bert-biobert
This model is a fine-tuned version of bert-base-uncased on the biobert_json dataset. It achieves the following results on the evaluation set:
- Loss: 0.1108
- Precision: 0.9431
- Recall: 0.9681
- F1: 0.9554
- Accuracy: 0.9766
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: Use 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: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.4604 | 1.0 | 612 | 0.1196 | 0.9239 | 0.9553 | 0.9393 | 0.9694 |
0.1399 | 2.0 | 1224 | 0.1072 | 0.9252 | 0.9724 | 0.9482 | 0.9726 |
0.0938 | 3.0 | 1836 | 0.0939 | 0.9425 | 0.9703 | 0.9562 | 0.9777 |
0.076 | 4.0 | 2448 | 0.0996 | 0.9393 | 0.9729 | 0.9558 | 0.9771 |
0.0543 | 5.0 | 3060 | 0.0976 | 0.9397 | 0.9699 | 0.9546 | 0.9766 |
0.0475 | 6.0 | 3672 | 0.1072 | 0.9392 | 0.9698 | 0.9543 | 0.9751 |
0.0399 | 7.0 | 4284 | 0.1060 | 0.9410 | 0.9698 | 0.9552 | 0.9765 |
0.038 | 8.0 | 4896 | 0.1097 | 0.9411 | 0.9707 | 0.9557 | 0.9770 |
0.0301 | 9.0 | 5508 | 0.1087 | 0.9438 | 0.9692 | 0.9563 | 0.9769 |
0.0281 | 10.0 | 6120 | 0.1108 | 0.9431 | 0.9681 | 0.9554 | 0.9766 |
Framework versions
- Transformers 4.46.2
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3
- Downloads last month
- 6
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for stivenacua17/bert-biobert
Base model
google-bert/bert-base-uncasedEvaluation results
- Precision on biobert_jsonvalidation set self-reported0.943
- Recall on biobert_jsonvalidation set self-reported0.968
- F1 on biobert_jsonvalidation set self-reported0.955
- Accuracy on biobert_jsonvalidation set self-reported0.977