Mindcast Emotion Classifier (Normal)

Model Description

한국어 일반 댓글에서 6가지 감정을 분류하는 모델입니다.

이 모델은 LoRA (Low-Rank Adaptation)를 사용하여 효율적으로 파인튜닝되었으며, 최종적으로 base model과 merge되어 배포되었습니다.

Training Date: 2025-12-15

Performance

Test Set Results

Metric Score
Accuracy 0.2383
F1 Score (Macro) 0.1915
F1 Score (Weighted) 0.2708

Confusion Matrix

[[33 34 24 30 34 21]
 [ 2  5  1  3 10  2]
 [ 1  3  2  3  3  1]
 [ 1  0  0  1  1  0]
 [ 7  8  3  2 17 12]
 [ 3  6  3  2  7 13]]

Detailed Classification Report

              precision    recall  f1-score   support

          분노     0.7021    0.1875    0.2960       176
          슬픔     0.0893    0.2174    0.1266        23
          불안     0.0606    0.1538    0.0870        13
          상처     0.0244    0.3333    0.0455         3
          당황     0.2361    0.3469    0.2810        49
          기쁨     0.2653    0.3824    0.3133        34

   micro avg     0.2383    0.2383    0.2383       298
   macro avg     0.2296    0.2702    0.1915       298
weighted avg     0.4936    0.2383    0.2708       298

Training Details

Hyperparameters

Hyperparameter Value
Base Model klue/roberta-base
Batch Size 64
Epochs 60
Learning Rate 0.0001
Warmup Ratio 0.1
Weight Decay 0.01
LoRA r 8
LoRA alpha 16
LoRA dropout 0.05

Training Data

  • Train samples: 973 (from 1 files)
  • Valid samples: 109
  • Test samples: 298 (from 1 files)
  • Number of labels: 6
  • Labels: 분노, 슬픔, 불안, 상처, 당황, 기쁨

Usage

Installation

pip install transformers torch

Quick Start

from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline

# Load model
model_name = "merrybabyxmas/mindcast-emotion-normal"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Create pipeline
classifier = pipeline("text-classification", model=model, tokenizer=tokenizer)

# Predict
text = "오늘 정말 행복한 하루였어요"
result = classifier(text)
print(result)

Model Architecture

  • Base Model: klue/roberta-base
  • Task: Sequence Classification
  • Number of Labels: N/A

Citation

If you use this model, please cite:

@misc{mindcast-model,
  author = {Mindcast Team},
  title = {Mindcast Emotion Classifier (Normal)},
  year = {2025},
  publisher = {HuggingFace},
  howpublished = {\url{https://huggingface.co/merrybabyxmas/mindcast-emotion-sc-only}},
}

Contact

For questions or feedback, please open an issue on the model repository.


This model card was automatically generated.

Downloads last month
-
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support