Commit
•
0d45503
1
Parent(s):
e68c6d8
Training in progress, step 2500
Browse files- {checkpoint-1300 → checkpoint-2400}/config.json +0 -0
- {checkpoint-1300 → checkpoint-2400}/optimizer.pt +1 -1
- {checkpoint-1300 → checkpoint-2400}/preprocessor_config.json +0 -0
- {checkpoint-1400 → checkpoint-2400}/pytorch_model.bin +1 -1
- {checkpoint-1400 → checkpoint-2400}/rng_state.pth +1 -1
- {checkpoint-1400 → checkpoint-2400}/scaler.pt +1 -1
- {checkpoint-1300 → checkpoint-2400}/scheduler.pt +1 -1
- {checkpoint-1400 → checkpoint-2400}/trainer_state.json +213 -3
- {checkpoint-1300 → checkpoint-2400}/training_args.bin +0 -0
- {checkpoint-1400 → checkpoint-2500}/config.json +0 -0
- {checkpoint-1400 → checkpoint-2500}/optimizer.pt +1 -1
- {checkpoint-1400 → checkpoint-2500}/preprocessor_config.json +0 -0
- {checkpoint-1300 → checkpoint-2500}/pytorch_model.bin +1 -1
- {checkpoint-1300 → checkpoint-2500}/rng_state.pth +2 -2
- {checkpoint-1300 → checkpoint-2500}/scaler.pt +1 -1
- {checkpoint-1400 → checkpoint-2500}/scheduler.pt +1 -1
- {checkpoint-1300 → checkpoint-2500}/trainer_state.json +255 -3
- {checkpoint-1400 → checkpoint-2500}/training_args.bin +0 -0
- pytorch_model.bin +1 -1
{checkpoint-1300 → checkpoint-2400}/config.json
RENAMED
|
File without changes
|
{checkpoint-1300 → checkpoint-2400}/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 sha256:21ec413dbf6dce3f66f0847998faf0727145082a010278c772b5854d5056f118
|
| 3 |
size 2490946501
|
{checkpoint-1300 → checkpoint-2400}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-1400 → checkpoint-2400}/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 sha256:66e4f13af4571dc4fc776d37a670bd005d274231a78ff352023ba5c4fadd8614
|
| 3 |
size 1262344621
|
{checkpoint-1400 → checkpoint-2400}/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 sha256:89ee801de1abf33cd6ebc4f0b38240fe2dfecaa66947acc7503d6a21ded7f76f
|
| 3 |
size 14639
|
{checkpoint-1400 → checkpoint-2400}/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 sha256:2dcc907e7c7cdb7f74446d1478d9321af62972312f69212a9da63fbf093fd591
|
| 3 |
size 557
|
{checkpoint-1300 → checkpoint-2400}/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 sha256:3212d7bddbe681e165c9fd72ab4f4e8de8c824c01389aa4a534800797bb190ca
|
| 3 |
size 627
|
{checkpoint-1400 → checkpoint-2400}/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,
|
|
@@ -300,11 +300,221 @@
|
|
| 300 |
"eval_samples_per_second": 23.669,
|
| 301 |
"eval_steps_per_second": 2.979,
|
| 302 |
"step": 1400
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 303 |
}
|
| 304 |
],
|
| 305 |
"max_steps": 3675,
|
| 306 |
"num_train_epochs": 25,
|
| 307 |
-
"total_flos":
|
| 308 |
"trial_name": null,
|
| 309 |
"trial_params": null
|
| 310 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 16.3265306122449,
|
| 5 |
+
"global_step": 2400,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 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 |
+
"epoch": 13.27,
|
| 411 |
+
"learning_rate": 0.0001632755905511811,
|
| 412 |
+
"loss": 0.3019,
|
| 413 |
+
"step": 1950
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 13.61,
|
| 417 |
+
"learning_rate": 0.00015855118110236219,
|
| 418 |
+
"loss": 0.2945,
|
| 419 |
+
"step": 2000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 13.61,
|
| 423 |
+
"eval_cer": 0.16634599706505787,
|
| 424 |
+
"eval_loss": 0.3725011348724365,
|
| 425 |
+
"eval_runtime": 25.0023,
|
| 426 |
+
"eval_samples_per_second": 23.518,
|
| 427 |
+
"eval_steps_per_second": 2.96,
|
| 428 |
+
"step": 2000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 13.95,
|
| 432 |
+
"learning_rate": 0.0001538267716535433,
|
| 433 |
+
"loss": 0.279,
|
| 434 |
+
"step": 2050
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 14.29,
|
| 438 |
+
"learning_rate": 0.0001491023622047244,
|
| 439 |
+
"loss": 0.2674,
|
| 440 |
+
"step": 2100
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 14.29,
|
| 444 |
+
"eval_cer": 0.15731289744007826,
|
| 445 |
+
"eval_loss": 0.3531067371368408,
|
| 446 |
+
"eval_runtime": 24.8381,
|
| 447 |
+
"eval_samples_per_second": 23.673,
