metadata
library_name: transformers
base_model: dmis-lab/biobert-base-cased-v1.2
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: BioBert_Medhhml
results: []
BioBert_Medhhml
This model is a fine-tuned version of dmis-lab/biobert-base-cased-v1.2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1382
- Accuracy: 0.776
- Auc: 0.896
- Precision: 0.867
- Recall: 0.632
- F1: 0.731
- F1-macro: 0.769
- F1-micro: 0.776
- F1-weighted: 0.771
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Auc | Precision | Recall | F1 | F1-macro | F1-micro | F1-weighted |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.3638 | 1.0 | 260 | 0.4250 | 0.78 | 0.908 | 0.913 | 0.603 | 0.726 | 0.771 | 0.78 | 0.773 |
| 0.2463 | 2.0 | 520 | 1.7793 | 0.645 | 0.652 | 0.811 | 0.347 | 0.486 | 0.607 | 0.645 | 0.611 |
| 0.1798 | 3.0 | 780 | 0.5889 | 0.768 | 0.926 | 0.883 | 0.6 | 0.715 | 0.76 | 0.768 | 0.761 |
| 0.1203 | 4.0 | 1040 | 0.6651 | 0.824 | 0.935 | 0.895 | 0.72 | 0.798 | 0.821 | 0.824 | 0.821 |
| 0.0785 | 5.0 | 1300 | 1.1382 | 0.776 | 0.896 | 0.867 | 0.632 | 0.731 | 0.769 | 0.776 | 0.771 |
Framework versions
- Transformers 4.53.0
- Pytorch 2.6.0+cu124
- Datasets 2.14.4
- Tokenizers 0.21.2