bert-base-finetuned-ynat

This model is a fine-tuned version of klue/bert-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3745
• F1: 0.8704

Model description

뉴스 제목을 입력하면 뉴스의 카테고리를 예측
label_map = {
'LABEL_0': 'IT/과학',
'LABEL_1': '경제',
'LABEL_2': '사회',
'LABEL_3': '생활문화',
'LABEL_4': '세계',
'LABEL_5': '스포츠',
'LABEL_6': '정치'
}

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: 256
• eval_batch_size: 256
• 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	F1
No log	1.0	179	0.3909	0.8655
No log	2.0	358	0.3788	0.8684
0.3774	3.0	537	0.3629	0.8699
0.3774	4.0	716	0.3776	0.8667
0.3774	5.0	895	0.3745	0.8704

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0