<!-- 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. -->
tokipona_model_v0.2
This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.9720
- Bleu: 11.6933
- Gen Len: 17.4433
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: 0.002
- 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: 20
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
---|---|---|---|---|---|
No log | 1.0 | 55 | 1.6640 | 0.2502 | 19.0 |
No log | 2.0 | 110 | 1.4393 | 3.6014 | 18.7685 |
No log | 3.0 | 165 | 1.3801 | 5.7834 | 17.7094 |
No log | 4.0 | 220 | 1.3014 | 6.6761 | 17.6798 |
No log | 5.0 | 275 | 1.3146 | 8.1891 | 17.3842 |
No log | 6.0 | 330 | 1.3197 | 7.9501 | 17.3399 |
No log | 7.0 | 385 | 1.3580 | 9.1859 | 17.5911 |
No log | 8.0 | 440 | 1.4023 | 10.4901 | 17.1823 |
No log | 9.0 | 495 | 1.4861 | 8.9722 | 17.4926 |
1.1 | 10.0 | 550 | 1.5296 | 10.0466 | 17.133 |
1.1 | 11.0 | 605 | 1.6126 | 9.7385 | 17.2217 |
1.1 | 12.0 | 660 | 1.7080 | 8.9794 | 17.3251 |
1.1 | 13.0 | 715 | 1.7778 | 10.5309 | 17.2759 |
1.1 | 14.0 | 770 | 1.8131 | 10.3548 | 17.202 |
1.1 | 15.0 | 825 | 1.8414 | 10.3668 | 17.468 |
1.1 | 16.0 | 880 | 1.8990 | 10.3459 | 17.4138 |
1.1 | 17.0 | 935 | 1.9324 | 11.7481 | 17.3448 |
1.1 | 18.0 | 990 | 1.9524 | 11.5339 | 17.2808 |
0.2722 | 19.0 | 1045 | 1.9732 | 11.1812 | 17.2365 |
0.2722 | 20.0 | 1100 | 1.9720 | 11.6933 | 17.4433 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3