stbl_clinical_bert_ft_rs5
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: 0.0936
- F1: 0.9268
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.2723 | 1.0 | 101 | 0.0875 | 0.8479 |
| 0.066 | 2.0 | 202 | 0.0688 | 0.9002 |
| 0.0328 | 3.0 | 303 | 0.0668 | 0.9070 |
| 0.0179 | 4.0 | 404 | 0.0689 | 0.9129 |
| 0.0098 | 5.0 | 505 | 0.0790 | 0.9147 |
| 0.0069 | 6.0 | 606 | 0.0805 | 0.9205 |
| 0.0033 | 7.0 | 707 | 0.0835 | 0.9268 |
| 0.0022 | 8.0 | 808 | 0.0904 | 0.9262 |
| 0.0021 | 9.0 | 909 | 0.0882 | 0.9263 |
| 0.0015 | 10.0 | 1010 | 0.0933 | 0.9289 |
| 0.0009 | 11.0 | 1111 | 0.0921 | 0.9311 |
| 0.0009 | 12.0 | 1212 | 0.0936 | 0.9268 |
Framework versions
- Transformers 4.22.1
- Pytorch 1.12.1+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1