vit-base-patch16-224-in21k-rotated-dungeons-v22
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the rotated_maps dataset. It achieves the following results on the evaluation set:
- Loss: 0.5265
- Accuracy: 0.8571
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: 2e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 1337
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.1127 | 4.3478 | 100 | 0.5795 | 0.8571 |
| 0.0985 | 8.6957 | 200 | 0.5159 | 0.8571 |
| 0.0799 | 13.0435 | 300 | 0.5685 | 0.8571 |
| 0.078 | 17.3913 | 400 | 0.5435 | 0.8571 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.19.1