<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
wbc-10-pretrain-20-epochsss
This model is a fine-tuned version of koobear/masked-10-pretraining-20-epoch on the image_folder dataset. It achieves the following results on the evaluation set:
- Loss: 0.4309
- Accuracy: 0.9688
- F1: 0.9693
- Precision: 0.9708
- Recall: 0.9688
- Balanced Acc: 0.9599
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: 2e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | Balanced Acc |
|---|---|---|---|---|---|---|---|---|
| 1.5411 | 1.0 | 7 | 1.3790 | 0.8895 | 0.8985 | 0.9194 | 0.8895 | 0.8473 |
| 1.3881 | 2.0 | 14 | 1.2476 | 0.9242 | 0.9189 | 0.9241 | 0.9242 | 0.8315 |
| 1.2796 | 3.0 | 21 | 1.1089 | 0.9334 | 0.9301 | 0.9382 | 0.9334 | 0.8724 |
| 1.158 | 4.0 | 28 | 1.0077 | 0.9468 | 0.9476 | 0.9530 | 0.9468 | 0.9180 |
| 1.0356 | 5.0 | 35 | 0.9019 | 0.9485 | 0.9495 | 0.9550 | 0.9485 | 0.9220 |
| 0.9418 | 6.0 | 42 | 0.8120 | 0.9514 | 0.9530 | 0.9585 | 0.9514 | 0.9324 |
| 0.8675 | 7.0 | 49 | 0.7336 | 0.9554 | 0.9563 | 0.9591 | 0.9554 | 0.9366 |
| 0.8061 | 8.0 | 56 | 0.6802 | 0.9566 | 0.9574 | 0.9595 | 0.9566 | 0.9457 |
| 0.7419 | 9.0 | 63 | 0.6255 | 0.9525 | 0.9541 | 0.9582 | 0.9525 | 0.9409 |
| 0.6886 | 10.0 | 70 | 0.5781 | 0.9618 | 0.9618 | 0.9634 | 0.9618 | 0.9391 |
| 0.655 | 11.0 | 77 | 0.5521 | 0.9560 | 0.9573 | 0.9607 | 0.9560 | 0.9487 |
| 0.6299 | 12.0 | 84 | 0.5187 | 0.9624 | 0.9628 | 0.9639 | 0.9624 | 0.9465 |
| 0.5987 | 13.0 | 91 | 0.4983 | 0.9624 | 0.9628 | 0.9639 | 0.9624 | 0.9498 |
| 0.5626 | 14.0 | 98 | 0.4799 | 0.9641 | 0.9647 | 0.9661 | 0.9641 | 0.9573 |
| 0.5531 | 15.0 | 105 | 0.4634 | 0.9676 | 0.9680 | 0.9691 | 0.9676 | 0.9565 |
| 0.5037 | 16.0 | 112 | 0.4493 | 0.9676 | 0.9679 | 0.9691 | 0.9676 | 0.9540 |
| 0.5202 | 17.0 | 119 | 0.4415 | 0.9664 | 0.9670 | 0.9685 | 0.9664 | 0.9548 |
| 0.5053 | 18.0 | 126 | 0.4346 | 0.9682 | 0.9686 | 0.9702 | 0.9682 | 0.9558 |
| 0.5102 | 19.0 | 133 | 0.4317 | 0.9699 | 0.9704 | 0.9718 | 0.9699 | 0.9606 |
| 0.4942 | 20.0 | 140 | 0.4309 | 0.9688 | 0.9693 | 0.9708 | 0.9688 | 0.9599 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
