<!-- 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_5_ternary_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: 2.0233
- F1: 0.7849
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 | 291 | 0.6352 | 0.7286 |
| 0.5477 | 2.0 | 582 | 0.5965 | 0.7682 |
| 0.5477 | 3.0 | 873 | 0.7696 | 0.7554 |
| 0.2383 | 4.0 | 1164 | 1.0119 | 0.7631 |
| 0.2383 | 5.0 | 1455 | 1.1300 | 0.7772 |
| 0.1068 | 6.0 | 1746 | 1.3515 | 0.7734 |
| 0.0401 | 7.0 | 2037 | 1.4935 | 0.7721 |
| 0.0401 | 8.0 | 2328 | 1.5418 | 0.7875 |
| 0.0213 | 9.0 | 2619 | 1.6902 | 0.7746 |
| 0.0213 | 10.0 | 2910 | 1.7091 | 0.7721 |
| 0.014 | 11.0 | 3201 | 1.7422 | 0.7836 |
| 0.014 | 12.0 | 3492 | 1.8603 | 0.7772 |
| 0.012 | 13.0 | 3783 | 1.8419 | 0.7734 |
| 0.0104 | 14.0 | 4074 | 1.9616 | 0.7657 |
| 0.0104 | 15.0 | 4365 | 1.9342 | 0.7823 |
| 0.005 | 16.0 | 4656 | 1.9646 | 0.7746 |
| 0.005 | 17.0 | 4947 | 1.9943 | 0.7772 |
| 0.0075 | 18.0 | 5238 | 1.9882 | 0.7798 |
| 0.0071 | 19.0 | 5529 | 1.9909 | 0.7849 |
| 0.0071 | 20.0 | 5820 | 2.0568 | 0.7798 |
| 0.0029 | 21.0 | 6111 | 2.0508 | 0.7759 |
| 0.0029 | 22.0 | 6402 | 2.0267 | 0.7823 |
| 0.0047 | 23.0 | 6693 | 2.0534 | 0.7785 |
| 0.0047 | 24.0 | 6984 | 2.0336 | 0.7849 |
| 0.0014 | 25.0 | 7275 | 2.0233 | 0.7849 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
