1.9090909090909094e-05,
|
| 1286 |
+
"loss": 0.0635,
|
| 1287 |
"step": 1680
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
"epoch": 28.0,
|
| 1291 |
+
"eval_cer": 0.18176670441676104,
|
| 1292 |
+
"eval_loss": 0.005886468570679426,
|
| 1293 |
+
"eval_runtime": 7.1012,
|
| 1294 |
+
"eval_samples_per_second": 67.594,
|
| 1295 |
+
"eval_steps_per_second": 2.112,
|
| 1296 |
+
"eval_wer": 0.19644180874722017,
|
| 1297 |
"step": 1680
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
"epoch": 28.17,
|
| 1301 |
"learning_rate": 1.8636363636363638e-05,
|
| 1302 |
+
"loss": 0.0981,
|
| 1303 |
"step": 1690
|
| 1304 |
},
|
| 1305 |
{
|
| 1306 |
"epoch": 28.33,
|
| 1307 |
"learning_rate": 1.8181818181818182e-05,
|
| 1308 |
+
"loss": 0.0408,
|
| 1309 |
"step": 1700
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
"epoch": 28.5,
|
| 1313 |
"learning_rate": 1.772727272727273e-05,
|
| 1314 |
+
"loss": 0.0735,
|
| 1315 |
"step": 1710
|
| 1316 |
},
|
| 1317 |
{
|
| 1318 |
"epoch": 28.67,
|
| 1319 |
"learning_rate": 1.7272727272727274e-05,
|
| 1320 |
+
"loss": 0.0649,
|
| 1321 |
"step": 1720
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
"epoch": 28.83,
|
| 1325 |
"learning_rate": 1.6818181818181818e-05,
|
| 1326 |
+
"loss": 0.0522,
|
| 1327 |
"step": 1730
|
| 1328 |
},
|
| 1329 |
{
|
| 1330 |
"epoch": 29.0,
|
| 1331 |
"learning_rate": 1.6363636363636366e-05,
|
| 1332 |
+
"loss": 0.0336,
|
| 1333 |
"step": 1740
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
"epoch": 29.0,
|
| 1337 |
+
"eval_cer": 0.15741789354473387,
|
| 1338 |
+
"eval_loss": 0.004836267791688442,
|
| 1339 |
+
"eval_runtime": 7.1076,
|
| 1340 |
+
"eval_samples_per_second": 67.533,
|
| 1341 |
+
"eval_steps_per_second": 2.11,
|
| 1342 |
+
"eval_wer": 0.18865826538176428,
|
| 1343 |
"step": 1740
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
"epoch": 29.17,
|
| 1347 |
"learning_rate": 1.590909090909091e-05,
|
| 1348 |
+
"loss": 0.0592,
|
| 1349 |
"step": 1750
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
"epoch": 29.33,
|
| 1353 |
"learning_rate": 1.5454545454545454e-05,
|
| 1354 |
+
"loss": 0.0499,
|
| 1355 |
"step": 1760
|
| 1356 |
},
|
| 1357 |
{
|
| 1358 |
"epoch": 29.5,
|
| 1359 |
"learning_rate": 1.5e-05,
|
| 1360 |
+
"loss": 0.0483,
|
| 1361 |
"step": 1770
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
"epoch": 29.67,
|
| 1365 |
"learning_rate": 1.4545454545454545e-05,
|
| 1366 |
+
"loss": 0.0572,
|
| 1367 |
"step": 1780
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
"epoch": 29.83,
|
| 1371 |
"learning_rate": 1.409090909090909e-05,
