Commit
•
37b1cf0
1
Parent(s):
bc93e6c
Training in progress, step 1900
Browse files- {checkpoint-700 → checkpoint-1800}/config.json +0 -0
- {checkpoint-700 → checkpoint-1800}/optimizer.pt +1 -1
- {checkpoint-700 → checkpoint-1800}/preprocessor_config.json +0 -0
- {checkpoint-800 → checkpoint-1800}/pytorch_model.bin +1 -1
- {checkpoint-700 → checkpoint-1800}/rng_state.pth +1 -1
- {checkpoint-800 → checkpoint-1800}/scaler.pt +1 -1
- {checkpoint-700 → checkpoint-1800}/scheduler.pt +1 -1
- {checkpoint-800 → checkpoint-1800}/trainer_state.json +213 -3
- {checkpoint-700 → checkpoint-1800}/training_args.bin +0 -0
- {checkpoint-800 → checkpoint-1900}/config.json +0 -0
- {checkpoint-800 → checkpoint-1900}/optimizer.pt +1 -1
- {checkpoint-800 → checkpoint-1900}/preprocessor_config.json +0 -0
- {checkpoint-700 → checkpoint-1900}/pytorch_model.bin +1 -1
- {checkpoint-800 → checkpoint-1900}/rng_state.pth +1 -1
- {checkpoint-700 → checkpoint-1900}/scaler.pt +1 -1
- {checkpoint-800 → checkpoint-1900}/scheduler.pt +1 -1
- {checkpoint-700 → checkpoint-1900}/trainer_state.json +255 -3
- {checkpoint-800 → checkpoint-1900}/training_args.bin +0 -0
- pytorch_model.bin +1 -1
{checkpoint-700 → checkpoint-1800}/config.json
RENAMED
|
File without changes
|
{checkpoint-700 → checkpoint-1800}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2490946501
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42f99a886bbf378c0dcaebf101e4c5328ee0274d00a62c2e1be3614a61c5e312
|
| 3 |
size 2490946501
|
{checkpoint-700 → checkpoint-1800}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-800 → checkpoint-1800}/pytorch_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262344621
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52a571823a5349c817772817b3ca92d24c7154a82c1ebe1934fc9b5c45efd5ae
|
| 3 |
size 1262344621
|
{checkpoint-700 → checkpoint-1800}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47d39bea761335d6d65b0b6d7eff2cfb4efe3fe039ea66b0c5cffcd47767de2b
|
| 3 |
size 14575
|
{checkpoint-800 → checkpoint-1800}/scaler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 557
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e961ca75ff911c1f4b25d656bc8175958d7392f62f4f35cdbbe7bb3902c5e7a6
|
| 3 |
size 557
|
{checkpoint-700 → checkpoint-1800}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae735c83da45a65907ba8c02193907649b6ed04e0ca77c0400a50dbb61401aaa
|
| 3 |
size 627
|
{checkpoint-800 → checkpoint-1800}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -174,11 +174,221 @@
|
|
| 174 |
"eval_samples_per_second": 23.985,
|
| 175 |
"eval_steps_per_second": 3.019,
|
| 176 |
"step": 800
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 177 |
}
|
| 178 |
],
|
| 179 |
"max_steps": 3675,
|
| 180 |
"num_train_epochs": 25,
|
| 181 |
-
"total_flos":
|
| 182 |
"trial_name": null,
|
| 183 |
"trial_params": null
|
| 184 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 12.244897959183673,
|
| 5 |
+
"global_step": 1800,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 174 |
"eval_samples_per_second": 23.985,
|
| 175 |
"eval_steps_per_second": 3.019,
|
| 176 |
"step": 800
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 5.78,
|
| 180 |
+
"learning_rate": 0.00026721259842519684,
|
| 181 |
+
"loss": 0.8337,
|
| 182 |
+
"step": 850
