<!-- 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. -->
opus-mt-en-de-finetuned-en-to-de
This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-de on the wmt16 dataset. It achieves the following results on the evaluation set:
- Loss: 1.2849
- Bleu: 30.529
- Rougelsum: 0.5587
- Gen Len: 27.0521
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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|
| 1.5584 | 1.0 | 12500 | 1.2921 | 30.5519 | 0.5601 | 27.0549 |
| 1.5649 | 2.0 | 25000 | 1.2877 | 30.578 | 0.5591 | 27.0415 |
| 1.5686 | 3.0 | 37500 | 1.2859 | 30.5509 | 0.5591 | 27.0401 |
| 1.5507 | 4.0 | 50000 | 1.2851 | 30.5396 | 0.5589 | 27.0526 |
| 1.5532 | 5.0 | 62500 | 1.2849 | 30.529 | 0.5587 | 27.0521 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
