| --- |
| library_name: transformers |
| base_model: dmis-lab/biobert-base-cased-v1.1 |
| tags: |
| - generated_from_trainer |
| metrics: |
| - accuracy |
| - precision |
| - recall |
| - f1 |
| model-index: |
| - name: Biobert_combo_v2 |
| results: [] |
| --- |
| |
| <!-- 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. --> |
|
|
| # Biobert_combo_v2 |
|
|
| This model is a fine-tuned version of [dmis-lab/biobert-base-cased-v1.1](https://huggingface.co/dmis-lab/biobert-base-cased-v1.1) on the None dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.1933 |
| - Accuracy: 0.924 |
| - Auc: 0.978 |
| - Precision: 0.938 |
| - Recall: 0.938 |
| - F1: 0.938 |
| - F1-macro: 0.919 |
| - F1-micro: 0.924 |
| - F1-weighted: 0.924 |
|
|
| ## 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: 8 |
| - eval_batch_size: 8 |
| - seed: 42 |
| - gradient_accumulation_steps: 4 |
| - 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 | |
| |:-------------:|:------:|:----:|:---------------:|:--------:|:-----:|:---------:|:------:|:-----:|:--------:|:--------:|:-----------:| |
| | 0.4506 | 0.2661 | 500 | 0.2929 | 0.883 | 0.944 | 0.902 | 0.91 | 0.906 | 0.876 | 0.883 | 0.883 | |
| | 0.2846 | 0.5323 | 1000 | 0.2606 | 0.897 | 0.957 | 0.904 | 0.934 | 0.918 | 0.89 | 0.897 | 0.897 | |
| | 0.2462 | 0.7984 | 1500 | 0.2316 | 0.907 | 0.966 | 0.915 | 0.938 | 0.926 | 0.901 | 0.907 | 0.907 | |
| | 0.2337 | 1.0644 | 2000 | 0.2297 | 0.91 | 0.967 | 0.926 | 0.928 | 0.927 | 0.904 | 0.91 | 0.91 | |
| | 0.21 | 1.3305 | 2500 | 0.2212 | 0.911 | 0.97 | 0.934 | 0.922 | 0.928 | 0.906 | 0.911 | 0.911 | |
| | 0.2033 | 1.5967 | 3000 | 0.2181 | 0.913 | 0.972 | 0.925 | 0.935 | 0.93 | 0.908 | 0.913 | 0.913 | |
| | 0.2029 | 1.8628 | 3500 | 0.2109 | 0.916 | 0.974 | 0.92 | 0.946 | 0.933 | 0.91 | 0.916 | 0.915 | |
| | 0.1948 | 2.1288 | 4000 | 0.2030 | 0.921 | 0.975 | 0.94 | 0.931 | 0.935 | 0.916 | 0.921 | 0.921 | |
| | 0.1812 | 2.3949 | 4500 | 0.2093 | 0.918 | 0.974 | 0.933 | 0.935 | 0.934 | 0.913 | 0.918 | 0.918 | |
| | 0.1822 | 2.6611 | 5000 | 0.1996 | 0.92 | 0.976 | 0.933 | 0.939 | 0.936 | 0.916 | 0.92 | 0.92 | |
| | 0.1818 | 2.9272 | 5500 | 0.1945 | 0.923 | 0.977 | 0.936 | 0.94 | 0.938 | 0.918 | 0.923 | 0.923 | |
| | 0.1707 | 3.1932 | 6000 | 0.1955 | 0.923 | 0.977 | 0.946 | 0.929 | 0.937 | 0.919 | 0.923 | 0.923 | |
| | 0.1635 | 3.4593 | 6500 | 0.2019 | 0.922 | 0.977 | 0.935 | 0.939 | 0.937 | 0.917 | 0.922 | 0.922 | |
| | 0.1747 | 3.7255 | 7000 | 0.1983 | 0.923 | 0.977 | 0.931 | 0.945 | 0.938 | 0.918 | 0.923 | 0.923 | |
| | 0.1735 | 3.9916 | 7500 | 0.1956 | 0.923 | 0.978 | 0.936 | 0.941 | 0.938 | 0.919 | 0.923 | 0.923 | |
| | 0.1646 | 4.2576 | 8000 | 0.1994 | 0.921 | 0.977 | 0.933 | 0.94 | 0.937 | 0.916 | 0.921 | 0.921 | |
| | 0.1616 | 4.5238 | 8500 | 0.1925 | 0.924 | 0.978 | 0.942 | 0.934 | 0.938 | 0.919 | 0.924 | 0.924 | |
| | 0.1615 | 4.7899 | 9000 | 0.1933 | 0.924 | 0.978 | 0.938 | 0.938 | 0.938 | 0.919 | 0.924 | 0.924 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.53.2 |
| - Pytorch 2.6.0+cu124 |
| - Datasets 2.14.4 |
| - Tokenizers 0.21.2 |
| |