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.12,
|
| 186 |
+
"learning_rate": 0.00026248818897637796,
|
| 187 |
+
"loss": 0.7717,
|
| 188 |
+
"step": 900
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 6.12,
|
| 192 |
+
"eval_cer": 0.20955486711234306,
|
| 193 |
+
"eval_loss": 0.5637524127960205,
|
| 194 |
+
"eval_runtime": 24.5794,
|
| 195 |
+
"eval_samples_per_second": 23.922,
|
| 196 |
+
"eval_steps_per_second": 3.011,
|
| 197 |
+
"step": 900
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 6.46,
|
| 201 |
+
"learning_rate": 0.00025776377952755903,
|
| 202 |
+
"loss": 0.6711,
|
| 203 |
+
"step": 950
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 6.8,
|
| 207 |
+
"learning_rate": 0.00025303937007874016,
|
| 208 |
+
"loss": 0.7003,
|
| 209 |
+
"step": 1000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 6.8,
|
| 213 |
+
"eval_cer": 0.19279308658079244,
|
| 214 |
+
"eval_loss": 0.5100580453872681,
|
| 215 |
+
"eval_runtime": 24.6049,
|
| 216 |
+
"eval_samples_per_second": 23.898,
|
| 217 |
+
"eval_steps_per_second": 3.008,
|
| 218 |
+
"step": 1000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 7.14,
|
| 222 |
+
"learning_rate": 0.0002483149606299212,
|
| 223 |
+
"loss": 0.6349,
|
| 224 |
+
"step": 1050
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 7.48,
|
| 228 |
+
"learning_rate": 0.00024359055118110235,
|
| 229 |
+
"loss": 0.5935,
|
| 230 |
+
"step": 1100
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 7.48,
|
| 234 |
+
"eval_cer": 0.17746616663949127,
|
| 235 |
+
"eval_loss": 0.46862325072288513,
|
| 236 |
+
"eval_runtime": 24.2024,
|
| 237 |
+
"eval_samples_per_second": 24.295,
|
| 238 |
+
"eval_steps_per_second": 3.058,
|
| 239 |
+
"step": 1100
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 7.82,
|
| 243 |
+
"learning_rate": 0.00023886614173228342,
|
| 244 |
+
"loss": 0.576,
|
| 245 |
+
"step": 1150
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 8.16,
|
| 249 |
+
"learning_rate": 0.00023414173228346455,
|
| 250 |
+
"loss": 0.5239,
|
| 251 |
+
"step": 1200
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 8.16,
|
| 255 |
+
"eval_cer": 0.1918799934779064,
|
| 256 |
+
"eval_loss": 0.4271094799041748,
|
| 257 |
+
"eval_runtime": 24.6968,
|
| 258 |
+
"eval_samples_per_second": 23.809,
|
| 259 |
+
"eval_steps_per_second": 2.996,
|
| 260 |
+
"step": 1200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 8.5,
|
| 264 |
+
"learning_rate": 0.00022941732283464564,
|
| 265 |
+
"loss": 0.5207,
|
| 266 |
+
"step": 1250
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 8.84,
|
| 270 |
+
"learning_rate": 0.00022469291338582677,
|
| 271 |
+
"loss": 0.4855,
|
| 272 |
+
"step": 1300
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 8.84,
|
| 276 |
+
"eval_cer": 0.17550953856187837,
|
| 277 |
+
"eval_loss": 0.40729257464408875,
|
| 278 |
+
"eval_runtime": 24.885,
|
| 279 |
+
"eval_samples_per_second": 23.629,
|
| 280 |
+
"eval_steps_per_second": 2.974,
|
| 281 |
+
"step": 1300
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 9.18,
|
| 285 |
+
"learning_rate": 0.00021996850393700784,
|
| 286 |
+
"loss": 0.4621,
|
| 287 |
+
"step": 1350
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 9.52,
|
| 291 |
+
"learning_rate": 0.00021524409448818896,
|
| 292 |
+
"loss": 0.4583,
|
| 293 |
+
"step": 1400