|
| 1372 |
+
"loss": 0.0371,
|
| 1373 |
"step": 1790
|
| 1374 |
},
|
| 1375 |
{
|
| 1376 |
"epoch": 30.0,
|
| 1377 |
"learning_rate": 1.3636363636363637e-05,
|
| 1378 |
+
"loss": 0.0455,
|
| 1379 |
"step": 1800
|
| 1380 |
},
|
| 1381 |
{
|
| 1382 |
"epoch": 30.0,
|
| 1383 |
+
"eval_cer": 0.16942242355605888,
|
| 1384 |
+
"eval_loss": 0.0036398672964423895,
|
| 1385 |
+
"eval_runtime": 7.0873,
|
| 1386 |
+
"eval_samples_per_second": 67.727,
|
| 1387 |
+
"eval_steps_per_second": 2.116,
|
| 1388 |
+
"eval_wer": 0.18421052631578946,
|
| 1389 |
"step": 1800
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 30.17,
|
| 1393 |
"learning_rate": 1.318181818181818e-05,
|
| 1394 |
+
"loss": 0.0679,
|
| 1395 |
"step": 1810
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
"epoch": 30.33,
|
| 1399 |
"learning_rate": 1.2727272727272727e-05,
|
| 1400 |
+
"loss": 0.0556,
|
| 1401 |
"step": 1820
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
"epoch": 30.5,
|
| 1405 |
"learning_rate": 1.2272727272727273e-05,
|
| 1406 |
+
"loss": 0.0555,
|
| 1407 |
"step": 1830
|
| 1408 |
},
|
| 1409 |
{
|
| 1410 |
"epoch": 30.67,
|
| 1411 |
"learning_rate": 1.1818181818181819e-05,
|
| 1412 |
+
"loss": 0.0486,
|
| 1413 |
"step": 1840
|
| 1414 |
},
|
| 1415 |
{
|
| 1416 |
"epoch": 30.83,
|
| 1417 |
"learning_rate": 1.1363636363636365e-05,
|
| 1418 |
+
"loss": 0.06,
|
| 1419 |
"step": 1850
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
"epoch": 31.0,
|
| 1423 |
"learning_rate": 1.0909090909090909e-05,
|
| 1424 |
+
"loss": 0.0672,
|
| 1425 |
"step": 1860
|
| 1426 |
},
|
| 1427 |
{
|
| 1428 |
"epoch": 31.0,
|
| 1429 |
+
"eval_cer": 0.1507361268403171,
|
| 1430 |
+
"eval_loss": 0.00381605327129364,
|
| 1431 |
+
"eval_runtime": 7.0984,
|
| 1432 |
+
"eval_samples_per_second": 67.621,
|
| 1433 |
+
"eval_steps_per_second": 2.113,
|
| 1434 |
+
"eval_wer": 0.18383988139362492,
|
| 1435 |
"step": 1860
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
"epoch": 31.17,
|
| 1439 |
"learning_rate": 1.0454545454545455e-05,
|
| 1440 |
+
"loss": 0.0414,
|
| 1441 |
"step": 1870
|
| 1442 |
},
|
| 1443 |
{
|
| 1444 |
"epoch": 31.33,
|
| 1445 |
"learning_rate": 1e-05,
|
| 1446 |
+
"loss": 0.041,
|
| 1447 |
"step": 1880
|
| 1448 |
},
|
| 1449 |
{
|
| 1450 |
"epoch": 31.5,
|
| 1451 |
"learning_rate": 9.545454545454547e-06,
|
| 1452 |
+
"loss": 0.0529,
|
| 1453 |
"step": 1890
|
| 1454 |
},
|
| 1455 |
{
|
| 1456 |
"epoch": 31.67,
|
| 1457 |
"learning_rate": 9.090909090909091e-06,
|
| 1458 |
+
"loss": 0.0744,
|
| 1459 |
"step": 1900
|
| 1460 |
},
|
| 1461 |
{
|
