42
This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the cifar100 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4122
- Accuracy: 0.8983
- Dt Accuracy: 0.8983
- Df Accuracy: 0.1827
- Unlearn Overall Accuracy: 0.8922
- Unlearn Time: None
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: 0.0002
- train_batch_size: 128
- eval_batch_size: 256
- 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 | Overall Accuracy | Unlearn Overall Accuracy | Time |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 391 | 0.4467 | 0.754 | 0.4033 | 0.4033 | None |
| 1.2106 | 2.0 | 782 | 0.4069 | 0.771 | 0.3831 | 0.3831 | None |
| 1.0503 | 3.0 | 1173 | 0.4355 | 0.7777 | 0.3738 | 0.3738 | None |
| 0.9645 | 4.0 | 1564 | 0.4543 | 0.762 | 0.3937 | 0.3937 | None |
| 0.9645 | 5.0 | 1955 | 0.4076 | 0.7477 | 0.4129 | 0.4129 | None |
| 0.8771 | 6.0 | 2346 | 0.4350 | 0.7047 | 0.4631 | 0.4631 | None |
| 0.7866 | 7.0 | 2737 | 0.4448 | 0.6643 | 0.5082 | 0.5082 | None |
| 0.7204 | 8.0 | 3128 | 0.4489 | 0.6137 | 0.5600 | 0.5600 | None |
| 0.6629 | 9.0 | 3519 | 0.4399 | 0.5727 | 0.6009 | 0.6009 | None |
| 0.6629 | 10.0 | 3910 | 0.4259 | 0.497 | 0.6689 | 0.6689 | None |
| 0.6048 | 11.0 | 4301 | 0.4372 | 0.439 | 0.7162 | 0.7162 | None |
| 0.5634 | 12.0 | 4692 | 0.4355 | 0.4067 | 0.7409 | 0.7409 | None |
| 0.5194 | 13.0 | 5083 | 0.4413 | 0.3437 | 0.7854 | 0.7854 | None |
| 0.5194 | 14.0 | 5474 | 0.4299 | 0.3327 | 0.7963 | 0.7963 | None |
| 0.4786 | 15.0 | 5865 | 0.4409 | 0.2567 | 0.8449 | 0.8449 | None |
| 0.4425 | 16.0 | 6256 | 0.4293 | 0.2327 | 0.8609 | 0.8609 | None |
| 0.4103 | 17.0 | 6647 | 0.4281 | 0.2177 | 0.8700 | 0.8700 | None |
| 0.3901 | 18.0 | 7038 | 0.4197 | 0.1933 | 0.8854 | 0.8854 | None |
| 0.3901 | 19.0 | 7429 | 0.4177 | 0.1897 | 0.8883 | 0.8883 | None |
| 0.3635 | 20.0 | 7820 | 0.4122 | 0.1827 | 0.8922 | 0.8922 | None |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2+cu118
- Datasets 2.18.0
- Tokenizers 0.15.2