<!-- 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_F01_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: 0.0147
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 | 294 | 0.1889 |
| 1.8488 | 2.0 | 588 | 0.0392 |
| 1.8488 | 3.0 | 882 | 0.0260 |
| 0.0369 | 4.0 | 1176 | 0.0212 |
| 0.0369 | 5.0 | 1470 | 0.0170 |
| 0.0182 | 6.0 | 1764 | 0.0162 |
| 0.0105 | 7.0 | 2058 | 0.0173 |
| 0.0105 | 8.0 | 2352 | 0.0177 |
| 0.0083 | 9.0 | 2646 | 0.0176 |
| 0.0083 | 10.0 | 2940 | 0.0164 |
| 0.0057 | 11.0 | 3234 | 0.0149 |
| 0.0037 | 12.0 | 3528 | 0.0168 |
| 0.0037 | 13.0 | 3822 | 0.0164 |
| 0.003 | 14.0 | 4116 | 0.0156 |
| 0.003 | 15.0 | 4410 | 0.0151 |
| 0.003 | 16.0 | 4704 | 0.0152 |
| 0.003 | 17.0 | 4998 | 0.0149 |
| 0.0028 | 18.0 | 5292 | 0.0153 |
| 0.002 | 19.0 | 5586 | 0.0146 |
| 0.002 | 20.0 | 5880 | 0.0147 |
| 0.0021 | 21.0 | 6174 | 0.0146 |
| 0.0021 | 22.0 | 6468 | 0.0149 |
| 0.0025 | 23.0 | 6762 | 0.0146 |
| 0.0018 | 24.0 | 7056 | 0.0150 |
| 0.0018 | 25.0 | 7350 | 0.0147 |
| 0.0021 | 26.0 | 7644 | 0.0145 |
| 0.0021 | 27.0 | 7938 | 0.0147 |
| 0.0017 | 28.0 | 8232 | 0.0146 |
| 0.0014 | 29.0 | 8526 | 0.0147 |
| 0.0014 | 30.0 | 8820 | 0.0147 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.12.1+cu102
- Datasets 2.13.1
- Tokenizers 0.13.3
