|
|
--- |
|
|
title: Emotion Detection App |
|
|
emoji: 😊 |
|
|
colorFrom: indigo |
|
|
colorTo: pink |
|
|
sdk: gradio |
|
|
sdk_version: "4.27.0" |
|
|
app_file: app.py |
|
|
pinned: false |
|
|
--- |
|
|
|
|
|
# 😊 Emotion Detection from Text using BERT |
|
|
|
|
|
Welcome to the **Emotion Detection Web App**! This application uses a fine-tuned BERT model to detect human emotions from short pieces of text. |
|
|
|
|
|
--- |
|
|
|
|
|
## 🚀 Demo |
|
|
|
|
|
👉 Try the live app: [Click here to open the web app](https://huggingface.co/spaces/sujith13082003/emotion_detection) |
|
|
|
|
|
--- |
|
|
|
|
|
## 🔍 Description |
|
|
|
|
|
This project leverages the `nateraw/bert-base-uncased-emotion` model from Hugging Face Transformers to classify input text into one of six emotions: |
|
|
|
|
|
- 😢 Sadness |
|
|
- 😀 Joy |
|
|
- 💖 Love |
|
|
- 😡 Anger |
|
|
- 😱 Fear |
|
|
- 😲 Surprise |
|
|
|
|
|
It uses: |
|
|
- **Hugging Face Transformers** for model and tokenizer |
|
|
- **PyTorch** for deep learning inference |
|
|
- **Gradio** to build an interactive web interface |
|
|
|
|
|
--- |
|
|
|
|
|
## 🧠 Model Used |
|
|
|
|
|
- **Model Name**: `nateraw/bert-base-uncased-emotion` |
|
|
- **Base Architecture**: BERT (uncased) |
|
|
- **Dataset**: GoEmotions subset |
|
|
|
|
|
--- |
|
|
|
|
|
## 📦 Dependencies |
|
|
|
|
|
Dependencies are defined in `requirements.txt`: |
|
|
- `transformers` |
|
|
- `torch` |
|
|
- `gradio` |
|
|
|
|
|
--- |
|
|
|
|
|
## 📈 Use Cases |
|
|
|
|
|
- Social media sentiment analysis |
|
|
- Customer feedback classification |
|
|
- Chatbot emotion understanding |
|
|
- Mental health applications |
|
|
|
|
|
--- |
|
|
|
|
|
## 👨💻 Author |
|
|
|
|
|
- **Sujith Kumar** |
|
|
- Hugging Face: [@sujith13082003](https://huggingface.co/sujith13082003) |
|
|
|
|
|
--- |
|
|
|
|
|
## 📝 License |
|
|
|
|
|
This project is for educational and research purposes. Refer to individual library licenses for commercial use. |
|
|
|
