metadata
license: other
tags:
- generated_from_keras_callback
model-index:
- name: ChristianMDahl/segFormer-b3-horizontal-vertical
results: []
ChristianMDahl/segFormer-b3-horizontal-vertical
This model is a fine-tuned version of nvidia/mit-b3 on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.1745
- Validation Loss: 0.2206
- Epoch: 16
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:
- optimizer: {'name': 'Adam', 'learning_rate': 6e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: float32
Training results
| Train Loss | Validation Loss | Epoch |
|---|---|---|
| 0.3203 | 0.2831 | 0 |
| 0.2822 | 0.2688 | 1 |
| 0.2662 | 0.2578 | 2 |
| 0.2526 | 0.2484 | 3 |
| 0.2396 | 0.2442 | 4 |
| 0.2288 | 0.2416 | 5 |
| 0.2195 | 0.2381 | 6 |
| 0.2121 | 0.2361 | 7 |
| 0.2058 | 0.2314 | 8 |
| 0.1999 | 0.2277 | 9 |
| 0.1952 | 0.2287 | 10 |
| 0.1912 | 0.2221 | 11 |
| 0.1869 | 0.2205 | 12 |
| 0.1835 | 0.2226 | 13 |
| 0.1804 | 0.2209 | 14 |
| 0.1775 | 0.2181 | 15 |
| 0.1745 | 0.2206 | 16 |
Framework versions
- Transformers 4.29.2
- TensorFlow 2.10.1
- Tokenizers 0.13.3