metadata
license: apache-2.0
base_model: WinKawaks/vit-tiny-patch16-224
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: vit-tiny-patch16-224-finetuned-RESISC45_01
results: []
vit-tiny-patch16-224-finetuned-RESISC45_01
This model is a fine-tuned version of WinKawaks/vit-tiny-patch16-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2402
- Accuracy: 0.9302
- Precision: 0.9317
- Recall: 0.9302
- F1: 0.9301
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: 0.0001
- train_batch_size: 512
- eval_batch_size: 512
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 3.9864 | 1.0 | 37 | 2.0458 | 0.643 | 0.6573 | 0.643 | 0.6131 |
| 0.8947 | 2.0 | 74 | 0.5364 | 0.873 | 0.8821 | 0.873 | 0.8720 |
| 0.5981 | 3.0 | 111 | 0.3644 | 0.907 | 0.9137 | 0.907 | 0.9068 |
| 0.46 | 4.0 | 148 | 0.2821 | 0.914 | 0.9209 | 0.914 | 0.9130 |
| 0.3936 | 5.0 | 185 | 0.2343 | 0.929 | 0.9331 | 0.929 | 0.9289 |
| 0.3629 | 6.0 | 222 | 0.2191 | 0.935 | 0.9404 | 0.935 | 0.9351 |
| 0.3154 | 7.0 | 259 | 0.2000 | 0.939 | 0.9424 | 0.939 | 0.9388 |
| 0.317 | 8.0 | 296 | 0.1736 | 0.952 | 0.9548 | 0.952 | 0.9520 |
| 0.2921 | 9.0 | 333 | 0.1725 | 0.952 | 0.9545 | 0.952 | 0.9519 |
| 0.2922 | 10.0 | 370 | 0.1738 | 0.945 | 0.9481 | 0.945 | 0.9449 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1