<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
vit-base-patch16-224-finetuned-eurosat
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.6991
- Accuracy: 0.7842
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.5297 | 0.99 | 106 | 1.2732 | 0.6121 |
| 1.0644 | 2.0 | 213 | 0.9503 | 0.7008 |
| 0.8303 | 3.0 | 320 | 0.8145 | 0.7374 |
| 0.7296 | 4.0 | 427 | 0.7382 | 0.7688 |
| 0.6729 | 4.99 | 533 | 0.7149 | 0.7714 |
| 0.612 | 5.96 | 636 | 0.6991 | 0.7842 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
