Commit
•
85aa57d
1
Parent(s):
31df82b
End of training
Browse files- README.md +29 -4
- model.safetensors +1 -1
- trainer_state.json +376 -26
README.md
CHANGED
|
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 2.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -40,7 +40,7 @@ The following hyperparameters were used during training:
|
|
| 40 |
- seed: 42
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
-
- num_epochs:
|
| 44 |
- mixed_precision_training: Native AMP
|
| 45 |
|
| 46 |
### Training results
|
|
@@ -70,8 +70,33 @@ The following hyperparameters were used during training:
|
|
| 70 |
| 3.0847 | 21.0 | 2415 | 2.9321 |
|
| 71 |
| 3.0333 | 22.0 | 2530 | 2.9128 |
|
| 72 |
| 3.0126 | 23.0 | 2645 | 2.8627 |
|
| 73 |
-
| 2.
|
| 74 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
|
| 76 |
|
| 77 |
### Framework versions
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 2.8303
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 40 |
- seed: 42
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
+
- num_epochs: 50
|
| 44 |
- mixed_precision_training: Native AMP
|
| 45 |
|
| 46 |
### Training results
|
|
|
|
| 70 |
| 3.0847 | 21.0 | 2415 | 2.9321 |
|
| 71 |
| 3.0333 | 22.0 | 2530 | 2.9128 |
|
| 72 |
| 3.0126 | 23.0 | 2645 | 2.8627 |
|
| 73 |
+
| 2.9968 | 24.0 | 2760 | 3.0186 |
|
| 74 |
+
| 3.0295 | 25.0 | 2875 | 3.0148 |
|
| 75 |
+
| 3.0294 | 26.0 | 2990 | 3.0341 |
|
| 76 |
+
| 3.0395 | 27.0 | 3105 | 2.9997 |
|
| 77 |
+
| 3.0445 | 28.0 | 3220 | 3.0575 |
|
| 78 |
+
| 2.9761 | 29.0 | 3335 | 2.9707 |
|
| 79 |
+
| 3.0075 | 30.0 | 3450 | 2.9392 |
|
| 80 |
+
| 3.0198 | 31.0 | 3565 | 2.9122 |
|
| 81 |
+
| 2.9782 | 32.0 | 3680 | 2.9471 |
|
| 82 |
+
| 2.9773 | 33.0 | 3795 | 3.0306 |
|
| 83 |
+
| 2.9528 | 34.0 | 3910 | 2.8513 |
|
| 84 |
+
| 2.9228 | 35.0 | 4025 | 2.8997 |
|
| 85 |
+
| 2.9221 | 36.0 | 4140 | 2.8646 |
|
| 86 |
+
| 2.8933 | 37.0 | 4255 | 2.8871 |
|
| 87 |
+
| 2.8925 | 38.0 | 4370 | 2.9407 |
|
| 88 |
+
| 2.9069 | 39.0 | 4485 | 2.9625 |
|
| 89 |
+
| 2.9246 | 40.0 | 4600 | 2.9946 |
|
| 90 |
+
| 2.9089 | 41.0 | 4715 | 2.8936 |
|
| 91 |
+
| 2.8573 | 42.0 | 4830 | 2.8272 |
|
| 92 |
+
| 2.8378 | 43.0 | 4945 | 2.8543 |
|
| 93 |
+
| 2.8957 | 44.0 | 5060 | 2.8590 |
|
| 94 |
+
| 2.8454 | 45.0 | 5175 | 2.8525 |
|
| 95 |
+
| 2.7964 | 46.0 | 5290 | 2.8658 |
|
| 96 |
+
| 2.8172 | 47.0 | 5405 | 2.8872 |
|
| 97 |
+
| 2.8472 | 48.0 | 5520 | 2.8654 |
|
| 98 |
+
| 2.811 | 49.0 | 5635 | 2.8572 |
|
| 99 |
+
| 2.801 | 50.0 | 5750 | 2.8303 |
|
| 100 |
|
| 101 |
|
| 102 |
### Framework versions
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 173870884
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79e5d32da4245509b017fc3052add7ff321633c4ad8033144e685fcd78b1f3c8
|
| 3 |
size 173870884
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 2.
