<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->
Imene/vit-base-patch16-384-wi4
This model is a fine-tuned version of google/vit-base-patch16-384 on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.1742
- Train Accuracy: 0.9982
- Train Top-3-accuracy: 0.9997
- Validation Loss: 1.5010
- Validation Accuracy: 0.5746
- Validation Top-3-accuracy: 0.8040
- Epoch: 10
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:
- optimizer: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 1800, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
- training_precision: mixed_float16
Training results
| Train Loss | Train Accuracy | Train Top-3-accuracy | Validation Loss | Validation Accuracy | Validation Top-3-accuracy | Epoch |
|---|---|---|---|---|---|---|
| 3.7777 | 0.0845 | 0.1855 | 3.3754 | 0.1543 | 0.3014 | 0 |
| 2.7253 | 0.3277 | 0.5560 | 2.4975 | 0.3452 | 0.5892 | 1 |
| 2.0079 | 0.5236 | 0.7589 | 2.1228 | 0.4234 | 0.6882 | 2 |
| 1.5256 | 0.6663 | 0.8549 | 1.9117 | 0.4734 | 0.7445 | 3 |
| 1.1602 | 0.7712 | 0.9270 | 1.8059 | 0.5162 | 0.7560 | 4 |
| 0.8509 | 0.8659 | 0.9614 | 1.6534 | 0.5516 | 0.7758 | 5 |
| 0.5955 | 0.9353 | 0.9836 | 1.6139 | 0.5610 | 0.7935 | 6 |
| 0.4229 | 0.9687 | 0.9940 | 1.5655 | 0.5631 | 0.7925 | 7 |
| 0.3045 | 0.9859 | 0.9979 | 1.5290 | 0.5714 | 0.7987 | 8 |
| 0.2221 | 0.9958 | 0.9990 | 1.5061 | 0.5954 | 0.8008 | 9 |
| 0.1742 | 0.9982 | 0.9997 | 1.5010 | 0.5746 | 0.8040 | 10 |
Framework versions
- Transformers 4.21.3
- TensorFlow 2.8.2
- Datasets 2.4.0
- Tokenizers 0.12.1
