metadata
license: apache-2.0
base_model: Helsinki-NLP/opus-mt-ar-en
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: ar-en-model
results: []
ar-en-model
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: 0.2988
- Bleu: 31.7475
- Gen Len: 33.5269
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 |
---|---|---|---|---|---|
No log | 1.0 | 312 | 0.3562 | 25.2145 | 33.397 |
0.6311 | 2.0 | 624 | 0.3311 | 27.4389 | 32.9487 |
0.6311 | 3.0 | 936 | 0.3178 | 28.887 | 33.6215 |
0.2523 | 4.0 | 1248 | 0.3102 | 29.9143 | 33.648 |
0.219 | 5.0 | 1560 | 0.3057 | 30.6256 | 33.4651 |
0.219 | 6.0 | 1872 | 0.3028 | 30.9421 | 33.4459 |
0.2025 | 7.0 | 2184 | 0.3008 | 31.5297 | 33.8605 |
0.2025 | 8.0 | 2496 | 0.2996 | 31.8919 | 33.8492 |
0.1895 | 9.0 | 2808 | 0.2989 | 31.7071 | 33.854 |
0.182 | 10.0 | 3120 | 0.2988 | 31.7475 | 33.5269 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.2.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1