raygx/Albert-Bhai-Nepali
This model is a fine-tuned version of Shushant/nepaliBERT on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 7.8042
- Validation Loss: 7.7339
- Epoch: 9
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': 'AdamWeightDecay', 'learning_rate': 5e-06, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.001}
- training_precision: float32
Training results
| Train Loss | Validation Loss | Epoch |
|---|---|---|
| 9.9774 | 9.7237 | 0 |
| 9.5636 | 9.3764 | 1 |
| 9.2524 | 9.0893 | 2 |
| 8.9711 | 8.8166 | 3 |
| 8.7139 | 8.5716 | 4 |
| 8.4774 | 8.3449 | 5 |
| 8.2628 | 8.1481 | 6 |
| 8.0779 | 7.9643 | 7 |
| 7.9268 | 7.8574 | 8 |
| 7.8042 | 7.7339 | 9 |
Framework versions
- Transformers 4.29.2
- TensorFlow 2.12.0
- Datasets 2.12.0
- Tokenizers 0.13.3