<!-- 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_8_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: 1.5821
- F1: 0.8265
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 | 0.3933 | 0.8222 |
| 0.4092 | 2.0 | 580 | 0.4431 | 0.8237 |
| 0.4092 | 3.0 | 870 | 0.6243 | 0.8292 |
| 0.1845 | 4.0 | 1160 | 0.6526 | 0.8300 |
| 0.1845 | 5.0 | 1450 | 0.9229 | 0.8203 |
| 0.0671 | 6.0 | 1740 | 0.9436 | 0.8279 |
| 0.0303 | 7.0 | 2030 | 1.1281 | 0.8260 |
| 0.0303 | 8.0 | 2320 | 1.1676 | 0.8327 |
| 0.0105 | 9.0 | 2610 | 1.2557 | 0.8291 |
| 0.0105 | 10.0 | 2900 | 1.3556 | 0.8326 |
| 0.0102 | 11.0 | 3190 | 1.3160 | 0.8413 |
| 0.0102 | 12.0 | 3480 | 1.3199 | 0.8344 |
| 0.0068 | 13.0 | 3770 | 1.3827 | 0.8314 |
| 0.0049 | 14.0 | 4060 | 1.5265 | 0.8197 |
| 0.0049 | 15.0 | 4350 | 1.5481 | 0.8215 |
| 0.0069 | 16.0 | 4640 | 1.3824 | 0.8292 |
| 0.0069 | 17.0 | 4930 | 1.4398 | 0.8305 |
| 0.0073 | 18.0 | 5220 | 1.5004 | 0.8255 |
| 0.0033 | 19.0 | 5510 | 1.5322 | 0.8253 |
| 0.0033 | 20.0 | 5800 | 1.5239 | 0.8237 |
| 0.0025 | 21.0 | 6090 | 1.5299 | 0.8286 |
| 0.0025 | 22.0 | 6380 | 1.5788 | 0.8271 |
| 0.0005 | 23.0 | 6670 | 1.5903 | 0.8298 |
| 0.0005 | 24.0 | 6960 | 1.5893 | 0.8232 |
| 0.0026 | 25.0 | 7250 | 1.5821 | 0.8265 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
