beit-base-patch16-224-hasta-65-fold2
This model is a fine-tuned version of microsoft/beit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.6728
- Accuracy: 0.75
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 100
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.5714 | 1 | 1.3013 | 0.3056 |
| No log | 1.7143 | 3 | 1.2799 | 0.2778 |
| No log | 2.8571 | 5 | 1.2588 | 0.3333 |
| No log | 4.0 | 7 | 1.1296 | 0.3889 |
| No log | 4.5714 | 8 | 1.1295 | 0.3611 |
| 1.1611 | 5.7143 | 10 | 1.2689 | 0.25 |
| 1.1611 | 6.8571 | 12 | 1.0895 | 0.3889 |
| 1.1611 | 8.0 | 14 | 1.0978 | 0.5 |
| 1.1611 | 8.5714 | 15 | 1.1168 | 0.5278 |
| 1.1611 | 9.7143 | 17 | 1.0844 | 0.4167 |
| 1.1611 | 10.8571 | 19 | 1.0476 | 0.5 |
| 0.9913 | 12.0 | 21 | 1.2315 | 0.4722 |
| 0.9913 | 12.5714 | 22 | 1.1444 | 0.4722 |
| 0.9913 | 13.7143 | 24 | 1.0242 | 0.5 |
| 0.9913 | 14.8571 | 26 | 1.0495 | 0.5278 |
| 0.9913 | 16.0 | 28 | 1.1234 | 0.4722 |
| 0.9913 | 16.5714 | 29 | 1.2332 | 0.5278 |
| 0.9206 | 17.7143 | 31 | 1.4389 | 0.3611 |
| 0.9206 | 18.8571 | 33 | 1.0300 | 0.5 |
| 0.9206 | 20.0 | 35 | 1.0028 | 0.5278 |
| 0.9206 | 20.5714 | 36 | 1.0322 | 0.5 |
| 0.9206 | 21.7143 | 38 | 1.0871 | 0.5278 |
| 0.7309 | 22.8571 | 40 | 0.9616 | 0.4722 |
| 0.7309 | 24.0 | 42 | 0.9571 | 0.5556 |
| 0.7309 | 24.5714 | 43 | 0.9855 | 0.5278 |
| 0.7309 | 25.7143 | 45 | 0.9598 | 0.5278 |
| 0.7309 | 26.8571 | 47 | 0.9774 | 0.5278 |
| 0.7309 | 28.0 | 49 | 0.9205 | 0.5556 |
| 0.6039 | 28.5714 | 50 | 0.9073 | 0.5556 |
| 0.6039 | 29.7143 | 52 | 0.8644 | 0.5833 |
| 0.6039 | 30.8571 | 54 | 0.8931 | 0.5833 |
| 0.6039 | 32.0 | 56 | 0.8686 | 0.6111 |
| 0.6039 | 32.5714 | 57 | 0.8381 | 0.5833 |
| 0.6039 | 33.7143 | 59 | 0.8658 | 0.5556 |
| 0.4784 | 34.8571 | 61 | 0.9915 | 0.5556 |
| 0.4784 | 36.0 | 63 | 0.7971 | 0.5833 |
| 0.4784 | 36.5714 | 64 | 0.7682 | 0.6111 |
| 0.4784 | 37.7143 | 66 | 0.9361 | 0.5833 |
| 0.4784 | 38.8571 | 68 | 0.9093 | 0.5833 |
| 0.4469 | 40.0 | 70 | 0.6728 | 0.75 |
| 0.4469 | 40.5714 | 71 | 0.6415 | 0.7222 |
| 0.4469 | 41.7143 | 73 | 0.7045 | 0.6667 |
| 0.4469 | 42.8571 | 75 | 0.8974 | 0.6389 |
| 0.4469 | 44.0 | 77 | 0.8032 | 0.6111 |
| 0.4469 | 44.5714 | 78 | 0.7134 | 0.6944 |
| 0.4329 | 45.7143 | 80 | 0.6975 | 0.7222 |
| 0.4329 | 46.8571 | 82 | 0.6758 | 0.7222 |
| 0.4329 | 48.0 | 84 | 0.8327 | 0.6111 |
| 0.4329 | 48.5714 | 85 | 0.9089 | 0.6111 |
| 0.4329 | 49.7143 | 87 | 0.9158 | 0.6111 |
| 0.4329 | 50.8571 | 89 | 0.8007 | 0.6389 |
| 0.4282 | 52.0 | 91 | 0.7363 | 0.6389 |
| 0.4282 | 52.5714 | 92 | 0.7378 | 0.6389 |
| 0.4282 | 53.7143 | 94 | 0.7449 | 0.6111 |
| 0.4282 | 54.8571 | 96 | 0.7605 | 0.6111 |
| 0.4282 | 56.0 | 98 | 0.7853 | 0.6111 |
| 0.4282 | 56.5714 | 99 | 0.7903 | 0.5833 |
| 0.3188 | 57.1429 | 100 | 0.7926 | 0.5833 |
Framework versions
- Transformers 4.41.0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1