metadata
tags:
- generated_from_trainer
datasets:
- jnlpba
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: biobert-base-cased-v1.2-finetuned-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: jnlpba
type: jnlpba
args: jnlpba
metrics:
- name: Precision
type: precision
value: 0.8948080842655547
- name: Recall
type: recall
value: 0.9282417121275703
- name: F1
type: f1
value: 0.9112183219652858
- name: Accuracy
type: accuracy
value: 0.9601644367242017
biobert-base-cased-v1.2-finetuned-ner
This model is a fine-tuned version of dmis-lab/biobert-base-cased-v1.2 on the jnlpba dataset. It achieves the following results on the evaluation set:
- Loss: 0.1265
- Precision: 0.8948
- Recall: 0.9282
- F1: 0.9112
- Accuracy: 0.9602
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.2278 | 1.0 | 1858 | 0.1826 | 0.8415 | 0.8815 | 0.8610 | 0.9384 |
| 0.151 | 2.0 | 3716 | 0.1443 | 0.8756 | 0.9162 | 0.8955 | 0.9530 |
| 0.1157 | 3.0 | 5574 | 0.1265 | 0.8948 | 0.9282 | 0.9112 | 0.9602 |
Framework versions
- Transformers 4.12.0.dev0
- Pytorch 1.9.1+cu102
- Datasets 1.12.1
- Tokenizers 0.10.3