<!-- 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_M04_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: 0.0178
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 | 269 | 0.2687 |
| 1.8467 | 2.0 | 538 | 0.0361 |
| 1.8467 | 3.0 | 807 | 0.0241 |
| 0.0357 | 4.0 | 1076 | 0.0198 |
| 0.0357 | 5.0 | 1345 | 0.0199 |
| 0.0159 | 6.0 | 1614 | 0.0175 |
| 0.0159 | 7.0 | 1883 | 0.0179 |
| 0.0077 | 8.0 | 2152 | 0.0189 |
| 0.0077 | 9.0 | 2421 | 0.0183 |
| 0.006 | 10.0 | 2690 | 0.0183 |
| 0.006 | 11.0 | 2959 | 0.0191 |
| 0.0044 | 12.0 | 3228 | 0.0186 |
| 0.0044 | 13.0 | 3497 | 0.0192 |
| 0.0033 | 14.0 | 3766 | 0.0189 |
| 0.0024 | 15.0 | 4035 | 0.0173 |
| 0.0024 | 16.0 | 4304 | 0.0171 |
| 0.0026 | 17.0 | 4573 | 0.0183 |
| 0.0026 | 18.0 | 4842 | 0.0181 |
| 0.0021 | 19.0 | 5111 | 0.0177 |
| 0.0021 | 20.0 | 5380 | 0.0174 |
| 0.0015 | 21.0 | 5649 | 0.0173 |
| 0.0015 | 22.0 | 5918 | 0.0174 |
| 0.0016 | 23.0 | 6187 | 0.0178 |
| 0.0016 | 24.0 | 6456 | 0.0180 |
| 0.0018 | 25.0 | 6725 | 0.0175 |
| 0.0018 | 26.0 | 6994 | 0.0171 |
| 0.0017 | 27.0 | 7263 | 0.0175 |
| 0.0014 | 28.0 | 7532 | 0.0177 |
| 0.0014 | 29.0 | 7801 | 0.0178 |
| 0.0013 | 30.0 | 8070 | 0.0178 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.12.1+cu102
- Datasets 2.13.1
- Tokenizers 0.13.3
