<!-- 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-3-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: 2.1003
- Bleu: 20.4006
- Meteor: 0.3752
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: 0.001
- 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 | 2.1601 | 8.8793 | 0.2344 |
No log | 2.0 | 404 | 1.8368 | 12.6317 | 0.2818 |
2.2884 | 3.0 | 606 | 1.7154 | 15.2843 | 0.3142 |
2.2884 | 4.0 | 808 | 1.6613 | 16.6089 | 0.332 |
1.4437 | 5.0 | 1010 | 1.6141 | 17.6498 | 0.3421 |
1.4437 | 6.0 | 1212 | 1.5957 | 18.0781 | 0.3491 |
1.4437 | 7.0 | 1414 | 1.5860 | 18.5857 | 0.3532 |
1.0727 | 8.0 | 1616 | 1.5875 | 19.1539 | 0.3605 |
1.0727 | 9.0 | 1818 | 1.6126 | 19.0598 | 0.3599 |
0.8623 | 10.0 | 2020 | 1.6371 | 19.2492 | 0.3608 |
0.8623 | 11.0 | 2222 | 1.6553 | 19.455 | 0.3626 |
0.8623 | 12.0 | 2424 | 1.6946 | 19.5787 | 0.365 |
0.6908 | 13.0 | 2626 | 1.7131 | 19.7264 | 0.3666 |
0.6908 | 14.0 | 2828 | 1.7351 | 19.9509 | 0.3695 |
0.5725 | 15.0 | 3030 | 1.7576 | 19.9999 | 0.369 |
0.5725 | 16.0 | 3232 | 1.7984 | 20.079 | 0.3707 |
0.5725 | 17.0 | 3434 | 1.8397 | 20.0174 | 0.3717 |
0.4773 | 18.0 | 3636 | 1.8763 | 19.7486 | 0.3683 |
0.4773 | 19.0 | 3838 | 1.9018 | 20.0435 | 0.3716 |
0.3998 | 20.0 | 4040 | 1.9268 | 20.2496 | 0.3735 |
0.3998 | 21.0 | 4242 | 1.9699 | 20.2014 | 0.3726 |
0.3998 | 22.0 | 4444 | 1.9935 | 20.0657 | 0.372 |
0.3426 | 23.0 | 4646 | 1.9934 | 20.1145 | 0.3734 |
0.3426 | 24.0 | 4848 | 2.0242 | 20.1118 | 0.3727 |
0.2999 | 25.0 | 5050 | 2.0496 | 20.1956 | 0.3734 |
0.2999 | 26.0 | 5252 | 2.0789 | 20.2959 | 0.3747 |
0.2999 | 27.0 | 5454 | 2.0850 | 20.2194 | 0.3738 |
0.2663 | 28.0 | 5656 | 2.0943 | 20.2931 | 0.3754 |
0.2663 | 29.0 | 5858 | 2.0941 | 20.3255 | 0.3745 |
0.2483 | 30.0 | 6060 | 2.1003 | 20.4006 | 0.3752 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2