<!-- 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_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: 1.8248
- F1: 0.8010
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 | 0.5418 | 0.7899 |
| 0.5486 | 2.0 | 578 | 0.6456 | 0.7727 |
| 0.5486 | 3.0 | 867 | 0.6705 | 0.8010 |
| 0.223 | 4.0 | 1156 | 0.8590 | 0.7998 |
| 0.223 | 5.0 | 1445 | 1.1503 | 0.7973 |
| 0.0936 | 6.0 | 1734 | 1.2678 | 0.7985 |
| 0.035 | 7.0 | 2023 | 1.3791 | 0.7973 |
| 0.035 | 8.0 | 2312 | 1.4720 | 0.7899 |
| 0.0249 | 9.0 | 2601 | 1.4755 | 0.7973 |
| 0.0249 | 10.0 | 2890 | 1.5383 | 0.7936 |
| 0.0177 | 11.0 | 3179 | 1.5926 | 0.8022 |
| 0.0177 | 12.0 | 3468 | 1.6685 | 0.8010 |
| 0.0124 | 13.0 | 3757 | 1.7205 | 0.7912 |
| 0.0047 | 14.0 | 4046 | 1.6311 | 0.7973 |
| 0.0047 | 15.0 | 4335 | 1.8000 | 0.7924 |
| 0.0051 | 16.0 | 4624 | 1.7546 | 0.7924 |
| 0.0051 | 17.0 | 4913 | 1.7800 | 0.7998 |
| 0.0018 | 18.0 | 5202 | 1.8507 | 0.7948 |
| 0.0018 | 19.0 | 5491 | 1.8112 | 0.7912 |
| 0.0052 | 20.0 | 5780 | 1.8149 | 0.7985 |
| 0.0026 | 21.0 | 6069 | 1.7938 | 0.7973 |
| 0.0026 | 22.0 | 6358 | 1.8643 | 0.8034 |
| 0.0013 | 23.0 | 6647 | 1.8311 | 0.7936 |
| 0.0013 | 24.0 | 6936 | 1.8298 | 0.7961 |
| 0.0031 | 25.0 | 7225 | 1.8248 | 0.8010 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