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 9.52,
|
| 297 |
+
"eval_cer": 0.1760313060492418,
|
| 298 |
+
"eval_loss": 0.41417357325553894,
|
| 299 |
+
"eval_runtime": 24.8425,
|
| 300 |
+
"eval_samples_per_second": 23.669,
|
| 301 |
+
"eval_steps_per_second": 2.979,
|
| 302 |
+
"step": 1400
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 9.86,
|
| 306 |
+
"learning_rate": 0.00021051968503937006,
|
| 307 |
+
"loss": 0.4415,
|
| 308 |
+
"step": 1450
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 10.2,
|
| 312 |
+
"learning_rate": 0.00020579527559055118,
|
| 313 |
+
"loss": 0.417,
|
| 314 |
+
"step": 1500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 10.2,
|
| 318 |
+
"eval_cer": 0.17156367193869232,
|
| 319 |
+
"eval_loss": 0.3833578824996948,
|
| 320 |
+
"eval_runtime": 24.455,
|
| 321 |
+
"eval_samples_per_second": 24.044,
|
| 322 |
+
"eval_steps_per_second": 3.026,
|
| 323 |
+
"step": 1500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 10.54,
|
| 327 |
+
"learning_rate": 0.00020107086614173225,
|
| 328 |
+
"loss": 0.3959,
|
| 329 |
+
"step": 1550
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 10.88,
|
| 333 |
+
"learning_rate": 0.00019634645669291338,
|
| 334 |
+
"loss": 0.4074,
|
| 335 |
+
"step": 1600
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 10.88,
|
| 339 |
+
"eval_cer": 0.16301972933311593,
|
| 340 |
+
"eval_loss": 0.3626195192337036,
|
| 341 |
+
"eval_runtime": 24.5184,
|
| 342 |
+
"eval_samples_per_second": 23.982,
|
| 343 |
+
"eval_steps_per_second": 3.018,
|
| 344 |
+
"step": 1600
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 11.22,
|
| 348 |
+
"learning_rate": 0.00019162204724409448,
|
| 349 |
+
"loss": 0.3584,
|
| 350 |
+
"step": 1650
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 11.56,
|
| 354 |
+
"learning_rate": 0.0001868976377952756,
|
| 355 |
+
"loss": 0.3682,
|
| 356 |
+
"step": 1700
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 11.56,
|
| 360 |
+
"eval_cer": 0.15685635088863525,
|
| 361 |
+
"eval_loss": 0.36927542090415955,
|
| 362 |
+
"eval_runtime": 24.3751,
|
| 363 |
+
"eval_samples_per_second": 24.123,
|
| 364 |
+
"eval_steps_per_second": 3.036,
|
| 365 |
+
"step": 1700
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 11.9,
|
| 369 |
+
"learning_rate": 0.00018217322834645667,
|
| 370 |
+
"loss": 0.3455,
|
| 371 |
+
"step": 1750
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 12.24,
|
| 375 |
+
"learning_rate": 0.00017744881889763777,
|
| 376 |
+
"loss": 0.3245,
|
| 377 |
+
"step": 1800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 12.24,
|
| 381 |
+
"eval_cer": 0.15816076960704387,
|
| 382 |
+
"eval_loss": 0.3740461468696594,
|
| 383 |
+
"eval_runtime": 24.7162,
|
| 384 |
+
"eval_samples_per_second": 23.79,
|
| 385 |
+
"eval_steps_per_second": 2.994,
|
| 386 |
+
"step": 1800
|
| 387 |
}
|
| 388 |
],
|
| 389 |
"max_steps": 3675,
|
| 390 |
"num_train_epochs": 25,
|
| 391 |
+
"total_flos": 6.982329131312831e+18,
|
| 392 |
"trial_name": null,
|
| 393 |
"trial_params": null
|
| 394 |
}
|
{checkpoint-700 → checkpoint-1800}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-800 → checkpoint-1900}/config.json
RENAMED
|
File without changes
|
{checkpoint-800 → checkpoint-1900}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2490946501
