<!-- 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. -->
train_model_yonsei
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.5148
- Accuracy: 0.87
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: 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.5711 | 0.98 | 11 | 1.4796 | 0.69 |
| 1.3855 | 1.96 | 22 | 1.2302 | 0.74 |
| 1.1544 | 2.93 | 33 | 1.0229 | 0.77 |
| 0.9292 | 4.0 | 45 | 0.8371 | 0.8 |
| 0.7715 | 4.98 | 56 | 0.7186 | 0.84 |
| 0.6521 | 5.96 | 67 | 0.6353 | 0.85 |
| 0.5736 | 6.93 | 78 | 0.5895 | 0.86 |
| 0.4745 | 8.0 | 90 | 0.5891 | 0.85 |
| 0.4361 | 8.98 | 101 | 0.5370 | 0.87 |
| 0.4431 | 9.78 | 110 | 0.5148 | 0.87 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