| 1462 |
"epoch": 31.83,
|
| 1463 |
"learning_rate": 8.636363636363637e-06,
|
| 1464 |
+
"loss": 0.0495,
|
| 1465 |
"step": 1910
|
| 1466 |
},
|
| 1467 |
{
|
| 1468 |
"epoch": 32.0,
|
| 1469 |
"learning_rate": 8.181818181818183e-06,
|
| 1470 |
+
"loss": 0.0315,
|
| 1471 |
"step": 1920
|
| 1472 |
},
|
| 1473 |
{
|
| 1474 |
"epoch": 32.0,
|
| 1475 |
+
"eval_cer": 0.15549263873159683,
|
| 1476 |
+
"eval_loss": 0.003280089935287833,
|
| 1477 |
+
"eval_runtime": 7.1329,
|
| 1478 |
+
"eval_samples_per_second": 67.294,
|
| 1479 |
+
"eval_steps_per_second": 2.103,
|
| 1480 |
+
"eval_wer": 0.18532246108228317,
|
| 1481 |
"step": 1920
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
"epoch": 32.17,
|
| 1485 |
"learning_rate": 7.727272727272727e-06,
|
| 1486 |
+
"loss": 0.0525,
|
| 1487 |
"step": 1930
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
"epoch": 32.33,
|
| 1491 |
"learning_rate": 7.272727272727272e-06,
|
| 1492 |
+
"loss": 0.0946,
|
| 1493 |
"step": 1940
|
| 1494 |
},
|
| 1495 |
{
|
| 1496 |
"epoch": 32.5,
|
| 1497 |
"learning_rate": 6.818181818181818e-06,
|
| 1498 |
+
"loss": 0.0302,
|
| 1499 |
"step": 1950
|
| 1500 |
},
|
| 1501 |
{
|
| 1502 |
"epoch": 32.67,
|
| 1503 |
"learning_rate": 6.363636363636363e-06,
|
| 1504 |
+
"loss": 0.0365,
|
| 1505 |
"step": 1960
|
| 1506 |
},
|
| 1507 |
{
|
| 1508 |
"epoch": 32.83,
|
| 1509 |
"learning_rate": 5.909090909090909e-06,
|
| 1510 |
+
"loss": 0.0357,
|
| 1511 |
"step": 1970
|
| 1512 |
},
|
| 1513 |
{
|
| 1514 |
"epoch": 33.0,
|
| 1515 |
"learning_rate": 5.4545454545454545e-06,
|
| 1516 |
+
"loss": 0.0466,
|
| 1517 |
"step": 1980
|
| 1518 |
},
|
| 1519 |
{
|
| 1520 |
"epoch": 33.0,
|
| 1521 |
+
"eval_cer": 0.15685164212910532,
|
| 1522 |
+
"eval_loss": 0.0033369685988873243,
|
| 1523 |
+
"eval_runtime": 7.0736,
|
| 1524 |
+
"eval_samples_per_second": 67.858,
|
| 1525 |
+
"eval_steps_per_second": 2.121,
|
| 1526 |
+
"eval_wer": 0.1827279466271312,
|
| 1527 |
"step": 1980
|
| 1528 |
},
|
| 1529 |
{
|
| 1530 |
"epoch": 33.17,
|
| 1531 |
"learning_rate": 5e-06,
|
| 1532 |
+
"loss": 0.0872,
|
| 1533 |
"step": 1990
|
| 1534 |
},
|
| 1535 |
{
|
| 1536 |
"epoch": 33.33,
|
| 1537 |
"learning_rate": 4.5454545454545455e-06,
|
| 1538 |
+
"loss": 0.0405,
|
| 1539 |
"step": 2000
|
| 1540 |
},
|
| 1541 |
{
|
| 1542 |
"epoch": 33.5,
|
| 1543 |
"learning_rate": 4.0909090909090915e-06,
|
| 1544 |
+
"loss": 0.0443,
|
| 1545 |
"step": 2010
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
"epoch": 33.67,
|
| 1549 |
"learning_rate": 3.636363636363636e-06,
|
| 1550 |
+
"loss": 0.0367,
|
| 1551 |
