metadata
library_name: transformers
license: mit
base_model: IAmSkyDra/BARTBana_v5
tags:
- generated_from_trainer
metrics:
- sacrebleu
model-index:
- name: BARTBana_Translation_v5
results: []
BARTBana_Translation_v5
This model is a fine-tuned version of IAmSkyDra/BARTBana_v5 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5602
- Sacrebleu: 7.0962
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
|---|---|---|---|---|
| 0.8121 | 1.0 | 742 | 0.7206 | 1.9030 |
| 0.7484 | 2.0 | 1484 | 0.6697 | 2.9685 |
| 0.6761 | 3.0 | 2226 | 0.6374 | 4.0287 |
| 0.6557 | 4.0 | 2968 | 0.6162 | 4.8115 |
| 0.6185 | 5.0 | 3710 | 0.6006 | 5.2849 |
| 0.6033 | 6.0 | 4452 | 0.5889 | 5.7585 |
| 0.573 | 7.0 | 5194 | 0.5807 | 6.1391 |
| 0.5573 | 8.0 | 5936 | 0.5758 | 6.3215 |
| 0.5491 | 9.0 | 6678 | 0.5688 | 6.4682 |
| 0.534 | 10.0 | 7420 | 0.5653 | 6.7433 |
| 0.5204 | 11.0 | 8162 | 0.5630 | 6.8459 |
| 0.5144 | 12.0 | 8904 | 0.5620 | 6.9725 |
| 0.5069 | 13.0 | 9646 | 0.5602 | 7.0593 |
| 0.5021 | 14.0 | 10388 | 0.5596 | 7.0546 |
| 0.4929 | 15.0 | 11130 | 0.5602 | 7.0962 |
Framework versions
- Transformers 4.48.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0