YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Japanese Sentiment Analysis v1
Model Description
Fine-tuned Japanese BERT model for 3-class sentiment analysis (Positive / Neutral / Negative) on e-commerce reviews and customer feedback.
- Base model: cl-tohoku/bert-base-japanese-v2
- Task: Text Classification
- Language: Japanese
- License: MIT
- Developer: Retro099 (portfolio project)
How to Use
from transformers import pipeline
classifier = pipeline("sentiment-analysis",
model="Retro099/japanese-sentiment-analysis-v1",
device=-1) # CPU for local / Streamlit Cloud
result = classifier("この商品は最高です!とても満足しています。")
print(result)
Label Mapping
LABEL_0 → Negative 😞
LABEL_1 → Neutral 😐
LABEL_2 → Positive 😊
日本語要約
日本語のレビューやテキストに対する感情分析モデルを構築しました。Hugging Face Transformers + PyTorchを使用してBERTをファインチューニングし、Positive/Negative/Neutralの3クラス分類を実現。eコマースにおける顧客満足度分析や自動レビュー分類に活用可能です。GradioデモとStreamlit Cloudデプロイ済みです。
Live Demos
Streamlit Cloud: https://cx7v54ehccpwnarlaplxt.streamlit.app/
Gradio (temporary): https://f50c787d7b105f7bf9.gradio.live/
GitHub Repo: https://github.com/Retro099/ML-Projects/tree/main/Japanese_Sentiment_Analysis
- Downloads last month
- 73
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support