<!-- 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. -->
nmt-mpst-id-en-lr_1e-05-ep_30-seq_128_bs-16
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.5612
- Bleu: 0.0476
- Meteor: 0.1643
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: 1e-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: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Meteor |
|---|---|---|---|---|---|
| No log | 1.0 | 404 | 3.1116 | 0.0198 | 0.0892 |
| 3.6027 | 2.0 | 808 | 2.9875 | 0.0255 | 0.1079 |
| 3.2803 | 3.0 | 1212 | 2.9296 | 0.0276 | 0.1135 |
| 3.1743 | 4.0 | 1616 | 2.8869 | 0.0287 | 0.116 |
| 3.1283 | 5.0 | 2020 | 2.8564 | 0.03 | 0.1208 |
| 3.1283 | 6.0 | 2424 | 2.8257 | 0.0309 | 0.1237 |
| 3.0739 | 7.0 | 2828 | 2.8007 | 0.0324 | 0.1281 |
| 3.0296 | 8.0 | 3232 | 2.7758 | 0.0334 | 0.131 |
| 3.0059 | 9.0 | 3636 | 2.7548 | 0.0346 | 0.134 |
| 2.965 | 10.0 | 4040 | 2.7349 | 0.0362 | 0.1371 |
| 2.965 | 11.0 | 4444 | 2.7176 | 0.0374 | 0.1403 |
| 2.9403 | 12.0 | 4848 | 2.6994 | 0.0382 | 0.1425 |
| 2.9166 | 13.0 | 5252 | 2.6841 | 0.0393 | 0.1448 |
| 2.9023 | 14.0 | 5656 | 2.6681 | 0.0404 | 0.1471 |
| 2.8742 | 15.0 | 6060 | 2.6548 | 0.0411 | 0.1508 |
| 2.8742 | 16.0 | 6464 | 2.6419 | 0.0422 | 0.1529 |
| 2.8523 | 17.0 | 6868 | 2.6286 | 0.0428 | 0.1538 |
| 2.8378 | 18.0 | 7272 | 2.6194 | 0.0434 | 0.1555 |
| 2.8258 | 19.0 | 7676 | 2.6095 | 0.0441 | 0.1568 |
| 2.8019 | 20.0 | 8080 | 2.6005 | 0.0447 | 0.1576 |
| 2.8019 | 21.0 | 8484 | 2.5938 | 0.0455 | 0.1598 |
| 2.7927 | 22.0 | 8888 | 2.5872 | 0.0459 | 0.1603 |
| 2.7846 | 23.0 | 9292 | 2.5800 | 0.0462 | 0.161 |
| 2.7775 | 24.0 | 9696 | 2.5757 | 0.0463 | 0.1621 |
| 2.77 | 25.0 | 10100 | 2.5712 | 0.0466 | 0.1624 |
| 2.7608 | 26.0 | 10504 | 2.5673 | 0.0469 | 0.1633 |
| 2.7608 | 27.0 | 10908 | 2.5645 | 0.0472 | 0.1634 |
| 2.7572 | 28.0 | 11312 | 2.5626 | 0.0474 | 0.1637 |
| 2.7578 | 29.0 | 11716 | 2.5617 | 0.0476 | 0.1641 |
| 2.7568 | 30.0 | 12120 | 2.5612 | 0.0476 | 0.1643 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.0
- Tokenizers 0.13.2
