metadata
tags:
- generated_from_trainer
datasets:
- ncbi_disease
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: biobert-base-cased-v1.2_ncbi_disease-sm-first-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: ncbi_disease
type: ncbi_disease
args: ncbi_disease
metrics:
- name: Precision
type: precision
value: 0.8522139160437032
- name: Recall
type: recall
value: 0.8826682549136391
- name: F1
type: f1
value: 0.8671737858396723
- name: Accuracy
type: accuracy
value: 0.9826972482743678
biobert-base-cased-v1.2_ncbi_disease-sm-first-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: 0.0865
- Precision: 0.8522
- Recall: 0.8827
- F1: 0.8672
- Accuracy: 0.9827
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 |
|---|---|---|---|---|---|---|---|
| 0.0858 | 1.0 | 1359 | 0.0985 | 0.7929 | 0.8005 | 0.7967 | 0.9730 |
| 0.042 | 2.0 | 2718 | 0.0748 | 0.8449 | 0.8856 | 0.8648 | 0.9820 |
| 0.0124 | 3.0 | 4077 | 0.0865 | 0.8522 | 0.8827 | 0.8672 | 0.9827 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.2+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1