|
--- |
|
library_name: transformers |
|
license: cc0-1.0 |
|
base_model: bionlp/bluebert_pubmed_uncased_L-24_H-1024_A-16 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: medical-ner-bluebert |
|
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. --> |
|
|
|
# medical-ner-bluebert |
|
|
|
This model is a fine-tuned version of [bionlp/bluebert_pubmed_uncased_L-24_H-1024_A-16](https://huggingface.co/bionlp/bluebert_pubmed_uncased_L-24_H-1024_A-16) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0894 |
|
- Precision: 0.9365 |
|
- Recall: 0.9705 |
|
- F1: 0.9532 |
|
- Accuracy: 0.9810 |
|
|
|
## 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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: cosine |
|
- num_epochs: 20 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 1.0 | 90 | 0.7138 | 0.5501 | 0.6082 | 0.5777 | 0.7757 | |
|
| No log | 2.0 | 180 | 0.6697 | 0.5391 | 0.7574 | 0.6298 | 0.7863 | |
|
| No log | 3.0 | 270 | 0.4996 | 0.6284 | 0.8024 | 0.7048 | 0.8429 | |
|
| No log | 4.0 | 360 | 0.3957 | 0.6779 | 0.8332 | 0.7476 | 0.8760 | |
|
| No log | 5.0 | 450 | 0.2886 | 0.7603 | 0.8658 | 0.8096 | 0.9160 | |
|
| 0.4932 | 6.0 | 540 | 0.2345 | 0.8026 | 0.8839 | 0.8413 | 0.9321 | |
|
| 0.4932 | 7.0 | 630 | 0.2061 | 0.8290 | 0.9121 | 0.8686 | 0.9419 | |
|
| 0.4932 | 8.0 | 720 | 0.1715 | 0.8537 | 0.9226 | 0.8868 | 0.9518 | |
|
| 0.4932 | 9.0 | 810 | 0.1454 | 0.8701 | 0.9374 | 0.9025 | 0.9603 | |
|
| 0.4932 | 10.0 | 900 | 0.1422 | 0.8857 | 0.9437 | 0.9137 | 0.9635 | |
|
| 0.4932 | 11.0 | 990 | 0.1134 | 0.9081 | 0.9516 | 0.9293 | 0.9718 | |
|
| 0.0935 | 12.0 | 1080 | 0.1147 | 0.9075 | 0.9582 | 0.9322 | 0.9722 | |
|
| 0.0935 | 13.0 | 1170 | 0.1039 | 0.9165 | 0.9616 | 0.9385 | 0.9757 | |
|
| 0.0935 | 14.0 | 1260 | 0.0978 | 0.9256 | 0.9658 | 0.9453 | 0.9774 | |
|
| 0.0935 | 15.0 | 1350 | 0.0925 | 0.9283 | 0.9671 | 0.9473 | 0.9797 | |
|
| 0.0935 | 16.0 | 1440 | 0.0873 | 0.9378 | 0.9679 | 0.9526 | 0.9813 | |
|
| 0.0301 | 17.0 | 1530 | 0.0927 | 0.9334 | 0.9703 | 0.9515 | 0.9803 | |
|
| 0.0301 | 18.0 | 1620 | 0.0903 | 0.9355 | 0.97 | 0.9525 | 0.9804 | |
|
| 0.0301 | 19.0 | 1710 | 0.0890 | 0.9373 | 0.9711 | 0.9539 | 0.9811 | |
|
| 0.0301 | 20.0 | 1800 | 0.0894 | 0.9365 | 0.9705 | 0.9532 | 0.9810 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.2 |
|
- Pytorch 2.5.1+cu121 |
|
- Datasets 3.1.0 |
|
- Tokenizers 0.20.3 |
|
|
