<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
darija_english_V2
This model is a fine-tuned version of Helsinki-NLP/opus-mt-ar-en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.4036
- Bleu: 34.0871
- Gen Len: 11.2089
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: 6
- eval_batch_size: 6
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.5301 | 1.0 | 2576 | 1.4232 | 30.0353 | 11.585 |
| 1.2105 | 2.0 | 5152 | 1.3514 | 32.5286 | 11.4165 |
| 1.0023 | 3.0 | 7728 | 1.3325 | 33.6457 | 11.2198 |
| 0.86 | 4.0 | 10304 | 1.3465 | 33.1315 | 11.4064 |
| 0.7111 | 5.0 | 12880 | 1.3641 | 34.0438 | 11.1765 |
| 0.6332 | 6.0 | 15456 | 1.3834 | 34.4185 | 11.0991 |
| 0.591 | 7.0 | 18032 | 1.3966 | 34.0332 | 11.1895 |
| 0.5355 | 8.0 | 20608 | 1.4036 | 34.0871 | 11.2089 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
