README.md

---
title: Content Classifier
emoji: 🔍
colorFrom: blue
colorTo: purple
sdk: docker
pinned: false
license: mit
app_port: 7860
---

# 🔍 Content Classifier

A powerful FastAPI-based content classification service using ONNX for threat detection and sentiment analysis.

## 🚀 Features

- **Threat Detection**: Classify content for potential threats
- **Sentiment Analysis**: Analyze text sentiment (positive/negative)
- **ONNX Runtime**: High-performance model inference
- **REST API**: Easy-to-use HTTP endpoints
- **Auto Documentation**: Interactive Swagger UI at `/docs`
- **Health Monitoring**: Built-in health checks

## 📡 API Endpoints

| Endpoint | Method | Description |
|----------|--------|-------------|
| `/predict` | POST | Classify text content |
| `/health` | GET | Check API health status |
| `/model-info` | GET | Get model information |
| `/docs` | GET | Interactive API documentation |

## 🔧 Usage

### Example Request

```bash
curl -X POST "https://YOUR-SPACE-NAME.hf.space/predict" \
     -H "Content-Type: application/json" \
     -d '{"text": "This is a sample text to classify"}'
```

### Example Response

```json
{
    "is_threat": false,
    "final_confidence": 0.75,
    "threat_prediction": 0.25,
    "sentiment_analysis": {
        "label": "POSITIVE",
        "score": 0.5
    },
    "onnx_prediction": {
        "threat_probability": 0.25,
        "raw_output": [[0.75, 0.25]],
        "output_shape": [1, 2]
    },
    "models_used": ["contextClassifier.onnx"],
    "raw_predictions": {
        "onnx": {
            "threat_probability": 0.25,
            "raw_output": [[0.75, 0.25]]
        },
        "sentiment": {
            "label": "POSITIVE",
            "score": 0.5
        }
    }
}
```

## 📊 Response Format

The API returns a structured response with:

- **`is_threat`**: Boolean indicating if content is threatening
- **`final_confidence`**: Confidence score (0.0 to 1.0)
- **`threat_prediction`**: Raw threat probability
- **`sentiment_analysis`**: Sentiment classification and score
- **`onnx_prediction`**: Raw ONNX model output
- **`models_used`**: List of models used for prediction
- **`raw_predictions`**: Complete prediction data

## 🛠️ Local Development

1. **Install dependencies:**
   ```bash
   pip install -r requirements.txt
   ```

2. **Place your model:**
   Ensure `contextClassifier.onnx` is in the project root

3. **Run the API:**
   ```bash
   python app.py
   ```

4. **Visit:** `http://localhost:7860/docs`

## 🐳 Docker

```bash
# Build
docker build -t content-classifier .

# Run
docker run -p 7860:7860 content-classifier
```

## 📝 Model Requirements

Your `contextClassifier.onnx` model should:
- Accept text-based inputs
- Return classification predictions
- Be compatible with ONNX Runtime

## ⚙️ Configuration

Customize the preprocessing and postprocessing functions in `app.py` based on your specific model requirements.

## 🔍 Monitoring

- **Health Check**: `/health` - Monitor API status
- **Model Info**: `/model-info` - View model details
- **Logs**: Check application logs for debugging

## 📜 License

MIT License - feel free to use and modify!