<!-- 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_0.0001-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: 1.7399
- Bleu: 0.1512
- Meteor: 0.3113
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: 0.0001
- 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 | 2.7159 | 0.0379 | 0.1409 |
| 3.1207 | 2.0 | 808 | 2.4769 | 0.0518 | 0.1728 |
| 2.7244 | 3.0 | 1212 | 2.3261 | 0.0667 | 0.1988 |
| 2.4735 | 4.0 | 1616 | 2.2345 | 0.0783 | 0.2172 |
| 2.3206 | 5.0 | 2020 | 2.1375 | 0.0876 | 0.2305 |
| 2.3206 | 6.0 | 2424 | 2.0815 | 0.0952 | 0.241 |
| 2.1735 | 7.0 | 2828 | 2.0141 | 0.1027 | 0.2506 |
| 2.0667 | 8.0 | 3232 | 1.9810 | 0.1107 | 0.2603 |
| 1.9851 | 9.0 | 3636 | 1.9394 | 0.1157 | 0.2664 |
| 1.9022 | 10.0 | 4040 | 1.9124 | 0.1205 | 0.2721 |
| 1.9022 | 11.0 | 4444 | 1.8832 | 0.1239 | 0.2776 |
| 1.8341 | 12.0 | 4848 | 1.8702 | 0.1287 | 0.2835 |
| 1.7756 | 13.0 | 5252 | 1.8432 | 0.1328 | 0.2881 |
| 1.7389 | 14.0 | 5656 | 1.8327 | 0.1338 | 0.2898 |
| 1.6861 | 15.0 | 6060 | 1.8225 | 0.1372 | 0.2934 |
| 1.6861 | 16.0 | 6464 | 1.8020 | 0.1391 | 0.2956 |
| 1.647 | 17.0 | 6868 | 1.7913 | 0.14 | 0.2973 |
| 1.6145 | 18.0 | 7272 | 1.7842 | 0.1434 | 0.302 |
| 1.585 | 19.0 | 7676 | 1.7746 | 0.144 | 0.3015 |
| 1.554 | 20.0 | 8080 | 1.7699 | 0.1442 | 0.3028 |
| 1.554 | 21.0 | 8484 | 1.7654 | 0.1462 | 0.3051 |
| 1.5332 | 22.0 | 8888 | 1.7573 | 0.147 | 0.3062 |
| 1.5109 | 23.0 | 9292 | 1.7537 | 0.1489 | 0.3083 |
| 1.4981 | 24.0 | 9696 | 1.7507 | 0.1493 | 0.3089 |
| 1.4834 | 25.0 | 10100 | 1.7445 | 0.1501 | 0.3094 |
| 1.4732 | 26.0 | 10504 | 1.7427 | 0.1502 | 0.3095 |
| 1.4732 | 27.0 | 10908 | 1.7414 | 0.1502 | 0.3102 |
| 1.4623 | 28.0 | 11312 | 1.7417 | 0.1511 | 0.3111 |
| 1.4601 | 29.0 | 11716 | 1.7405 | 0.1512 | 0.3114 |
| 1.4543 | 30.0 | 12120 | 1.7399 | 0.1512 | 0.3113 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.0
- Tokenizers 0.13.2
