<!-- 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-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: 1.8218
- Bleu: 0.1371
- Meteor: 0.294
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: 0.0001
- 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 | 2.6357 | 0.042 | 0.1513 |
| No log | 2.0 | 404 | 2.4891 | 0.0526 | 0.1749 |
| 2.781 | 3.0 | 606 | 2.3754 | 0.062 | 0.1918 |
| 2.781 | 4.0 | 808 | 2.2946 | 0.0693 | 0.2047 |
| 2.4692 | 5.0 | 1010 | 2.2262 | 0.0779 | 0.2175 |
| 2.4692 | 6.0 | 1212 | 2.1729 | 0.0825 | 0.2231 |
| 2.4692 | 7.0 | 1414 | 2.1226 | 0.0897 | 0.2328 |
| 2.2484 | 8.0 | 1616 | 2.0789 | 0.0932 | 0.2381 |
| 2.2484 | 9.0 | 1818 | 2.0450 | 0.1007 | 0.2478 |
| 2.099 | 10.0 | 2020 | 2.0132 | 0.1041 | 0.255 |
| 2.099 | 11.0 | 2222 | 1.9818 | 0.1085 | 0.2584 |
| 2.099 | 12.0 | 2424 | 1.9608 | 0.113 | 0.2639 |
| 1.9729 | 13.0 | 2626 | 1.9422 | 0.1165 | 0.2689 |
| 1.9729 | 14.0 | 2828 | 1.9223 | 0.1186 | 0.2717 |
| 1.8885 | 15.0 | 3030 | 1.9114 | 0.1219 | 0.2757 |
| 1.8885 | 16.0 | 3232 | 1.9020 | 0.1238 | 0.2794 |
| 1.8885 | 17.0 | 3434 | 1.8827 | 0.1254 | 0.2793 |
| 1.8171 | 18.0 | 3636 | 1.8762 | 0.1278 | 0.2824 |
| 1.8171 | 19.0 | 3838 | 1.8686 | 0.1298 | 0.285 |
| 1.7597 | 20.0 | 4040 | 1.8595 | 0.1307 | 0.2864 |
| 1.7597 | 21.0 | 4242 | 1.8533 | 0.1328 | 0.2891 |
| 1.7597 | 22.0 | 4444 | 1.8453 | 0.1335 | 0.2901 |
| 1.7183 | 23.0 | 4646 | 1.8400 | 0.1347 | 0.2912 |
| 1.7183 | 24.0 | 4848 | 1.8342 | 0.135 | 0.2914 |
| 1.6893 | 25.0 | 5050 | 1.8308 | 0.1355 | 0.2919 |
| 1.6893 | 26.0 | 5252 | 1.8258 | 0.1357 | 0.2924 |
| 1.6893 | 27.0 | 5454 | 1.8248 | 0.1365 | 0.2933 |
| 1.6667 | 28.0 | 5656 | 1.8233 | 0.137 | 0.294 |
| 1.6667 | 29.0 | 5858 | 1.8223 | 0.1371 | 0.2941 |
| 1.6585 | 30.0 | 6060 | 1.8218 | 0.1371 | 0.294 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.0
- Tokenizers 0.13.2
