<!-- 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. -->
spell_correction_M03_LM
This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0205
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: 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 |
|---|---|---|---|
| No log | 1.0 | 261 | 0.3991 |
| 1.9255 | 2.0 | 522 | 0.0588 |
| 1.9255 | 3.0 | 783 | 0.0308 |
| 0.05 | 4.0 | 1044 | 0.0239 |
| 0.05 | 5.0 | 1305 | 0.0205 |
| 0.0192 | 6.0 | 1566 | 0.0203 |
| 0.0192 | 7.0 | 1827 | 0.0201 |
| 0.012 | 8.0 | 2088 | 0.0191 |
| 0.012 | 9.0 | 2349 | 0.0199 |
| 0.0077 | 10.0 | 2610 | 0.0193 |
| 0.0077 | 11.0 | 2871 | 0.0189 |
| 0.0069 | 12.0 | 3132 | 0.0196 |
| 0.0069 | 13.0 | 3393 | 0.0203 |
| 0.0063 | 14.0 | 3654 | 0.0193 |
| 0.0063 | 15.0 | 3915 | 0.0198 |
| 0.0057 | 16.0 | 4176 | 0.0196 |
| 0.0057 | 17.0 | 4437 | 0.0198 |
| 0.0034 | 18.0 | 4698 | 0.0202 |
| 0.0034 | 19.0 | 4959 | 0.0204 |
| 0.0032 | 20.0 | 5220 | 0.0208 |
| 0.0032 | 21.0 | 5481 | 0.0206 |
| 0.0033 | 22.0 | 5742 | 0.0207 |
| 0.003 | 23.0 | 6003 | 0.0208 |
| 0.003 | 24.0 | 6264 | 0.0208 |
| 0.0025 | 25.0 | 6525 | 0.0206 |
| 0.0025 | 26.0 | 6786 | 0.0206 |
| 0.0029 | 27.0 | 7047 | 0.0205 |
| 0.0029 | 28.0 | 7308 | 0.0204 |
| 0.0026 | 29.0 | 7569 | 0.0206 |
| 0.0026 | 30.0 | 7830 | 0.0205 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.12.1+cu102
- Datasets 2.13.1
- Tokenizers 0.13.3
