---
base_model: dmis-lab/biobert-base-cased-v1.1
tags:
- generated_from_trainer
model-index:
- name: ner-cdr-finetuned
  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. -->

# ner-cdr-finetuned

This model is a fine-tuned version of [dmis-lab/biobert-base-cased-v1.1](https://huggingface.co/dmis-lab/biobert-base-cased-v1.1) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1871
- Chemical: {'precision': 0.9215938303341902, 'recall': 0.9263565891472868, 'f1': 0.9239690721649483, 'number': 774}
- Disease: {'precision': 0.7865546218487395, 'recall': 0.8327402135231317, 'f1': 0.8089887640449438, 'number': 562}
- Overall Precision: 0.8631
- Overall Recall: 0.8870
- Overall F1: 0.8749
- Overall Accuracy: 0.9480

## 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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | Chemical                                                                                                 | Disease                                                                                                  | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
| No log        | 1.0   | 57   | 0.2923          | {'precision': 0.8410757946210269, 'recall': 0.8888888888888888, 'f1': 0.864321608040201, 'number': 774}  | {'precision': 0.5264483627204031, 'recall': 0.7437722419928826, 'f1': 0.616519174041298, 'number': 562}  | 0.6861            | 0.8278         | 0.7503     | 0.8980           |
| No log        | 2.0   | 114  | 0.2079          | {'precision': 0.9134487350199734, 'recall': 0.8863049095607235, 'f1': 0.8996721311475411, 'number': 774} | {'precision': 0.7549909255898367, 'recall': 0.7402135231316725, 'f1': 0.7475292003593891, 'number': 562} | 0.8464            | 0.8249         | 0.8355     | 0.9339           |
| No log        | 3.0   | 171  | 0.1910          | {'precision': 0.9088575096277278, 'recall': 0.9147286821705426, 'f1': 0.9117836445589182, 'number': 774} | {'precision': 0.7578125, 'recall': 0.8629893238434164, 'f1': 0.8069883527454244, 'number': 562}          | 0.8407            | 0.8930         | 0.8661     | 0.9464           |
| No log        | 4.0   | 228  | 0.1856          | {'precision': 0.9296875, 'recall': 0.9224806201550387, 'f1': 0.9260700389105059, 'number': 774}          | {'precision': 0.770764119601329, 'recall': 0.8256227758007118, 'f1': 0.7972508591065292, 'number': 562}  | 0.8599            | 0.8817         | 0.8707     | 0.9461           |
| No log        | 5.0   | 285  | 0.1871          | {'precision': 0.9215938303341902, 'recall': 0.9263565891472868, 'f1': 0.9239690721649483, 'number': 774} | {'precision': 0.7865546218487395, 'recall': 0.8327402135231317, 'f1': 0.8089887640449438, 'number': 562} | 0.8631            | 0.8870         | 0.8749     | 0.9480           |


### Framework versions

- Transformers 4.41.1
- Pytorch 2.7.0+cu118
- Datasets 3.6.0
- Tokenizers 0.19.1