metadata
license: mit
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: Bio_ClinicalBERT_fold_4_binary_v1
results: []
Bio_ClinicalBERT_fold_4_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: 1.4627
- F1: 0.8342
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 | 289 | 0.3641 | 0.8394 |
| 0.3953 | 2.0 | 578 | 0.3729 | 0.8294 |
| 0.3953 | 3.0 | 867 | 0.6156 | 0.8126 |
| 0.189 | 4.0 | 1156 | 0.7389 | 0.8326 |
| 0.189 | 5.0 | 1445 | 0.8925 | 0.8322 |
| 0.0783 | 6.0 | 1734 | 1.0909 | 0.8196 |
| 0.0219 | 7.0 | 2023 | 1.1241 | 0.8346 |
| 0.0219 | 8.0 | 2312 | 1.2684 | 0.8130 |
| 0.0136 | 9.0 | 2601 | 1.2615 | 0.8202 |
| 0.0136 | 10.0 | 2890 | 1.2477 | 0.8401 |
| 0.0143 | 11.0 | 3179 | 1.3211 | 0.8254 |
| 0.0143 | 12.0 | 3468 | 1.2627 | 0.8286 |
| 0.0165 | 13.0 | 3757 | 1.3804 | 0.8264 |
| 0.006 | 14.0 | 4046 | 1.3213 | 0.8414 |
| 0.006 | 15.0 | 4335 | 1.3152 | 0.8427 |
| 0.0117 | 16.0 | 4624 | 1.3373 | 0.8368 |
| 0.0117 | 17.0 | 4913 | 1.3599 | 0.8406 |
| 0.0021 | 18.0 | 5202 | 1.4072 | 0.8237 |
| 0.0021 | 19.0 | 5491 | 1.3893 | 0.8336 |
| 0.0045 | 20.0 | 5780 | 1.4331 | 0.8391 |
| 0.0049 | 21.0 | 6069 | 1.4128 | 0.8370 |
| 0.0049 | 22.0 | 6358 | 1.4660 | 0.8356 |
| 0.0029 | 23.0 | 6647 | 1.4721 | 0.8388 |
| 0.0029 | 24.0 | 6936 | 1.4636 | 0.8329 |
| 0.0023 | 25.0 | 7225 | 1.4627 | 0.8342 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1