opus-mt-en-es-finetuned-es-to-azz
This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-es on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.0112
- Bleu: 3.8287
- Gen Len: 93.0277
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: 32
- eval_batch_size: 32
- 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 | 199 | 2.8353 | 0.5636 | 134.5673 |
No log | 2.0 | 398 | 2.4731 | 1.5372 | 110.2931 |
3.173 | 3.0 | 597 | 2.3122 | 2.2405 | 103.761 |
3.173 | 4.0 | 796 | 2.2080 | 2.7367 | 97.6428 |
3.173 | 5.0 | 995 | 2.1358 | 3.0156 | 95.995 |
2.3722 | 6.0 | 1194 | 2.0908 | 3.417 | 94.6667 |
2.3722 | 7.0 | 1393 | 2.0513 | 3.6661 | 93.2415 |
2.184 | 8.0 | 1592 | 2.0306 | 3.5118 | 93.766 |
2.184 | 9.0 | 1791 | 2.0161 | 4.0349 | 93.3233 |
2.184 | 10.0 | 1990 | 2.0112 | 3.8287 | 93.0277 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
- Downloads last month
- 11