<!-- 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. -->
Bio_ClinicalBERT_fold_10_ternary_v1
This model is a fine-tuned version of emilyalsentzer/Bio_ClinicalBERT on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.0706
- F1: 0.7748
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: 2e-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: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| No log | 1.0 | 290 | 0.6097 | 0.7290 |
| 0.555 | 2.0 | 580 | 0.6106 | 0.7649 |
| 0.555 | 3.0 | 870 | 0.6608 | 0.7847 |
| 0.2449 | 4.0 | 1160 | 0.8894 | 0.7809 |
| 0.2449 | 5.0 | 1450 | 1.1049 | 0.7760 |
| 0.1055 | 6.0 | 1740 | 1.2951 | 0.7884 |
| 0.0338 | 7.0 | 2030 | 1.4809 | 0.7760 |
| 0.0338 | 8.0 | 2320 | 1.4751 | 0.7698 |
| 0.0225 | 9.0 | 2610 | 1.6648 | 0.7809 |
| 0.0225 | 10.0 | 2900 | 1.7174 | 0.7772 |
| 0.006 | 11.0 | 3190 | 1.7872 | 0.7735 |
| 0.006 | 12.0 | 3480 | 1.7803 | 0.7748 |
| 0.0161 | 13.0 | 3770 | 1.9302 | 0.7735 |
| 0.0005 | 14.0 | 4060 | 1.9853 | 0.7748 |
| 0.0005 | 15.0 | 4350 | 2.0043 | 0.7735 |
| 0.0062 | 16.0 | 4640 | 1.9969 | 0.7760 |
| 0.0062 | 17.0 | 4930 | 2.0173 | 0.7760 |
| 0.0068 | 18.0 | 5220 | 1.9891 | 0.7785 |
| 0.0034 | 19.0 | 5510 | 1.9951 | 0.7797 |
| 0.0034 | 20.0 | 5800 | 2.0283 | 0.7748 |
| 0.0049 | 21.0 | 6090 | 1.9985 | 0.7834 |
| 0.0049 | 22.0 | 6380 | 2.0131 | 0.7760 |
| 0.0011 | 23.0 | 6670 | 2.0526 | 0.7748 |
| 0.0011 | 24.0 | 6960 | 2.0662 | 0.7748 |
| 0.001 | 25.0 | 7250 | 2.0706 | 0.7748 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
