<!-- 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-100-no-pretrain-20-epoch
This model is a fine-tuned version of on the image_folder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1354
- Accuracy: 0.9595
- F1: 0.9600
- Precision: 0.9610
- Recall: 0.9595
- Balanced Acc: 0.9472
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 |
|---|---|---|---|---|---|---|---|---|
| 1.3674 | 1.0 | 66 | 1.2838 | 0.3958 | 0.4072 | 0.4693 | 0.3958 | 0.4695 |
| 0.976 | 2.0 | 132 | 1.0647 | 0.5885 | 0.5904 | 0.6686 | 0.5885 | 0.5967 |
| 0.7904 | 3.0 | 198 | 0.7471 | 0.6777 | 0.7007 | 0.7614 | 0.6777 | 0.7059 |
| 0.616 | 4.0 | 264 | 0.4532 | 0.8420 | 0.8470 | 0.8560 | 0.8420 | 0.7918 |
| 0.489 | 5.0 | 330 | 0.4153 | 0.8385 | 0.8370 | 0.8718 | 0.8385 | 0.7628 |
| 0.3829 | 6.0 | 396 | 0.3386 | 0.8675 | 0.8765 | 0.9063 | 0.8675 | 0.8731 |
| 0.2713 | 7.0 | 462 | 0.3027 | 0.8912 | 0.8965 | 0.9110 | 0.8912 | 0.9081 |
| 0.2308 | 8.0 | 528 | 0.2875 | 0.8837 | 0.8942 | 0.9237 | 0.8837 | 0.9152 |
| 0.1831 | 9.0 | 594 | 0.2027 | 0.9311 | 0.9341 | 0.9421 | 0.9311 | 0.9212 |
| 0.1754 | 10.0 | 660 | 0.1712 | 0.9387 | 0.9411 | 0.9473 | 0.9387 | 0.9428 |
| 0.1515 | 11.0 | 726 | 0.1999 | 0.9259 | 0.9327 | 0.9523 | 0.9259 | 0.9292 |
| 0.1302 | 12.0 | 792 | 0.1740 | 0.9427 | 0.9438 | 0.9459 | 0.9427 | 0.9332 |
| 0.1186 | 13.0 | 858 | 0.1465 | 0.9479 | 0.9493 | 0.9535 | 0.9479 | 0.9425 |
| 0.1004 | 14.0 | 924 | 0.1509 | 0.9473 | 0.9481 | 0.9500 | 0.9473 | 0.9371 |
| 0.0909 | 15.0 | 990 | 0.1638 | 0.9456 | 0.9471 | 0.9506 | 0.9456 | 0.9482 |
| 0.0725 | 16.0 | 1056 | 0.1484 | 0.9583 | 0.9584 | 0.9585 | 0.9583 | 0.9318 |
| 0.0687 | 17.0 | 1122 | 0.1553 | 0.9543 | 0.9550 | 0.9565 | 0.9543 | 0.9437 |
| 0.061 | 18.0 | 1188 | 0.1367 | 0.9549 | 0.9555 | 0.9566 | 0.9549 | 0.9422 |
| 0.0478 | 19.0 | 1254 | 0.1380 | 0.9601 | 0.9603 | 0.9607 | 0.9601 | 0.9423 |
| 0.0406 | 20.0 | 1320 | 0.1354 | 0.9595 | 0.9600 | 0.9610 | 0.9595 | 0.9472 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