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:edfc1df136634c5b701601b6ff6af9b1881f7cd09119136764e47cb758be8c01
|
| 3 |
size 2490946501
|
{checkpoint-800 → checkpoint-1900}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-700 → checkpoint-1900}/pytorch_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262344621
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:944bbfe53ed8e3678f0df597947df90c1f1fd1a592c5e2f1c1893cd644db8b25
|
| 3 |
size 1262344621
|
{checkpoint-800 → checkpoint-1900}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14639
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:19558c49bacfb403056aca7f61bd635fce6a3b4d6a2d60ba71c39efd76e04aa8
|
| 3 |
size 14639
|
{checkpoint-700 → checkpoint-1900}/scaler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 557
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c12154f7b08b94398fffa27beb8a5986462659e12ab1223bd629316877afb771
|
| 3 |
size 557
|
{checkpoint-800 → checkpoint-1900}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0ea1eff427e80a898e7d7eedbf89dfef56c38dc29788f8cb29b845574ba7ec4
|
| 3 |
size 627
|
{checkpoint-700 → checkpoint-1900}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -153,11 +153,263 @@
|
|
| 153 |
"eval_samples_per_second": 24.068,
|
| 154 |
"eval_steps_per_second": 3.029,
|
| 155 |
"step": 700
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 156 |
}
|
| 157 |
],
|
| 158 |
"max_steps": 3675,
|
| 159 |
"num_train_epochs": 25,
|
| 160 |
-
"total_flos":
|
| 161 |
"trial_name": null,
|
| 162 |
"trial_params": null
|
| 163 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 12.92517006802721,
|
| 5 |
+
"global_step": 1900,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 153 |
"eval_samples_per_second": 24.068,
|
| 154 |
"eval_steps_per_second": 3.029,
|
| 155 |
"step": 700
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 5.1,
|
| 159 |
+
"learning_rate": 0.0002766614173228346,
|
| 160 |
+
"loss": 1.0153,
|
| 161 |
+
"step": 750
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.44,
|
| 165 |
+
"learning_rate": 0.0002719370078740157,
|
| 166 |
+
"loss": 0.8879,
|
| 167 |
+
"step": 800
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 5.44,
|
| 171 |
+
"eval_cer": 0.23668677645524214,
|
| 172 |
+
"eval_loss": 0.6693652868270874,
|
| 173 |
+
"eval_runtime": 24.5154,
|
| 174 |
+
"eval_samples_per_second": 23.985,
|
| 175 |
+
"eval_steps_per_second": 3.019,
|
| 176 |
+
"step": 800
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 5.78,
|
| 180 |
+
"learning_rate": 0.00026721259842519684,
|
| 181 |
+
"loss": 0.8337,
|
| 182 |
+
"step": 850
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.12,
|
| 186 |
+
"learning_rate": 0.00026248818897637796,
|
| 187 |
+
"loss": 0.7717,
|
| 188 |
+
"step": 900
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 6.12,
|
| 192 |
+
"eval_cer": 0.20955486711234306,
|
| 193 |
+
"eval_loss": 0.5637524127960205,
|
| 194 |
+
"eval_runtime": 24.5794,
|
| 195 |
+
"eval_samples_per_second": 23.922,
|
| 196 |
+
"eval_steps_per_second": 3.011,
|
| 197 |
+
"step": 900
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 6.46,
|
| 201 |
+
"learning_rate": 0.00025776377952755903,
|
| 202 |
+
"loss": 0.6711,
|
| 203 |
+
"step": 950
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 6.8,