|
| 3 |
-
"best_model_checkpoint": "cdetr-mist1-brain-gt-tumors-8ah-6l/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -332,47 +332,397 @@
|
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 24.0,
|
| 335 |
-
"learning_rate":
|
| 336 |
-
"loss": 2.
|
| 337 |
"step": 2760
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 24.0,
|
| 341 |
-
"eval_loss":
|
| 342 |
-
"eval_runtime":
|
| 343 |
-
"eval_samples_per_second": 6.
|
| 344 |
-
"eval_steps_per_second": 0.
|
| 345 |
"step": 2760
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 25.0,
|
| 349 |
-
"learning_rate":
|
| 350 |
-
"loss":
|
| 351 |
"step": 2875
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 25.0,
|
| 355 |
-
"eval_loss":
|
| 356 |
-
"eval_runtime":
|
| 357 |
-
"eval_samples_per_second": 6.
|
| 358 |
-
"eval_steps_per_second": 0.
|
| 359 |
"step": 2875
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch":
|
| 363 |
-
"
|
| 364 |
-
"
|
| 365 |
-
"
|
| 366 |
-
|
| 367 |
-
|
| 368 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 369 |
}
|
| 370 |
],
|
| 371 |
"logging_steps": 500,
|
| 372 |
-
"max_steps":
|
| 373 |
-
"num_train_epochs":
|
| 374 |
"save_steps": 500,
|
| 375 |
-
"total_flos":
|
| 376 |
"trial_name": null,
|
| 377 |
"trial_params": null
|
| 378 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 2.827199935913086,
|
| 3 |
+
"best_model_checkpoint": "cdetr-mist1-brain-gt-tumors-8ah-6l/checkpoint-4830",
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 5750,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 24.0,
|
| 335 |
+
"learning_rate": 5.234782608695652e-06,
|
| 336 |
+
"loss": 2.9968,
|
| 337 |
"step": 2760
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 24.0,
|
| 341 |
+
"eval_loss": 3.0186123847961426,
|
| 342 |
+
"eval_runtime": 5.7749,
|
| 343 |
+
"eval_samples_per_second": 6.926,
|
| 344 |
+
"eval_steps_per_second": 0.866,
|
| 345 |
"step": 2760
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 25.0,
|
| 349 |
+
"learning_rate": 5.036521739130436e-06,
|
| 350 |
+
"loss": 3.0295,
|
| 351 |
"step": 2875
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 25.0,
|
| 355 |
+
"eval_loss": 3.0147616863250732,
|
| 356 |
+
"eval_runtime": 5.799,
|
| 357 |
+
"eval_samples_per_second": 6.898,
|
| 358 |
+
"eval_steps_per_second": 0.862,
|
| 359 |
"step": 2875
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 26.0,
|
| 363 |
+
"learning_rate": 4.836521739130435e-06,
|
| 364 |
+
"loss": 3.0294,
|
| 365 |
+
"step": 2990
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 26.0,
|
| 369 |
+
"eval_loss": 3.034066915512085,
|
| 370 |
+
"eval_runtime": 5.8474,
|
| 371 |
+
"eval_samples_per_second": 6.841,
|
| 372 |
+
"eval_steps_per_second": 0.855,
|
| 373 |
+
"step": 2990
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 27.0,
|
| 377 |
+
"learning_rate": 4.636521739130435e-06,
|
| 378 |
+
"loss": 3.0395,
|
| 379 |
+
"step": 3105
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 27.0,
|
| 383 |
+
"eval_loss": 2.99969220161438,
|
| 384 |
+
"eval_runtime": 5.7579,
|
| 385 |
+
"eval_samples_per_second": 6.947,
|
| 386 |
+
"eval_steps_per_second": 0.868,
|
| 387 |
+
"step": 3105
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.0,
|
| 391 |
+
"learning_rate": 4.436521739130435e-06,
|
| 392 |
+
"loss": 3.0445,
|
| 393 |
+
"step": 3220
