Commit
•
82f4531
1
Parent(s):
e81dccb
huggingartists
Browse files- README.md +3 -3
- evaluation.txt +1 -1
- flax_model.msgpack +1 -1
- optimizer.pt +1 -1
- pytorch_model.bin +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +123 -7
- training_args.bin +1 -1
README.md
CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
|
|
45 |
dataset = load_dataset("huggingartists/platina")
|
46 |
```
|
47 |
|
48 |
-
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Платина (Platina)'s lyrics.
|
53 |
|
54 |
-
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/
|
55 |
|
56 |
-
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/
|
57 |
|
58 |
## How to use
|
59 |
|
|
|
45 |
dataset = load_dataset("huggingartists/platina")
|
46 |
```
|
47 |
|
48 |
+
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2ih365j7/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Платина (Platina)'s lyrics.
|
53 |
|
54 |
+
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1quasiz0) for full transparency and reproducibility.
|
55 |
|
56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1quasiz0/artifacts) is logged and versioned.
|
57 |
|
58 |
## How to use
|
59 |
|
evaluation.txt
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"eval_loss": 1.
|
|
|
1 |
+
{"eval_loss": 1.3523887395858765, "eval_runtime": 6.2976, "eval_samples_per_second": 21.119, "eval_steps_per_second": 2.699, "epoch": 5.0}
|
flax_model.msgpack
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 497764120
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1a6eabf9f9233b440eedfafd6710ddc3daeea0fffdfe8c2a281b4758902fa78c
|
3 |
size 497764120
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 995604017
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:554836147560545c9373bbde2a72b0db4282ab87a5b7a113a7d10b992febc8fa
|
3 |
size 995604017
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 510403817
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:058b060e53ad2583dcda7b9ab91cc8ba684ee7591ea5fab893e07e5d17e956f3
|
3 |
size 510403817
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14567
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3ebfb9f7c5ef30fc6c5169e74f8cece8ea746e80139907978a5ace953a7c1b9a
|
3 |
size 14567
|
scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:56a03ee9c4f7d46c39bb609d5652834cf98daba2d29c096120b392bb7cba62b1
|
3 |
size 623
|
trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"best_metric": 1.
|
3 |
-
"best_model_checkpoint": "output/platina/checkpoint-
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -452,11 +452,127 @@
|
|
452 |
"eval_samples_per_second": 22.648,
|
453 |
"eval_steps_per_second": 2.939,
|
454 |
"step": 348
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
455 |
}
|
456 |
],
|
457 |
-
"max_steps":
|
458 |
-
"num_train_epochs":
|
459 |
-
"total_flos":
|
460 |
"trial_name": null,
|
461 |
"trial_params": null
|
462 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 1.3523887395858765,
|
3 |
+
"best_model_checkpoint": "output/platina/checkpoint-435",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"global_step": 435,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
452 |
"eval_samples_per_second": 22.648,
|
453 |
"eval_steps_per_second": 2.939,
|
454 |
"step": 348
|
455 |
+
},
|
456 |
+
{
|
457 |
+
"epoch": 4.02,
|
458 |
+
"learning_rate": 1.788243178686371e-07,
|
459 |
+
"loss": 1.5412,
|
460 |
+
"step": 350
|
461 |
+
},
|
462 |
+
{
|
463 |
+
"epoch": 4.08,
|
464 |
+
"learning_rate": 2.1799062746641418e-06,
|
465 |
+
"loss": 1.4467,
|
466 |
+
"step": 355
|
467 |
+
},
|
468 |
+
{
|
469 |
+
"epoch": 4.14,
|
470 |
+
"learning_rate": 6.340326210572388e-06,
|
471 |
+
"loss": 1.5015,
|
472 |
+
"step": 360
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 4.2,
|
476 |
+
"learning_rate": 1.2524827574860016e-05,
|
477 |
+
"loss": 1.4381,
|
478 |
+
"step": 365
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 4.25,
|
482 |
+
"learning_rate": 2.053235029776879e-05,
|
483 |
+
"loss": 1.4247,
|
484 |
+
"step": 370
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 4.31,
|
488 |
+
"learning_rate": 3.0102567316140582e-05,
|
489 |
+
"loss": 1.414,
|
490 |
+
"step": 375
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 4.37,
|
494 |
+
"learning_rate": 4.092434788749928e-05,
|
495 |
+
"loss": 1.3577,
|
496 |
+
"step": 380
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 4.43,
|
500 |
+
"learning_rate": 5.264587254760211e-05,
|
501 |
+
"loss": 1.4768,
|
502 |
+
"step": 385
|
503 |
+
},
|
504 |
+
{
|
505 |
+
"epoch": 4.48,
|
506 |
+
"learning_rate": 6.48860708710403e-05,
|
507 |
+
"loss": 1.4088,
|
508 |
+
"step": 390
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 4.54,
|
512 |
+
"learning_rate": 7.724701018971169e-05,
|
513 |
+
"loss": 1.4472,
|
514 |
+
"step": 395
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 4.6,
|
518 |
+
"learning_rate": 8.932683250807266e-05,
|
519 |
+
"loss": 1.415,
|
520 |
+
"step": 400
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 4.66,
|
524 |
+
"learning_rate": 0.00010073281903200549,
|
525 |
+
"loss": 1.4059,
|
526 |
+
"step": 405
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 4.71,
|
530 |
+
"learning_rate": 0.00011109415757939859,
|
531 |
+
"loss": 1.4512,
|
532 |
+
"step": 410
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 4.77,
|
536 |
+
"learning_rate": 0.00012007399779992913,
|
537 |
+
"loss": 1.5409,
|
538 |
+
"step": 415
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 4.83,
|
542 |
+
"learning_rate": 0.00012738040228509664,
|
543 |
+
"loss": 1.3528,
|
544 |
+
"step": 420
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 4.89,
|
548 |
+
"learning_rate": 0.0001327758375444962,
|
549 |
+
"loss": 1.5104,
|
550 |
+
"step": 425
|
551 |
+
},
|
552 |
+
{
|
553 |
+
"epoch": 4.94,
|
554 |
+
"learning_rate": 0.0001360848962937354,
|
555 |
+
"loss": 1.4315,
|
556 |
+
"step": 430
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 5.0,
|
560 |
+
"learning_rate": 0.0001372,
|
561 |
+
"loss": 1.4081,
|
562 |
+
"step": 435
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 5.0,
|
566 |
+
"eval_loss": 1.3523887395858765,
|
567 |
+
"eval_runtime": 6.0439,
|
568 |
+
"eval_samples_per_second": 22.006,
|
569 |
+
"eval_steps_per_second": 2.813,
|
570 |
+
"step": 435
|
571 |
}
|
572 |
],
|
573 |
+
"max_steps": 435,
|
574 |
+
"num_train_epochs": 5,
|
575 |
+
"total_flos": 452296507392000.0,
|
576 |
"trial_name": null,
|
577 |
"trial_params": null
|
578 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2863
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2cf0d38b97ef8d76fefde84a017dca88b69ddcb5c6903588227fab13bcb9da45
|
3 |
size 2863
|