<!-- 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-es-finetuned-es-to-kog
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: 2.0199
- Bleu: 1.7728
- Gen Len: 92.4257
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 | 2.7533 | 0.3297 | 124.1411 |
| No log | 2.0 | 398 | 2.4277 | 0.8514 | 111.2292 |
| 3.0539 | 3.0 | 597 | 2.2794 | 1.2877 | 99.2128 |
| 3.0539 | 4.0 | 796 | 2.1943 | 1.5011 | 97.7683 |
| 3.0539 | 5.0 | 995 | 2.1241 | 1.5278 | 95.4471 |
| 2.3246 | 6.0 | 1194 | 2.0844 | 1.9169 | 93.6738 |
| 2.3246 | 7.0 | 1393 | 2.0604 | 1.7087 | 92.2065 |
| 2.1625 | 8.0 | 1592 | 2.0352 | 1.9762 | 93.0164 |
| 2.1625 | 9.0 | 1791 | 2.0239 | 1.9638 | 93.1385 |
| 2.1625 | 10.0 | 1990 | 2.0199 | 1.7728 | 92.4257 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