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 28.0,
|
| 397 |
+
"eval_loss": 3.0574753284454346,
|
| 398 |
+
"eval_runtime": 5.6888,
|
| 399 |
+
"eval_samples_per_second": 7.031,
|
| 400 |
+
"eval_steps_per_second": 0.879,
|
| 401 |
+
"step": 3220
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 29.0,
|
| 405 |
+
"learning_rate": 4.236521739130436e-06,
|
| 406 |
+
"loss": 2.9761,
|
| 407 |
+
"step": 3335
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 29.0,
|
| 411 |
+
"eval_loss": 2.97066068649292,
|
| 412 |
+
"eval_runtime": 5.7458,
|
| 413 |
+
"eval_samples_per_second": 6.962,
|
| 414 |
+
"eval_steps_per_second": 0.87,
|
| 415 |
+
"step": 3335
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 30.0,
|
| 419 |
+
"learning_rate": 4.038260869565218e-06,
|
| 420 |
+
"loss": 3.0075,
|
| 421 |
+
"step": 3450
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 30.0,
|
| 425 |
+
"eval_loss": 2.939246416091919,
|
| 426 |
+
"eval_runtime": 5.722,
|
| 427 |
+
"eval_samples_per_second": 6.991,
|
| 428 |
+
"eval_steps_per_second": 0.874,
|
| 429 |
+
"step": 3450
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 31.0,
|
| 433 |
+
"learning_rate": 3.838260869565218e-06,
|
| 434 |
+
"loss": 3.0198,
|
| 435 |
+
"step": 3565
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 31.0,
|
| 439 |
+
"eval_loss": 2.9121549129486084,
|
| 440 |
+
"eval_runtime": 5.8674,
|
| 441 |
+
"eval_samples_per_second": 6.817,
|
| 442 |
+
"eval_steps_per_second": 0.852,
|
| 443 |
+
"step": 3565
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 32.0,
|
| 447 |
+
"learning_rate": 3.6382608695652177e-06,
|
| 448 |
+
"loss": 2.9782,
|
| 449 |
+
"step": 3680
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 32.0,
|
| 453 |
+
"eval_loss": 2.947079658508301,
|
| 454 |
+
"eval_runtime": 5.7346,
|
| 455 |
+
"eval_samples_per_second": 6.975,
|
| 456 |
+
"eval_steps_per_second": 0.872,
|
| 457 |
+
"step": 3680
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 33.0,
|
| 461 |
+
"learning_rate": 3.4382608695652175e-06,
|
| 462 |
+
"loss": 2.9773,
|
| 463 |
+
"step": 3795
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 33.0,
|
| 467 |
+
"eval_loss": 3.030561923980713,
|
| 468 |
+
"eval_runtime": 5.7197,
|
| 469 |
+
"eval_samples_per_second": 6.993,
|
| 470 |
+
"eval_steps_per_second": 0.874,
|
| 471 |
+
"step": 3795
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 34.0,
|
| 475 |
+
"learning_rate": 3.2382608695652177e-06,
|
| 476 |
+
"loss": 2.9528,
|
| 477 |
+
"step": 3910
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 34.0,
|
| 481 |
+
"eval_loss": 2.8513386249542236,
|
| 482 |
+
"eval_runtime": 5.6599,
|
| 483 |
+
"eval_samples_per_second": 7.067,
|
| 484 |
+
"eval_steps_per_second": 0.883,
|
| 485 |
+
"step": 3910
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 35.0,
|
| 489 |
+
"learning_rate": 3.038260869565218e-06,
|
| 490 |
+
"loss": 2.9228,
|
| 491 |
+
"step": 4025
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 35.0,
|
| 495 |
+
"eval_loss": 2.899709701538086,
|
| 496 |
+
"eval_runtime": 5.6812,
|
| 497 |
+
"eval_samples_per_second": 7.041,
|
| 498 |
+
"eval_steps_per_second": 0.88,
|
| 499 |
+
"step": 4025
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 36.0,
|
| 503 |
+
"learning_rate": 2.8382608695652177e-06,
|
| 504 |
+
"loss": 2.9221,
|
| 505 |
+
"step": 4140
