<!-- 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-32
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.6636
- Bleu: 0.0405
- Meteor: 0.1481
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: 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: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Meteor |
|---|---|---|---|---|---|
| No log | 1.0 | 202 | 3.0655 | 0.0225 | 0.0979 |
| No log | 2.0 | 404 | 2.9938 | 0.0254 | 0.1076 |
| 3.3387 | 3.0 | 606 | 2.9513 | 0.0273 | 0.1124 |
| 3.3387 | 4.0 | 808 | 2.9152 | 0.0283 | 0.115 |
| 3.1851 | 5.0 | 1010 | 2.8882 | 0.0288 | 0.1167 |
| 3.1851 | 6.0 | 1212 | 2.8676 | 0.0299 | 0.1202 |
| 3.1851 | 7.0 | 1414 | 2.8431 | 0.0306 | 0.1215 |
| 3.1032 | 8.0 | 1616 | 2.8280 | 0.0313 | 0.124 |
| 3.1032 | 9.0 | 1818 | 2.8119 | 0.0323 | 0.1258 |
| 3.052 | 10.0 | 2020 | 2.7964 | 0.0326 | 0.1279 |
| 3.052 | 11.0 | 2222 | 2.7843 | 0.0334 | 0.1311 |
| 3.052 | 12.0 | 2424 | 2.7702 | 0.0344 | 0.1326 |
| 3.0051 | 13.0 | 2626 | 2.7596 | 0.035 | 0.1344 |
| 3.0051 | 14.0 | 2828 | 2.7462 | 0.0353 | 0.1351 |
| 2.9733 | 15.0 | 3030 | 2.7361 | 0.0361 | 0.1377 |
| 2.9733 | 16.0 | 3232 | 2.7277 | 0.0366 | 0.1389 |
| 2.9733 | 17.0 | 3434 | 2.7180 | 0.0372 | 0.1405 |
| 2.9451 | 18.0 | 3636 | 2.7108 | 0.0378 | 0.1417 |
| 2.9451 | 19.0 | 3838 | 2.7026 | 0.0382 | 0.143 |
| 2.9197 | 20.0 | 4040 | 2.6948 | 0.0384 | 0.1439 |
| 2.9197 | 21.0 | 4242 | 2.6904 | 0.0389 | 0.1448 |
| 2.9197 | 22.0 | 4444 | 2.6846 | 0.0395 | 0.1461 |
| 2.9004 | 23.0 | 4646 | 2.6792 | 0.0398 | 0.1466 |
| 2.9004 | 24.0 | 4848 | 2.6759 | 0.0401 | 0.1472 |
| 2.891 | 25.0 | 5050 | 2.6719 | 0.0403 | 0.1474 |
| 2.891 | 26.0 | 5252 | 2.6685 | 0.0405 | 0.1473 |
| 2.891 | 27.0 | 5454 | 2.6667 | 0.0408 | 0.1484 |
| 2.8783 | 28.0 | 5656 | 2.6650 | 0.0406 | 0.1481 |
| 2.8783 | 29.0 | 5858 | 2.6639 | 0.0406 | 0.1482 |
| 2.8784 | 30.0 | 6060 | 2.6636 | 0.0405 | 0.1481 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.0
- Tokenizers 0.13.2
