BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext-finetuned-ner-30-v2
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1852
- Precision: 0.7087
- Recall: 0.8328
- F1: 0.7657
- Accuracy: 0.9473
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 OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 239 | 0.1539 | 0.6390 | 0.8594 | 0.7330 | 0.9409 |
| No log | 2.0 | 478 | 0.1355 | 0.6693 | 0.8223 | 0.7380 | 0.9438 |
| 0.2038 | 3.0 | 717 | 0.1333 | 0.7145 | 0.7808 | 0.7462 | 0.9460 |
| 0.2038 | 4.0 | 956 | 0.1428 | 0.6857 | 0.8467 | 0.7577 | 0.9454 |
| 0.0881 | 5.0 | 1195 | 0.1539 | 0.7317 | 0.7941 | 0.7616 | 0.9484 |
| 0.0881 | 6.0 | 1434 | 0.1852 | 0.7087 | 0.8328 | 0.7657 | 0.9473 |
| 0.0515 | 7.0 | 1673 | 0.2085 | 0.7061 | 0.8234 | 0.7603 | 0.9460 |
| 0.0515 | 8.0 | 1912 | 0.2273 | 0.7134 | 0.7938 | 0.7515 | 0.9448 |
| 0.0272 | 9.0 | 2151 | 0.2724 | 0.7149 | 0.8126 | 0.7606 | 0.9475 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.1