basso4_eng_to_vie_model
This model was trained from scratch on the mt_eng_vietnamese dataset. It achieves the following results on the evaluation set:
- Loss: 0.9999
- Bleu: 1.3777
- Gen Len: 18.7953
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.3485 | 1.0 | 6666 | 1.2151 | 0.9294 | 18.7893 |
| 1.2969 | 2.0 | 13332 | 1.1510 | 1.0593 | 18.7881 |
| 1.2426 | 3.0 | 19998 | 1.1053 | 1.1598 | 18.7873 |
| 1.2067 | 4.0 | 26664 | 1.0713 | 1.2393 | 18.789 |
| 1.1809 | 5.0 | 33330 | 1.0469 | 1.2853 | 18.7894 |
| 1.1575 | 6.0 | 39996 | 1.0288 | 1.3227 | 18.7899 |
| 1.1407 | 7.0 | 46662 | 1.0154 | 1.3405 | 18.7937 |
| 1.1295 | 8.0 | 53328 | 1.0071 | 1.3657 | 18.7943 |
| 1.1241 | 9.0 | 59994 | 1.0014 | 1.3753 | 18.7953 |
| 1.1209 | 10.0 | 66660 | 0.9999 | 1.3777 | 18.7953 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3