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 36.0,
|
| 509 |
+
"eval_loss": 2.8646340370178223,
|
| 510 |
+
"eval_runtime": 5.6561,
|
| 511 |
+
"eval_samples_per_second": 7.072,
|
| 512 |
+
"eval_steps_per_second": 0.884,
|
| 513 |
+
"step": 4140
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 37.0,
|
| 517 |
+
"learning_rate": 2.638260869565218e-06,
|
| 518 |
+
"loss": 2.8933,
|
| 519 |
+
"step": 4255
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 37.0,
|
| 523 |
+
"eval_loss": 2.8870697021484375,
|
| 524 |
+
"eval_runtime": 5.7007,
|
| 525 |
+
"eval_samples_per_second": 7.017,
|
| 526 |
+
"eval_steps_per_second": 0.877,
|
| 527 |
+
"step": 4255
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 38.0,
|
| 531 |
+
"learning_rate": 2.4382608695652177e-06,
|
| 532 |
+
"loss": 2.8925,
|
| 533 |
+
"step": 4370
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 38.0,
|
| 537 |
+
"eval_loss": 2.94067120552063,
|
| 538 |
+
"eval_runtime": 5.6567,
|
| 539 |
+
"eval_samples_per_second": 7.071,
|
| 540 |
+
"eval_steps_per_second": 0.884,
|
| 541 |
+
"step": 4370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 39.0,
|
| 545 |
+
"learning_rate": 2.2382608695652175e-06,
|
| 546 |
+
"loss": 2.9069,
|
| 547 |
+
"step": 4485
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 39.0,
|
| 551 |
+
"eval_loss": 2.9625325202941895,
|
| 552 |
+
"eval_runtime": 5.6615,
|
| 553 |
+
"eval_samples_per_second": 7.065,
|
| 554 |
+
"eval_steps_per_second": 0.883,
|
| 555 |
+
"step": 4485
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 40.0,
|
| 559 |
+
"learning_rate": 2.0382608695652177e-06,
|
| 560 |
+
"loss": 2.9246,
|
| 561 |
+
"step": 4600
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 40.0,
|
| 565 |
+
"eval_loss": 2.994633197784424,
|
| 566 |
+
"eval_runtime": 5.6894,
|
| 567 |
+
"eval_samples_per_second": 7.031,
|
| 568 |
+
"eval_steps_per_second": 0.879,
|
| 569 |
+
"step": 4600
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 41.0,
|
| 573 |
+
"learning_rate": 1.8382608695652178e-06,
|
| 574 |
+
"loss": 2.9089,
|
| 575 |
+
"step": 4715
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 41.0,
|
| 579 |
+
"eval_loss": 2.8935651779174805,
|
| 580 |
+
"eval_runtime": 5.7226,
|
| 581 |
+
"eval_samples_per_second": 6.99,
|
| 582 |
+
"eval_steps_per_second": 0.874,
|
| 583 |
+
"step": 4715
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 42.0,
|
| 587 |
+
"learning_rate": 1.6382608695652176e-06,
|
| 588 |
+
"loss": 2.8573,
|
| 589 |
+
"step": 4830
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 42.0,
|
| 593 |
+
"eval_loss": 2.827199935913086,
|
| 594 |
+
"eval_runtime": 5.6411,
|
| 595 |
+
"eval_samples_per_second": 7.091,
|
| 596 |
+
"eval_steps_per_second": 0.886,
|
| 597 |
+
"step": 4830
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 43.0,
|
| 601 |
+
"learning_rate": 1.4382608695652176e-06,
|
| 602 |
+
"loss": 2.8378,
|
| 603 |
+
"step": 4945
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 43.0,
|
| 607 |
+
"eval_loss": 2.8543009757995605,
|
| 608 |
+
"eval_runtime": 5.6643,
|
| 609 |
+
"eval_samples_per_second": 7.062,
|
| 610 |
+
"eval_steps_per_second": 0.883,
|
| 611 |
+
"step": 4945
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 44.0,
|
| 615 |
+
"learning_rate": 1.2382608695652176e-06,
|
| 616 |
+
"loss": 2.8957,
|
| 617 |
+
"step": 5060
