hushem_40x_deit_base_sgd_00001_fold3
This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.4462
- Accuracy: 0.2326
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.395 | 1.0 | 217 | 1.4656 | 0.2326 |
1.3668 | 2.0 | 434 | 1.4647 | 0.2326 |
1.3731 | 3.0 | 651 | 1.4638 | 0.2326 |
1.4025 | 4.0 | 868 | 1.4629 | 0.2326 |
1.3639 | 5.0 | 1085 | 1.4620 | 0.2326 |
1.3879 | 6.0 | 1302 | 1.4612 | 0.2326 |
1.3778 | 7.0 | 1519 | 1.4604 | 0.2326 |
1.3612 | 8.0 | 1736 | 1.4596 | 0.2326 |
1.3705 | 9.0 | 1953 | 1.4589 | 0.2326 |
1.3789 | 10.0 | 2170 | 1.4582 | 0.2326 |
1.361 | 11.0 | 2387 | 1.4575 | 0.2326 |
1.3859 | 12.0 | 2604 | 1.4568 | 0.2326 |
1.3896 | 13.0 | 2821 | 1.4562 | 0.2326 |
1.3672 | 14.0 | 3038 | 1.4556 | 0.2326 |
1.3726 | 15.0 | 3255 | 1.4550 | 0.2326 |
1.3714 | 16.0 | 3472 | 1.4544 | 0.2326 |
1.3679 | 17.0 | 3689 | 1.4539 | 0.2326 |
1.3686 | 18.0 | 3906 | 1.4533 | 0.2326 |
1.3759 | 19.0 | 4123 | 1.4528 | 0.2326 |
1.3715 | 20.0 | 4340 | 1.4524 | 0.2326 |
1.3586 | 21.0 | 4557 | 1.4519 | 0.2326 |
1.3602 | 22.0 | 4774 | 1.4515 | 0.2326 |
1.3784 | 23.0 | 4991 | 1.4511 | 0.2326 |
1.3361 | 24.0 | 5208 | 1.4507 | 0.2326 |
1.3639 | 25.0 | 5425 | 1.4503 | 0.2326 |
1.3719 | 26.0 | 5642 | 1.4499 | 0.2326 |
1.37 | 27.0 | 5859 | 1.4496 | 0.2326 |
1.3566 | 28.0 | 6076 | 1.4493 | 0.2326 |
1.3167 | 29.0 | 6293 | 1.4490 | 0.2326 |
1.3522 | 30.0 | 6510 | 1.4487 | 0.2326 |
1.3828 | 31.0 | 6727 | 1.4484 | 0.2326 |
1.3664 | 32.0 | 6944 | 1.4481 | 0.2326 |
1.3407 | 33.0 | 7161 | 1.4479 | 0.2326 |
1.3633 | 34.0 | 7378 | 1.4477 | 0.2326 |
1.3611 | 35.0 | 7595 | 1.4475 | 0.2326 |
1.3425 | 36.0 | 7812 | 1.4473 | 0.2326 |
1.3418 | 37.0 | 8029 | 1.4471 | 0.2326 |
1.3479 | 38.0 | 8246 | 1.4470 | 0.2326 |
1.3576 | 39.0 | 8463 | 1.4468 | 0.2326 |
1.3408 | 40.0 | 8680 | 1.4467 | 0.2326 |
1.3433 | 41.0 | 8897 | 1.4466 | 0.2326 |
1.3828 | 42.0 | 9114 | 1.4465 | 0.2326 |
1.3558 | 43.0 | 9331 | 1.4464 | 0.2326 |
1.3616 | 44.0 | 9548 | 1.4463 | 0.2326 |
1.36 | 45.0 | 9765 | 1.4463 | 0.2326 |
1.3465 | 46.0 | 9982 | 1.4462 | 0.2326 |
1.3484 | 47.0 | 10199 | 1.4462 | 0.2326 |
1.3826 | 48.0 | 10416 | 1.4462 | 0.2326 |
1.3656 | 49.0 | 10633 | 1.4462 | 0.2326 |
1.3363 | 50.0 | 10850 | 1.4462 | 0.2326 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2
- Downloads last month
- 5