README.md

metadata

library_name: transformers
license: apache-2.0
base_model: monologg/koelectra-base-v3-discriminator
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: MyMbti_classification_model
    results: []

MyMbti_classification_model

This model is a fine-tuned version of monologg/koelectra-base-v3-discriminator on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.5286
• Accuracy: 0.1898
• F1: 0.1547

Model description

이 모델은 16개의 MBTI를 라벨로 분류해 해당 라벨을 예측하는 모델입니다. 모델의 정확도가 낮은것은 학습에 사용한 데이터가 정제되지 않았습니다. 테스트용으로 만들었기 때문에 성능은 보장하지 못합니다.

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
2.6213	0.1673	500	2.6180	0.1142	0.0241
2.6412	0.3347	1000	2.6167	0.1318	0.0336
2.5861	0.5020	1500	2.6111	0.1320	0.0385
2.6183	0.6693	2000	2.6133	0.1222	0.0461
2.5954	0.8367	2500	2.5958	0.1411	0.0607
2.5828	1.0040	3000	2.5822	0.1479	0.0703
2.5803	1.1714	3500	2.5685	0.1553	0.0826
2.5615	1.3387	4000	2.5566	0.1645	0.0977
2.5463	1.5060	4500	2.5531	0.1687	0.1111
2.5511	1.6734	5000	2.5446	0.1679	0.1170
2.5242	1.8407	5500	2.5342	0.1726	0.1215
2.5191	2.0080	6000	2.5246	0.1825	0.1384
2.4866	2.1754	6500	2.5306	0.1834	0.1428
2.5005	2.3427	7000	2.5325	0.1803	0.1399
2.5131	2.5100	7500	2.5195	0.1877	0.1473
2.4918	2.6774	8000	2.5204	0.1876	0.1489
2.4755	2.8447	8500	2.5218	0.1877	0.1568
2.4223	3.0120	9000	2.5286	0.1898	0.1547
2.4297	3.1794	9500	2.5364	0.1874	0.1599
2.4213	3.3467	10000	2.5432	0.1866	0.1584
2.4619	3.5141	10500	2.5393	0.1879	0.1585
2.4383	3.6814	11000	2.5424	0.1849	0.1590
2.4368	3.8487	11500	2.5414	0.1866	0.1599

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4