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 44.0,
|
| 621 |
+
"eval_loss": 2.859046459197998,
|
| 622 |
+
"eval_runtime": 5.687,
|
| 623 |
+
"eval_samples_per_second": 7.034,
|
| 624 |
+
"eval_steps_per_second": 0.879,
|
| 625 |
+
"step": 5060
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 45.0,
|
| 629 |
+
"learning_rate": 1.0382608695652174e-06,
|
| 630 |
+
"loss": 2.8454,
|
| 631 |
+
"step": 5175
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 45.0,
|
| 635 |
+
"eval_loss": 2.8524787425994873,
|
| 636 |
+
"eval_runtime": 5.6623,
|
| 637 |
+
"eval_samples_per_second": 7.064,
|
| 638 |
+
"eval_steps_per_second": 0.883,
|
| 639 |
+
"step": 5175
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 46.0,
|
| 643 |
+
"learning_rate": 8.382608695652175e-07,
|
| 644 |
+
"loss": 2.7964,
|
| 645 |
+
"step": 5290
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 46.0,
|
| 649 |
+
"eval_loss": 2.8657748699188232,
|
| 650 |
+
"eval_runtime": 5.6625,
|
| 651 |
+
"eval_samples_per_second": 7.064,
|
| 652 |
+
"eval_steps_per_second": 0.883,
|
| 653 |
+
"step": 5290
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 47.0,
|
| 657 |
+
"learning_rate": 6.382608695652175e-07,
|
| 658 |
+
"loss": 2.8172,
|
| 659 |
+
"step": 5405
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 47.0,
|
| 663 |
+
"eval_loss": 2.8872039318084717,
|
| 664 |
+
"eval_runtime": 5.6798,
|
| 665 |
+
"eval_samples_per_second": 7.043,
|
| 666 |
+
"eval_steps_per_second": 0.88,
|
| 667 |
+
"step": 5405
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 48.0,
|
| 671 |
+
"learning_rate": 4.382608695652174e-07,
|
| 672 |
+
"loss": 2.8472,
|
| 673 |
+
"step": 5520
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 48.0,
|
| 677 |
+
"eval_loss": 2.8653695583343506,
|
| 678 |
+
"eval_runtime": 5.6595,
|
| 679 |
+
"eval_samples_per_second": 7.068,
|
| 680 |
+
"eval_steps_per_second": 0.883,
|
| 681 |
+
"step": 5520
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 49.0,
|
| 685 |
+
"learning_rate": 2.3826086956521743e-07,
|
| 686 |
+
"loss": 2.811,
|
| 687 |
+
"step": 5635
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 49.0,
|
| 691 |
+
"eval_loss": 2.8571643829345703,
|
| 692 |
+
"eval_runtime": 5.7272,
|
| 693 |
+
"eval_samples_per_second": 6.984,
|
| 694 |
+
"eval_steps_per_second": 0.873,
|
| 695 |
+
"step": 5635
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 50.0,
|
| 699 |
+
"learning_rate": 3.82608695652174e-08,
|
| 700 |
+
"loss": 2.801,
|
| 701 |
+
"step": 5750
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 50.0,
|
| 705 |
+
"eval_loss": 2.830277919769287,
|
| 706 |
+
"eval_runtime": 5.6412,
|
| 707 |
+
"eval_samples_per_second": 7.091,
|
| 708 |
+
"eval_steps_per_second": 0.886,
|
| 709 |
+
"step": 5750
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 50.0,
|
| 713 |
+
"step": 5750,
|
| 714 |
+
"total_flos": 1.147766207616e+19,
|
| 715 |
+
"train_loss": 1.5786318518597147,
|
| 716 |
+
"train_runtime": 2642.836,
|
| 717 |
+
"train_samples_per_second": 8.703,
|
| 718 |
+
"train_steps_per_second": 2.176
|
| 719 |
}
|
| 720 |
],
|
| 721 |
"logging_steps": 500,
|
| 722 |
+
"max_steps": 5750,
|
| 723 |
+
"num_train_epochs": 50,
|
| 724 |
"save_steps": 500,
|
| 725 |
+
"total_flos": 1.147766207616e+19,
|
| 726 |
"trial_name": null,
|
| 727 |
"trial_params": null
|
| 728 |
}
|