Downloads · 30 days
5
10% of all-time downloads
merrybabyxmas/mindcast-emotion-sarcastic
mindcast-emotion-sarcastic is a text classification model from merrybabyxmas. Use it when you need a label for a piece of text. The card lists the license as apache-2.0.
한국어 비꼬는 댓글(sarcasm)에서 6가지 감정을 분류하는 모델입니다.
Downloads · 30 days
5
10% of all-time downloads
All-time downloads
52
Public
Parameters
111M
2.7 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors443 MB · 100%
From the Hugging Face model README
한국어 비꼬는 댓글(sarcasm)에서 6가지 감정을 분류하는 모델입니다.
이 모델은 LoRA (Low-Rank Adaptation)를 사용하여 효율적으로 파인튜닝되었으며, 최종적으로 base model과 merge되어 배포되었습니다.
Training Date: 2025-12-15
| Metric | Score |
|---|---|
| Accuracy | 0.3140 |
| F1 Score (Macro) | 0.2097 |
| F1 Score (Weighted) | 0.3518 |
[[41 21 15 3 26 33]
[ 5 4 1 0 6 3]
[ 4 1 2 0 1 0]
[ 0 0 0 0 0 0]
[ 6 7 3 0 20 16]
[ 3 4 2 0 6 9]]
precision recall f1-score support
분노 0.6949 0.2950 0.4141 139
슬픔 0.1081 0.2105 0.1429 19
불안 0.0870 0.2500 0.1290 8
상처 0.0000 0.0000 0.0000 0
당황 0.3390 0.3846 0.3604 52
기쁨 0.1475 0.3750 0.2118 24
micro avg 0.3140 0.3140 0.3140 242
macro avg 0.2294 0.2525 0.2097 242
weighted avg 0.4980 0.3140 0.3518 242
| Hyperparameter | Value |
|---|---|
| Base Model | klue/roberta-base |
| Batch Size | 64 |
| Epochs | 150 |
| Learning Rate | 0.0001 |
| Warmup Ratio | 0.1 |
| Weight Decay | 0.01 |
| LoRA r | 8 |
| LoRA alpha | 16 |
| LoRA dropout | 0.05 |
pip install transformers torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
# Load model
model_name = "merrybabyxmas/mindcast-emotion-sarcastic"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)
# Create pipeline
classifier = pipeline("text-classification", model=model, tokenizer=tokenizer)
# Predict (sarcastic text)
text = "아 진짜 최고다~ 오늘도 망했네"
result = classifier(text)
print(result)
If you use this model, please cite:
@misc{mindcast-model,
author = {Mindcast Team},
title = {Mindcast Emotion Classifier (Sarcastic)},
year = {2025},
publisher = {HuggingFace},
howpublished = {\url{https://huggingface.co/merrybabyxmas/mindcast-emotion-sc-only}},
}
For questions or feedback, please open an issue on the model repository.
This model card was automatically generated.