stbl_clinical_bert_ft_rs8
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: 0.1016
- F1: 0.9189
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 |
|---|---|---|---|---|
| 0.283 | 1.0 | 101 | 0.1003 | 0.8531 |
| 0.0682 | 2.0 | 202 | 0.0709 | 0.8999 |
| 0.0293 | 3.0 | 303 | 0.0782 | 0.9149 |
| 0.0172 | 4.0 | 404 | 0.0731 | 0.9160 |
| 0.0089 | 5.0 | 505 | 0.0827 | 0.9142 |
| 0.0056 | 6.0 | 606 | 0.0878 | 0.9199 |
| 0.0033 | 7.0 | 707 | 0.0953 | 0.9193 |
| 0.0024 | 8.0 | 808 | 0.0961 | 0.9200 |
| 0.0021 | 9.0 | 909 | 0.0947 | 0.9209 |
| 0.0018 | 10.0 | 1010 | 0.0980 | 0.9195 |
| 0.0011 | 11.0 | 1111 | 0.1013 | 0.9178 |
| 0.0011 | 12.0 | 1212 | 0.1016 | 0.9189 |
Framework versions
- Transformers 4.22.2
- Pytorch 1.12.1+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1