<!-- 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. -->
Nso-En_update3
This model is a fine-tuned version of Helsinki-NLP/opus-mt-nso-en on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.6854
- Bleu: 21.2223
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: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu |
|---|---|---|---|---|
| 3.5054 | 1.0 | 1734 | 3.1912 | 16.5243 |
| 3.0368 | 2.0 | 3468 | 2.9680 | 18.0237 |
| 2.6866 | 3.0 | 5202 | 2.8594 | 19.5515 |
| 2.51 | 4.0 | 6936 | 2.7916 | 20.1468 |
| 2.3754 | 5.0 | 8670 | 2.7438 | 20.0535 |
| 2.2534 | 6.0 | 10404 | 2.7186 | 20.7329 |
| 2.144 | 7.0 | 12138 | 2.7034 | 20.9116 |
| 2.0709 | 8.0 | 13872 | 2.6945 | 21.0866 |
| 2.0191 | 9.0 | 15606 | 2.6880 | 21.1577 |
| 1.9973 | 10.0 | 17340 | 2.6854 | 21.1386 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
