<!-- 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. -->
europarl-de-en
This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-de on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.4246
- Bleu: 24.8548
- Gen Len: 26.14
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
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 107 | 1.4276 | 24.5835 | 26.1167 |
| No log | 2.0 | 214 | 1.4223 | 24.8629 | 26.26 |
| No log | 3.0 | 321 | 1.4217 | 24.9621 | 26.0933 |
| No log | 4.0 | 428 | 1.4226 | 25.092 | 26.13 |
| 1.2607 | 5.0 | 535 | 1.4246 | 24.8548 | 26.14 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.10.0+cu113
- Datasets 2.9.0
- Tokenizers 0.13.2
