<!-- 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. -->
translation_1
This model is a fine-tuned version of t5-base on the opus_books dataset. It achieves the following results on the evaluation set:
- Loss: 2.5005
- Bleu: 5.03
- Gen Len: 16.0434
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 3.4771 | 1.0 | 875 | 2.8863 | 2.2211 | 16.4083 |
| 2.9851 | 2.0 | 1750 | 2.7086 | 3.2474 | 16.4271 |
| 2.8137 | 3.0 | 2625 | 2.6050 | 3.8481 | 16.378 |
| 2.6962 | 4.0 | 3500 | 2.5268 | 4.3032 | 16.2734 |
| 2.6439 | 5.0 | 4375 | 2.4694 | 4.5354 | 16.2414 |
| 2.5633 | 6.0 | 5250 | 2.4227 | 4.8672 | 16.2483 |
| 2.5122 | 7.0 | 6125 | 2.4068 | 5.0916 | 16.256 |
| 2.693 | 8.0 | 7000 | 2.5069 | 5.0708 | 16.0449 |
| 2.6754 | 9.0 | 7875 | 2.5003 | 5.0422 | 16.0389 |
| 2.6714 | 10.0 | 8750 | 2.5003 | 5.0467 | 16.0357 |
| 2.6679 | 11.0 | 9625 | 2.5003 | 5.0482 | 16.038 |
| 2.6812 | 12.0 | 10500 | 2.5004 | 5.0456 | 16.0377 |
| 2.6733 | 13.0 | 11375 | 2.5004 | 5.0456 | 16.036 |
| 2.6802 | 14.0 | 12250 | 2.5005 | 5.0454 | 16.0403 |
| 2.6652 | 15.0 | 13125 | 2.5005 | 5.0403 | 16.0391 |
| 2.6718 | 16.0 | 14000 | 2.5005 | 5.0483 | 16.0403 |
| 2.6756 | 17.0 | 14875 | 2.5005 | 5.0324 | 16.04 |
| 2.6751 | 18.0 | 15750 | 2.5005 | 5.03 | 16.0434 |
| 2.6689 | 19.0 | 16625 | 2.5005 | 5.03 | 16.0434 |
| 2.6687 | 20.0 | 17500 | 2.5005 | 5.03 | 16.0434 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3
