test-hasy-3
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the HASY dataset. It achieves the following results on the evaluation set:
- Loss: 1.1780
- Accuracy: 0.7256
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: 16
- eval_batch_size: 16
- seed: 1787
- 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 |
|---|---|---|---|---|
| No log | 1.0 | 271 | 1.5038 | 0.6528 |
| 1.9073 | 2.0 | 542 | 1.4066 | 0.6694 |
| 1.9073 | 3.0 | 813 | 1.3478 | 0.6819 |
| 1.8233 | 4.0 | 1084 | 1.3028 | 0.6881 |
| 1.8233 | 5.0 | 1355 | 1.2658 | 0.7089 |
| 1.7738 | 6.0 | 1626 | 1.2334 | 0.7152 |
| 1.7738 | 7.0 | 1897 | 1.2151 | 0.7173 |
| 1.7149 | 8.0 | 2168 | 1.1893 | 0.7235 |
| 1.7149 | 9.0 | 2439 | 1.1848 | 0.7214 |
| 1.7125 | 10.0 | 2710 | 1.1780 | 0.7256 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 8
Model tree for khaingsmon/test-hasy-3
Base model
google/vit-base-patch16-224-in21k