|
| 207 |
+
"learning_rate": 0.00025303937007874016,
|
| 208 |
+
"loss": 0.7003,
|
| 209 |
+
"step": 1000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 6.8,
|
| 213 |
+
"eval_cer": 0.19279308658079244,
|
| 214 |
+
"eval_loss": 0.5100580453872681,
|
| 215 |
+
"eval_runtime": 24.6049,
|
| 216 |
+
"eval_samples_per_second": 23.898,
|
| 217 |
+
"eval_steps_per_second": 3.008,
|
| 218 |
+
"step": 1000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 7.14,
|
| 222 |
+
"learning_rate": 0.0002483149606299212,
|
| 223 |
+
"loss": 0.6349,
|
| 224 |
+
"step": 1050
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 7.48,
|
| 228 |
+
"learning_rate": 0.00024359055118110235,
|
| 229 |
+
"loss": 0.5935,
|
| 230 |
+
"step": 1100
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 7.48,
|
| 234 |
+
"eval_cer": 0.17746616663949127,
|
| 235 |
+
"eval_loss": 0.46862325072288513,
|
| 236 |
+
"eval_runtime": 24.2024,
|
| 237 |
+
"eval_samples_per_second": 24.295,
|
| 238 |
+
"eval_steps_per_second": 3.058,
|
| 239 |
+
"step": 1100
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 7.82,
|
| 243 |
+
"learning_rate": 0.00023886614173228342,
|
| 244 |
+
"loss": 0.576,
|
| 245 |
+
"step": 1150
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 8.16,
|
| 249 |
+
"learning_rate": 0.00023414173228346455,
|
| 250 |
+
"loss": 0.5239,
|
| 251 |
+
"step": 1200
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 8.16,
|
| 255 |
+
"eval_cer": 0.1918799934779064,
|
| 256 |
+
"eval_loss": 0.4271094799041748,
|
| 257 |
+
"eval_runtime": 24.6968,
|
| 258 |
+
"eval_samples_per_second": 23.809,
|
| 259 |
+
"eval_steps_per_second": 2.996,
|
| 260 |
+
"step": 1200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 8.5,
|
| 264 |
+
"learning_rate": 0.00022941732283464564,
|
| 265 |
+
"loss": 0.5207,
|
| 266 |
+
"step": 1250
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 8.84,
|
| 270 |
+
"learning_rate": 0.00022469291338582677,
|
| 271 |
+
"loss": 0.4855,
|
| 272 |
+
"step": 1300
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 8.84,
|
| 276 |
+
"eval_cer": 0.17550953856187837,
|
| 277 |
+
"eval_loss": 0.40729257464408875,
|
| 278 |
+
"eval_runtime": 24.885,
|
| 279 |
+
"eval_samples_per_second": 23.629,
|
| 280 |
+
"eval_steps_per_second": 2.974,
|
| 281 |
+
"step": 1300
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 9.18,
|
| 285 |
+
"learning_rate": 0.00021996850393700784,
|
| 286 |
+
"loss": 0.4621,
|
| 287 |
+
"step": 1350
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 9.52,
|
| 291 |
+
"learning_rate": 0.00021524409448818896,
|
| 292 |
+
"loss": 0.4583,
|
| 293 |
+
"step": 1400
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 9.52,
|
| 297 |
+
"eval_cer": 0.1760313060492418,
|
| 298 |
+
"eval_loss": 0.41417357325553894,
|
| 299 |
+
"eval_runtime": 24.8425,
|
| 300 |
+
"eval_samples_per_second": 23.669,
|
| 301 |
+
"eval_steps_per_second": 2.979,
|
| 302 |
+
"step": 1400
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 9.86,
|
| 306 |
+
"learning_rate": 0.00021051968503937006,
|
| 307 |
+
"loss": 0.4415,
|
| 308 |
+
"step": 1450
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 10.2,
|
| 312 |
+
"learning_rate": 0.00020579527559055118,
|
| 313 |
+
"loss": 0.417,
|
| 314 |
+
"step": 1500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 10.2,
