<!-- 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_M02_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: 0.0138
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 | 263 | 0.3208 |
| 1.8633 | 2.0 | 526 | 0.0425 |
| 1.8633 | 3.0 | 789 | 0.0253 |
| 0.037 | 4.0 | 1052 | 0.0210 |
| 0.037 | 5.0 | 1315 | 0.0172 |
| 0.015 | 6.0 | 1578 | 0.0162 |
| 0.015 | 7.0 | 1841 | 0.0156 |
| 0.01 | 8.0 | 2104 | 0.0151 |
| 0.01 | 9.0 | 2367 | 0.0150 |
| 0.0063 | 10.0 | 2630 | 0.0153 |
| 0.0063 | 11.0 | 2893 | 0.0141 |
| 0.0049 | 12.0 | 3156 | 0.0140 |
| 0.0049 | 13.0 | 3419 | 0.0150 |
| 0.0039 | 14.0 | 3682 | 0.0136 |
| 0.0039 | 15.0 | 3945 | 0.0147 |
| 0.0027 | 16.0 | 4208 | 0.0142 |
| 0.0027 | 17.0 | 4471 | 0.0128 |
| 0.0028 | 18.0 | 4734 | 0.0128 |
| 0.0028 | 19.0 | 4997 | 0.0128 |
| 0.0029 | 20.0 | 5260 | 0.0131 |
| 0.0022 | 21.0 | 5523 | 0.0139 |
| 0.0022 | 22.0 | 5786 | 0.0144 |
| 0.0018 | 23.0 | 6049 | 0.0136 |
| 0.0018 | 24.0 | 6312 | 0.0134 |
| 0.002 | 25.0 | 6575 | 0.0143 |
| 0.002 | 26.0 | 6838 | 0.0137 |
| 0.0016 | 27.0 | 7101 | 0.0137 |
| 0.0016 | 28.0 | 7364 | 0.0140 |
| 0.0018 | 29.0 | 7627 | 0.0139 |
| 0.0018 | 30.0 | 7890 | 0.0138 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.12.1+cu102
- Datasets 2.13.1
- Tokenizers 0.13.3
