vit-base-patch16-224-RU2-10
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: 0.6429
- Accuracy: 0.85
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: 5.5e-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: 0.05
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.1641 | 0.99 | 38 | 0.9789 | 0.7333 |
0.5847 | 2.0 | 77 | 0.6371 | 0.8167 |
0.2844 | 2.99 | 115 | 0.6706 | 0.75 |
0.2275 | 4.0 | 154 | 0.5359 | 0.8167 |
0.1539 | 4.99 | 192 | 0.6067 | 0.8167 |
0.1113 | 6.0 | 231 | 0.7887 | 0.7667 |
0.1117 | 6.99 | 269 | 0.6443 | 0.8167 |
0.1088 | 8.0 | 308 | 0.6429 | 0.85 |
0.0824 | 8.99 | 346 | 0.6499 | 0.8333 |
0.0834 | 9.87 | 380 | 0.6802 | 0.8167 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 2
Finetuned from
Evaluation results
- Accuracy on imagefoldervalidation set self-reported0.850