beit-base-patch16-224-hasta-55-fold5
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: 1.4708
- Accuracy: 0.6111
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 | 1.2240 | 0.3056 |
| No log | 1.7143 | 3 | 1.2263 | 0.3056 |
| No log | 2.8571 | 5 | 1.2222 | 0.3889 |
| No log | 4.0 | 7 | 1.1690 | 0.3889 |
| No log | 4.5714 | 8 | 1.1691 | 0.3889 |
| 1.1249 | 5.7143 | 10 | 1.0999 | 0.3889 |
| 1.1249 | 6.8571 | 12 | 1.1605 | 0.4167 |
| 1.1249 | 8.0 | 14 | 1.1912 | 0.4167 |
| 1.1249 | 8.5714 | 15 | 1.1771 | 0.3889 |
| 1.1249 | 9.7143 | 17 | 1.2370 | 0.4722 |
| 1.1249 | 10.8571 | 19 | 1.2607 | 0.5 |
| 0.9274 | 12.0 | 21 | 1.2756 | 0.4722 |
| 0.9274 | 12.5714 | 22 | 1.2208 | 0.4722 |
| 0.9274 | 13.7143 | 24 | 1.3705 | 0.5 |
| 0.9274 | 14.8571 | 26 | 1.2191 | 0.5278 |
| 0.9274 | 16.0 | 28 | 1.3502 | 0.5278 |
| 0.9274 | 16.5714 | 29 | 1.2628 | 0.5278 |
| 0.7889 | 17.7143 | 31 | 1.0868 | 0.5 |
| 0.7889 | 18.8571 | 33 | 1.3983 | 0.5 |
| 0.7889 | 20.0 | 35 | 1.2537 | 0.5556 |
| 0.7889 | 20.5714 | 36 | 1.1540 | 0.4722 |
| 0.7889 | 21.7143 | 38 | 1.2135 | 0.5556 |
| 0.7027 | 22.8571 | 40 | 1.4271 | 0.5 |
| 0.7027 | 24.0 | 42 | 1.1828 | 0.5 |
| 0.7027 | 24.5714 | 43 | 1.2126 | 0.4444 |
| 0.7027 | 25.7143 | 45 | 1.4980 | 0.5556 |
| 0.7027 | 26.8571 | 47 | 1.3495 | 0.5556 |
| 0.7027 | 28.0 | 49 | 1.1969 | 0.5278 |
| 0.6037 | 28.5714 | 50 | 1.2063 | 0.5556 |
| 0.6037 | 29.7143 | 52 | 1.3115 | 0.5833 |
| 0.6037 | 30.8571 | 54 | 1.1726 | 0.5278 |
| 0.6037 | 32.0 | 56 | 1.1872 | 0.5556 |
| 0.6037 | 32.5714 | 57 | 1.2399 | 0.5556 |
| 0.6037 | 33.7143 | 59 | 1.2566 | 0.5278 |
| 0.5147 | 34.8571 | 61 | 1.1848 | 0.5278 |
| 0.5147 | 36.0 | 63 | 1.2614 | 0.5556 |
| 0.5147 | 36.5714 | 64 | 1.3975 | 0.5556 |
| 0.5147 | 37.7143 | 66 | 1.4708 | 0.6111 |
| 0.5147 | 38.8571 | 68 | 1.3233 | 0.5833 |
| 0.4004 | 40.0 | 70 | 1.2994 | 0.5556 |
| 0.4004 | 40.5714 | 71 | 1.3582 | 0.5278 |
| 0.4004 | 41.7143 | 73 | 1.3577 | 0.5278 |
| 0.4004 | 42.8571 | 75 | 1.1985 | 0.5833 |
| 0.4004 | 44.0 | 77 | 1.1448 | 0.5556 |
| 0.4004 | 44.5714 | 78 | 1.1714 | 0.6111 |
| 0.4323 | 45.7143 | 80 | 1.3707 | 0.6111 |
| 0.4323 | 46.8571 | 82 | 1.5477 | 0.5833 |
| 0.4323 | 48.0 | 84 | 1.4254 | 0.5833 |
| 0.4323 | 48.5714 | 85 | 1.3031 | 0.5833 |
| 0.4323 | 49.7143 | 87 | 1.1843 | 0.6111 |
| 0.4323 | 50.8571 | 89 | 1.1835 | 0.6111 |
| 0.3568 | 52.0 | 91 | 1.2399 | 0.6111 |
| 0.3568 | 52.5714 | 92 | 1.2606 | 0.6111 |
| 0.3568 | 53.7143 | 94 | 1.2997 | 0.5833 |
| 0.3568 | 54.8571 | 96 | 1.3184 | 0.5833 |
| 0.3568 | 56.0 | 98 | 1.3294 | 0.5833 |
| 0.3568 | 56.5714 | 99 | 1.3337 | 0.5833 |
| 0.3308 | 57.1429 | 100 | 1.3367 | 0.5833 |
Framework versions
- Transformers 4.41.0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1