<!-- 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. -->
t5-base-finetuned-en-to-it-hrs
This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.4678
- Bleu: 22.3501
- Gen Len: 50.294
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: 40
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.4526 | 1.0 | 1125 | 1.9406 | 11.7289 | 57.5773 |
| 1.2548 | 2.0 | 2250 | 1.8509 | 14.9652 | 53.1013 |
| 1.1458 | 3.0 | 3375 | 1.7841 | 16.7549 | 52.4607 |
| 1.048 | 4.0 | 4500 | 1.7393 | 18.0223 | 51.4573 |
| 0.9922 | 5.0 | 5625 | 1.6980 | 18.6182 | 51.4733 |
| 0.9691 | 6.0 | 6750 | 1.6702 | 19.1118 | 51.994 |
| 0.9382 | 7.0 | 7875 | 1.6493 | 19.9025 | 51.128 |
| 0.8995 | 8.0 | 9000 | 1.6272 | 20.2594 | 51.2807 |
| 0.8843 | 9.0 | 10125 | 1.6106 | 20.4571 | 50.9607 |
| 0.8634 | 10.0 | 11250 | 1.5819 | 20.6829 | 51.0007 |
| 0.8507 | 11.0 | 12375 | 1.5752 | 20.6869 | 51.46 |
| 0.824 | 12.0 | 13500 | 1.5612 | 20.8633 | 51.2387 |
| 0.8124 | 13.0 | 14625 | 1.5496 | 21.3232 | 50.684 |
| 0.8081 | 14.0 | 15750 | 1.5425 | 21.4131 | 50.544 |
| 0.7837 | 15.0 | 16875 | 1.5302 | 21.2258 | 51.0287 |
| 0.7752 | 16.0 | 18000 | 1.5244 | 21.6548 | 50.312 |
| 0.7698 | 17.0 | 19125 | 1.5197 | 21.6719 | 50.7993 |
| 0.7606 | 18.0 | 20250 | 1.5168 | 21.7322 | 50.5947 |
| 0.7527 | 19.0 | 21375 | 1.5128 | 21.8434 | 50.4273 |
| 0.7515 | 20.0 | 22500 | 1.5008 | 21.6784 | 50.4933 |
| 0.7436 | 21.0 | 23625 | 1.5010 | 21.955 | 50.2093 |
| 0.7307 | 22.0 | 24750 | 1.4976 | 21.9676 | 50.7 |
| 0.7311 | 23.0 | 25875 | 1.4919 | 22.1018 | 50.5687 |
| 0.7206 | 24.0 | 27000 | 1.4890 | 22.0666 | 50.198 |
| 0.7142 | 25.0 | 28125 | 1.4843 | 22.1885 | 50.312 |
| 0.7125 | 26.0 | 29250 | 1.4796 | 22.1068 | 50.3167 |
| 0.7069 | 27.0 | 30375 | 1.4843 | 22.2135 | 50.144 |
| 0.701 | 28.0 | 31500 | 1.4761 | 22.168 | 50.574 |
| 0.6968 | 29.0 | 32625 | 1.4777 | 22.1219 | 50.5933 |
| 0.704 | 30.0 | 33750 | 1.4745 | 22.179 | 50.4773 |
| 0.698 | 31.0 | 34875 | 1.4733 | 22.1779 | 50.3713 |
| 0.6816 | 32.0 | 36000 | 1.4756 | 22.3355 | 50.3967 |
| 0.681 | 33.0 | 37125 | 1.4713 | 22.3124 | 50.192 |
| 0.6896 | 34.0 | 38250 | 1.4701 | 22.2848 | 50.1133 |
| 0.6798 | 35.0 | 39375 | 1.4677 | 22.2537 | 50.1573 |
| 0.6908 | 36.0 | 40500 | 1.4686 | 22.2789 | 50.202 |
| 0.6765 | 37.0 | 41625 | 1.4687 | 22.2854 | 50.1687 |
| 0.679 | 38.0 | 42750 | 1.4675 | 22.3388 | 50.3127 |
| 0.6788 | 39.0 | 43875 | 1.4672 | 22.2971 | 50.2687 |
| 0.6744 | 40.0 | 45000 | 1.4678 | 22.3501 | 50.294 |
Framework versions
- Transformers 4.22.1
- Pytorch 1.12.1
- Datasets 2.5.1
- Tokenizers 0.11.0
