|
--- |
|
license: cc-by-sa-4.0 |
|
base_model: klue/bert-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- klue |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: klue_ner_bert_model |
|
results: |
|
- task: |
|
name: Token Classification |
|
type: token-classification |
|
dataset: |
|
name: klue |
|
type: klue |
|
config: ner |
|
split: validation |
|
args: ner |
|
metrics: |
|
- name: Precision |
|
type: precision |
|
value: 0.883861132284665 |
|
- name: Recall |
|
type: recall |
|
value: 0.8966608084358524 |
|
- name: F1 |
|
type: f1 |
|
value: 0.890214963707426 |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.9781297871646948 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# klue_ner_bert_model |
|
|
|
This model is a fine-tuned version of [klue/bert-base](https://huggingface.co/klue/bert-base) on the klue dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0843 |
|
- Precision: 0.8839 |
|
- Recall: 0.8967 |
|
- F1: 0.8902 |
|
- Accuracy: 0.9781 |
|
|
|
## Model description |
|
|
|
KLUE BERT base is a pre-trained BERT Model on Korean Language. The developers of KLUE BERT base developed the model in the context of the development of the [Korean Language Understanding Evaluation (KLUE) Benchmark](https://arxiv.org/pdf/2105.09680.pdf). |
|
|
|
## Intended uses & limitations |
|
|
|
## How to Get Started With the Model |
|
|
|
```python |
|
from transformers import AutoModel, AutoTokenizer |
|
|
|
model = AutoModel.from_pretrained("klue/bert-base") |
|
tokenizer = AutoTokenizer.from_pretrained("klue/bert-base") |
|
``` |
|
|
|
## 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: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.0638 | 1.0 | 2626 | 0.0807 | 0.8623 | 0.8702 | 0.8662 | 0.9747 | |
|
| 0.0402 | 2.0 | 5252 | 0.0780 | 0.8756 | 0.8896 | 0.8825 | 0.9770 | |
|
| 0.025 | 3.0 | 7878 | 0.0843 | 0.8839 | 0.8967 | 0.8902 | 0.9781 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.31.0 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.0 |
|
- Tokenizers 0.13.3 |
|
|