|
| 448 |
+
"eval_steps_per_second": 2.979,
|
| 449 |
+
"step": 2100
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 14.63,
|
| 453 |
+
"learning_rate": 0.0001443779527559055,
|
| 454 |
+
"loss": 0.2584,
|
| 455 |
+
"step": 2150
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 14.97,
|
| 459 |
+
"learning_rate": 0.0001396535433070866,
|
| 460 |
+
"loss": 0.2796,
|
| 461 |
+
"step": 2200
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 14.97,
|
| 465 |
+
"eval_cer": 0.14808413500733736,
|
| 466 |
+
"eval_loss": 0.3606802523136139,
|
| 467 |
+
"eval_runtime": 24.8151,
|
| 468 |
+
"eval_samples_per_second": 23.695,
|
| 469 |
+
"eval_steps_per_second": 2.982,
|
| 470 |
+
"step": 2200
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 15.31,
|
| 474 |
+
"learning_rate": 0.0001349291338582677,
|
| 475 |
+
"loss": 0.2462,
|
| 476 |
+
"step": 2250
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 15.65,
|
| 480 |
+
"learning_rate": 0.0001302047244094488,
|
| 481 |
+
"loss": 0.256,
|
| 482 |
+
"step": 2300
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 15.65,
|
| 486 |
+
"eval_cer": 0.15819338007500408,
|
| 487 |
+
"eval_loss": 0.3580550253391266,
|
| 488 |
+
"eval_runtime": 24.5695,
|
| 489 |
+
"eval_samples_per_second": 23.932,
|
| 490 |
+
"eval_steps_per_second": 3.012,
|
| 491 |
+
"step": 2300
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 15.99,
|
| 495 |
+
"learning_rate": 0.00012548031496062992,
|
| 496 |
+
"loss": 0.2524,
|
| 497 |
+
"step": 2350
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 16.33,
|
| 501 |
+
"learning_rate": 0.00012075590551181102,
|
| 502 |
+
"loss": 0.2219,
|
| 503 |
+
"step": 2400
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 16.33,
|
| 507 |
+
"eval_cer": 0.14801891407141693,
|
| 508 |
+
"eval_loss": 0.35925593972206116,
|
| 509 |
+
"eval_runtime": 24.982,
|
| 510 |
+
"eval_samples_per_second": 23.537,
|
| 511 |
+
"eval_steps_per_second": 2.962,
|
| 512 |
+
"step": 2400
|
| 513 |
}
|
| 514 |
],
|
| 515 |
"max_steps": 3675,
|
| 516 |
"num_train_epochs": 25,
|
| 517 |
+
"total_flos": 9.330801216434254e+18,
|
| 518 |
"trial_name": null,
|
| 519 |
"trial_params": null
|
| 520 |
}
|
{checkpoint-1300 → checkpoint-2400}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-1400 → checkpoint-2500}/config.json
RENAMED
|
File without changes
|
{checkpoint-1400 → checkpoint-2500}/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:c295d37c841b2d9274e1c2ff590e71f0a545a4dfe10d0f547dca1cf116d93838
|
| 3 |
size 2490946501
|
{checkpoint-1400 → checkpoint-2500}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-1300 → checkpoint-2500}/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:a368a48ff43ab4bc8e301dc5a41abb75b9de3b052aeffd6edcf4765b37523a8d
|
| 3 |
size 1262344621
|
{checkpoint-1300 → checkpoint-2500}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8575e4b2ab78e0afbc81dc9eacc30543d16798a08a6fac89cde25f7c7f9d9a06
|
| 3 |
+
size 14703
|
{checkpoint-1300 → checkpoint-2500}/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 sha256:0c0deb49ff9261d5b5d7609a0755c25d7a947f998b4734a1525c3a096a987646
|
| 3 |
size 557
|
{checkpoint-1400 → checkpoint-2500}/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:d1dc16f2d67a6830e5fca5f7b55d822dad401a11ddc497a9f782324d540f77d8
|
| 3 |
size 627
|
{checkpoint-1300 → checkpoint-2500}/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,
|
|
@@ -279,11 +279,263 @@
|
|
| 279 |
"eval_samples_per_second": 23.629,
|
| 280 |
"eval_steps_per_second": 2.974,
|
| 281 |
"step": 1300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 282 |
}
|
| 283 |
],
|
| 284 |
"max_steps": 3675,
|
| 285 |
"num_train_epochs": 25,
|
| 286 |
-
"total_flos":
|
| 287 |
"trial_name": null,
|
| 288 |
"trial_params": null
|
| 289 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 17.006802721088434,
|
| 5 |
+
"global_step": 2500,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 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 |
+
"epoch": 13.27,
|
| 411 |
+
"learning_rate": 0.0001632755905511811,
|
| 412 |
