<!-- 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_2_binary_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: 1.9317
- F1: 0.7921
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 | 0.4221 | 0.7856 |
| 0.4062 | 2.0 | 580 | 0.5184 | 0.7949 |
| 0.4062 | 3.0 | 870 | 0.6854 | 0.7840 |
| 0.1775 | 4.0 | 1160 | 0.9834 | 0.7840 |
| 0.1775 | 5.0 | 1450 | 1.3223 | 0.7804 |
| 0.0697 | 6.0 | 1740 | 1.2896 | 0.7923 |
| 0.0265 | 7.0 | 2030 | 1.4620 | 0.7914 |
| 0.0265 | 8.0 | 2320 | 1.5554 | 0.7835 |
| 0.0102 | 9.0 | 2610 | 1.7009 | 0.7880 |
| 0.0102 | 10.0 | 2900 | 1.6163 | 0.7923 |
| 0.015 | 11.0 | 3190 | 1.6851 | 0.7841 |
| 0.015 | 12.0 | 3480 | 1.7493 | 0.7901 |
| 0.0141 | 13.0 | 3770 | 1.8819 | 0.7827 |
| 0.0133 | 14.0 | 4060 | 1.7535 | 0.7939 |
| 0.0133 | 15.0 | 4350 | 1.6613 | 0.7966 |
| 0.0067 | 16.0 | 4640 | 1.6807 | 0.7999 |
| 0.0067 | 17.0 | 4930 | 1.6703 | 0.7978 |
| 0.0053 | 18.0 | 5220 | 1.7309 | 0.8013 |
| 0.0037 | 19.0 | 5510 | 1.8058 | 0.7942 |
| 0.0037 | 20.0 | 5800 | 1.8233 | 0.7916 |
| 0.0023 | 21.0 | 6090 | 1.8206 | 0.7913 |
| 0.0023 | 22.0 | 6380 | 1.8466 | 0.7949 |
| 0.0012 | 23.0 | 6670 | 1.8531 | 0.7985 |
| 0.0012 | 24.0 | 6960 | 1.9211 | 0.7944 |
| 0.0001 | 25.0 | 7250 | 1.9317 | 0.7921 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
