biobert-base-cased-v1.2_ncbi_disease-softmax-labelall-ner
This model is a fine-tuned version of dmis-lab/biobert-base-cased-v1.2 on the ncbi_disease dataset. It achieves the following results on the evaluation set:
- Loss: 0.0629
- Precision: 0.8289
- Recall: 0.8615
- F1: 0.8449
- Accuracy: 0.9861
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0554 | 1.0 | 1359 | 0.0659 | 0.7814 | 0.8132 | 0.7970 | 0.9825 |
| 0.0297 | 2.0 | 2718 | 0.0445 | 0.8284 | 0.8895 | 0.8578 | 0.9876 |
| 0.0075 | 3.0 | 4077 | 0.0629 | 0.8289 | 0.8615 | 0.8449 | 0.9861 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.2+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
- Downloads last month
- 114
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Dataset used to train jordyvl/biobert-base-cased-v1.2_ncbi_disease-softmax-labelall-ner
Evaluation results
- Precision on ncbi_diseaseself-reported0.829
- Recall on ncbi_diseaseself-reported0.861
- F1 on ncbi_diseaseself-reported0.845
- Accuracy on ncbi_diseaseself-reported0.986