metadata
license: cc-by-4.0
base_model: Helsinki-NLP/opus-mt-tc-big-en-ar
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: english-to-darija-2
results: []
english-to-darija-2
This model is a fine-tuned version of Helsinki-NLP/opus-mt-tc-big-en-ar on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0309
- Bleu: 66.9765
- Gen Len: 15.4554
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: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 2.5337 | 1.0 | 1392 | 2.0006 | 17.4965 | 15.0162 |
| 1.5827 | 2.0 | 2784 | 1.5484 | 26.9067 | 14.9165 |
| 1.133 | 3.0 | 4176 | 1.3098 | 38.9189 | 15.0717 |
| 0.8616 | 4.0 | 5568 | 1.1757 | 49.8352 | 15.2675 |
| 0.6862 | 5.0 | 6960 | 1.1013 | 57.7662 | 15.2131 |
| 0.5583 | 6.0 | 8352 | 1.0651 | 62.2156 | 15.2488 |
| 0.4734 | 7.0 | 9744 | 1.0460 | 65.0696 | 15.3751 |
| 0.4122 | 8.0 | 11136 | 1.0370 | 65.317 | 15.4677 |
| 0.3734 | 9.0 | 12528 | 1.0326 | 66.5766 | 15.4436 |
| 0.3354 | 10.0 | 13920 | 1.0309 | 66.9765 | 15.4554 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0