+
"loss": 0.3019,
|
| 413 |
+
"step": 1950
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 13.61,
|
| 417 |
+
"learning_rate": 0.00015855118110236219,
|
| 418 |
+
"loss": 0.2945,
|
| 419 |
+
"step": 2000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 13.61,
|
| 423 |
+
"eval_cer": 0.16634599706505787,
|
| 424 |
+
"eval_loss": 0.3725011348724365,
|
| 425 |
+
"eval_runtime": 25.0023,
|
| 426 |
+
"eval_samples_per_second": 23.518,
|
| 427 |
+
"eval_steps_per_second": 2.96,
|
| 428 |
+
"step": 2000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 13.95,
|
| 432 |
+
"learning_rate": 0.0001538267716535433,
|
| 433 |
+
"loss": 0.279,
|
| 434 |
+
"step": 2050
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 14.29,
|
| 438 |
+
"learning_rate": 0.0001491023622047244,
|
| 439 |
+
"loss": 0.2674,
|
| 440 |
+
"step": 2100
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 14.29,
|
| 444 |
+
"eval_cer": 0.15731289744007826,
|
| 445 |
+
"eval_loss": 0.3531067371368408,
|
| 446 |
+
"eval_runtime": 24.8381,
|
| 447 |
+
"eval_samples_per_second": 23.673,
|
| 448 |
+
"eval_steps_per_second": 2.979,
|
| 449 |
+
"step": 2100
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 14.63,
|
| 453 |
+
"learning_rate": 0.0001443779527559055,
|
| 454 |
+
"loss": 0.2584,
|
| 455 |
+
"step": 2150
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 14.97,
|
| 459 |
+
"learning_rate": 0.0001396535433070866,
|
| 460 |
+
"loss": 0.2796,
|
| 461 |
+
"step": 2200
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 14.97,
|
| 465 |
+
"eval_cer": 0.14808413500733736,
|
| 466 |
+
"eval_loss": 0.3606802523136139,
|
| 467 |
+
"eval_runtime": 24.8151,
|
| 468 |
+
"eval_samples_per_second": 23.695,
|
| 469 |
+
"eval_steps_per_second": 2.982,
|
| 470 |
+
"step": 2200
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 15.31,
|
| 474 |
+
"learning_rate": 0.0001349291338582677,
|
| 475 |
+
"loss": 0.2462,
|
| 476 |
+
"step": 2250
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 15.65,
|
| 480 |
+
"learning_rate": 0.0001302047244094488,
|
| 481 |
+
"loss": 0.256,
|
| 482 |
+
"step": 2300
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 15.65,
|
| 486 |
+
"eval_cer": 0.15819338007500408,
|
| 487 |
+
"eval_loss": 0.3580550253391266,
|
| 488 |
+
"eval_runtime": 24.5695,
|
| 489 |
+
"eval_samples_per_second": 23.932,
|
| 490 |
+
"eval_steps_per_second": 3.012,
|
| 491 |
+
"step": 2300
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 15.99,
|
| 495 |
+
"learning_rate": 0.00012548031496062992,
|
| 496 |
+
"loss": 0.2524,
|
| 497 |
+
"step": 2350
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 16.33,
|
| 501 |
+
"learning_rate": 0.00012075590551181102,
|
| 502 |
+
"loss": 0.2219,
|
| 503 |
+
"step": 2400
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 16.33,
|
| 507 |
+
"eval_cer": 0.14801891407141693,
|
| 508 |
+
"eval_loss": 0.35925593972206116,
|
| 509 |
+
"eval_runtime": 24.982,
|
| 510 |
+
"eval_samples_per_second": 23.537,
|
| 511 |
+
"eval_steps_per_second": 2.962,
|
| 512 |
+
"step": 2400
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 16.67,
|
| 516 |
+
"learning_rate": 0.0001160314960629921,
|
| 517 |
+
"loss": 0.2364,
|
| 518 |
+
"step": 2450
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 17.01,
|
| 522 |
+
"learning_rate": 0.00011130708661417321,
|
| 523 |
+
"loss": 0.2291,
|
| 524 |
+
"step": 2500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 17.01,
|
| 528 |
+
"eval_cer": 0.1471058209685309,
|
| 529 |
+
"eval_loss": 0.35567909479141235,
|
| 530 |
+
"eval_runtime": 24.4749,
|
| 531 |
+
"eval_samples_per_second": 24.025,
|
| 532 |
+
"eval_steps_per_second": 3.024,
|
| 533 |
+
"step": 2500
|
| 534 |
}
|
| 535 |
],
|
| 536 |
"max_steps": 3675,
|
| 537 |
"num_train_epochs": 25,
|
| 538 |
+
"total_flos": 9.701648869870967e+18,
|
| 539 |
"trial_name": null,
|
| 540 |
"trial_params": null
|
| 541 |
}
|
{checkpoint-1400 → checkpoint-2500}/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:a368a48ff43ab4bc8e301dc5a41abb75b9de3b052aeffd6edcf4765b37523a8d
|
| 3 |
size 1262344621
|