|
| 318 |
+
"eval_cer": 0.17156367193869232,
|
| 319 |
+
"eval_loss": 0.3833578824996948,
|
| 320 |
+
"eval_runtime": 24.455,
|
| 321 |
+
"eval_samples_per_second": 24.044,
|
| 322 |
+
"eval_steps_per_second": 3.026,
|
| 323 |
+
"step": 1500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 10.54,
|
| 327 |
+
"learning_rate": 0.00020107086614173225,
|
| 328 |
+
"loss": 0.3959,
|
| 329 |
+
"step": 1550
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 10.88,
|
| 333 |
+
"learning_rate": 0.00019634645669291338,
|
| 334 |
+
"loss": 0.4074,
|
| 335 |
+
"step": 1600
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 10.88,
|
| 339 |
+
"eval_cer": 0.16301972933311593,
|
| 340 |
+
"eval_loss": 0.3626195192337036,
|
| 341 |
+
"eval_runtime": 24.5184,
|
| 342 |
+
"eval_samples_per_second": 23.982,
|
| 343 |
+
"eval_steps_per_second": 3.018,
|
| 344 |
+
"step": 1600
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 11.22,
|
| 348 |
+
"learning_rate": 0.00019162204724409448,
|
| 349 |
+
"loss": 0.3584,
|
| 350 |
+
"step": 1650
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 11.56,
|
| 354 |
+
"learning_rate": 0.0001868976377952756,
|
| 355 |
+
"loss": 0.3682,
|
| 356 |
+
"step": 1700
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 11.56,
|
| 360 |
+
"eval_cer": 0.15685635088863525,
|
| 361 |
+
"eval_loss": 0.36927542090415955,
|
| 362 |
+
"eval_runtime": 24.3751,
|
| 363 |
+
"eval_samples_per_second": 24.123,
|
| 364 |
+
"eval_steps_per_second": 3.036,
|
| 365 |
+
"step": 1700
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 11.9,
|
| 369 |
+
"learning_rate": 0.00018217322834645667,
|
| 370 |
+
"loss": 0.3455,
|
| 371 |
+
"step": 1750
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 12.24,
|
| 375 |
+
"learning_rate": 0.00017744881889763777,
|
| 376 |
+
"loss": 0.3245,
|
| 377 |
+
"step": 1800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 12.24,
|
| 381 |
+
"eval_cer": 0.15816076960704387,
|
| 382 |
+
"eval_loss": 0.3740461468696594,
|
| 383 |
+
"eval_runtime": 24.7162,
|
| 384 |
+
"eval_samples_per_second": 23.79,
|
| 385 |
+
"eval_steps_per_second": 2.994,
|
| 386 |
+
"step": 1800
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 12.59,
|
| 390 |
+
"learning_rate": 0.0001727244094488189,
|
| 391 |
+
"loss": 0.3208,
|
| 392 |
+
"step": 1850
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 12.93,
|
| 396 |
+
"learning_rate": 0.000168,
|
| 397 |
+
"loss": 0.3063,
|
| 398 |
+
"step": 1900
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 12.93,
|
| 402 |
+
"eval_cer": 0.15904125224196966,
|
| 403 |
+
"eval_loss": 0.3622555434703827,
|
| 404 |
+
"eval_runtime": 24.4729,
|
| 405 |
+
"eval_samples_per_second": 24.027,
|
| 406 |
+
"eval_steps_per_second": 3.024,
|
| 407 |
+
"step": 1900
|
| 408 |
}
|
| 409 |
],
|
| 410 |
"max_steps": 3675,
|
| 411 |
"num_train_epochs": 25,
|
| 412 |
+
"total_flos": 7.38349957947191e+18,
|
| 413 |
"trial_name": null,
|
| 414 |
"trial_params": null
|
| 415 |
}
|
{checkpoint-800 → checkpoint-1900}/training_args.bin
RENAMED
|
File without changes
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262344621
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:944bbfe53ed8e3678f0df597947df90c1f1fd1a592c5e2f1c1893cd644db8b25
|
| 3 |
size 1262344621
|