"step": 2020
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
"epoch": 33.83,
|
| 1555 |
"learning_rate": 3.1818181818181817e-06,
|
| 1556 |
+
"loss": 0.0358,
|
| 1557 |
"step": 2030
|
| 1558 |
},
|
| 1559 |
{
|
| 1560 |
"epoch": 34.0,
|
| 1561 |
"learning_rate": 2.7272727272727272e-06,
|
| 1562 |
+
"loss": 0.0491,
|
| 1563 |
"step": 2040
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
"epoch": 34.0,
|
| 1567 |
+
"eval_cer": 0.15560588901472253,
|
| 1568 |
+
"eval_loss": 0.0035262603778392076,
|
| 1569 |
+
"eval_runtime": 7.1013,
|
| 1570 |
+
"eval_samples_per_second": 67.593,
|
| 1571 |
+
"eval_steps_per_second": 2.112,
|
| 1572 |
+
"eval_wer": 0.18346923647146035,
|
| 1573 |
"step": 2040
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
"epoch": 34.17,
|
| 1577 |
"learning_rate": 2.2727272727272728e-06,
|
| 1578 |
+
"loss": 0.0409,
|
| 1579 |
"step": 2050
|
| 1580 |
},
|
| 1581 |
{
|
| 1582 |
"epoch": 34.33,
|
| 1583 |
"learning_rate": 1.818181818181818e-06,
|
| 1584 |
+
"loss": 0.0755,
|
| 1585 |
"step": 2060
|
| 1586 |
},
|
| 1587 |
{
|
| 1588 |
"epoch": 34.5,
|
| 1589 |
"learning_rate": 1.3636363636363636e-06,
|
| 1590 |
+
"loss": 0.0617,
|
| 1591 |
"step": 2070
|
| 1592 |
},
|
| 1593 |
{
|
| 1594 |
"epoch": 34.67,
|
| 1595 |
"learning_rate": 9.09090909090909e-07,
|
| 1596 |
+
"loss": 0.0672,
|
| 1597 |
"step": 2080
|
| 1598 |
},
|
| 1599 |
{
|
| 1600 |
"epoch": 34.83,
|
| 1601 |
"learning_rate": 4.545454545454545e-07,
|
| 1602 |
+
"loss": 0.0605,
|
| 1603 |
"step": 2090
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
"epoch": 35.0,
|
| 1607 |
"learning_rate": 0.0,
|
| 1608 |
+
"loss": 0.0315,
|
| 1609 |
"step": 2100
|
| 1610 |
},
|
| 1611 |
{
|
| 1612 |
"epoch": 35.0,
|
| 1613 |
+
"eval_cer": 0.15571913929784825,
|
| 1614 |
+
"eval_loss": 0.003189487848430872,
|
| 1615 |
+
"eval_runtime": 7.1308,
|
| 1616 |
+
"eval_samples_per_second": 67.314,
|
| 1617 |
+
"eval_steps_per_second": 2.104,
|
| 1618 |
+
"eval_wer": 0.18346923647146035,
|
| 1619 |
"step": 2100
|
| 1620 |
},
|
| 1621 |
{
|
| 1622 |
"epoch": 35.0,
|
| 1623 |
"step": 2100,
|
| 1624 |
+
"total_flos": 5.305190022011399e+18,
|
| 1625 |
+
"train_loss": 1.105300986397834,
|
| 1626 |
+
"train_runtime": 4221.7996,
|
| 1627 |
+
"train_samples_per_second": 15.917,
|
| 1628 |
"train_steps_per_second": 0.497
|
| 1629 |
}
|
| 1630 |
],
|
|
|
|
| 1632 |
"max_steps": 2100,
|
| 1633 |
"num_train_epochs": 35,
|
| 1634 |
"save_steps": 500,
|
| 1635 |
+
"total_flos": 5.305190022011399e+18,
|
| 1636 |
"trial_name": null,
|
| 1637 |
"trial_params": null
|
| 1638 |
}
|