BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext-finetuned-ner-30
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: 0.1185
- Precision: 0.7453
- Recall: 0.8757
- F1: 0.8053
- Accuracy: 0.9574
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 | 364 | 0.1245 | 0.7017 | 0.8745 | 0.7786 | 0.9518 |
| 0.1928 | 2.0 | 728 | 0.1148 | 0.7292 | 0.8577 | 0.7882 | 0.9550 |
| 0.0979 | 3.0 | 1092 | 0.1165 | 0.7243 | 0.8842 | 0.7963 | 0.9562 |
| 0.0979 | 4.0 | 1456 | 0.1185 | 0.7453 | 0.8757 | 0.8053 | 0.9574 |
| 0.0701 | 5.0 | 1820 | 0.1417 | 0.7635 | 0.8327 | 0.7966 | 0.9550 |
| 0.0455 | 6.0 | 2184 | 0.1593 | 0.7625 | 0.8263 | 0.7931 | 0.9556 |
| 0.03 | 7.0 | 2548 | 0.1737 | 0.7686 | 0.8390 | 0.8023 | 0.9568 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.1