metadata

library_name: transformers
language:
  - ko
license: apache-2.0
base_model: monologg/koelectra-base-v3-discriminator
tags:
  - text-classification
  - koELECTRA
  - Korean-NLP
  - topic-classification
  - news-classification
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: ynat-model
    results: []

ynat-model

This model is a fine-tuned version of monologg/koelectra-base-v3-discriminator on the klue-ynat dataset. It achieves the following results on the evaluation set:

Loss: 0.4131
Accuracy: 0.8601
F1: 0.8614
Precision: 0.8477
Recall: 0.8773

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: 5e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.3952	1.0	714	0.4250	0.8523	0.8525	0.8336	0.8755
0.2963	2.0	1428	0.3992	0.8574	0.8583	0.8454	0.8746
0.2176	3.0	2142	0.4131	0.8601	0.8614	0.8477	0.8773

Framework versions

Transformers 4.54.1
Pytorch 2.6.0+cu124
Datasets 4.0.0
Tokenizers 0.21.4