Vertex AI FastAPI Classifier

This project is a FastAPI-based web service that uses Google Vertex AI for text classification. It provides endpoints to classify questions, store responses, and collect user feedback. The backend uses SQLAlchemy for database operations.

Features

• /api/classify: Classifies a question using a Vertex AI model.
• /api/feedback: Allows users to submit feedback on the classification.
• /api/feedback/{response_id}: Retrieves feedback for a specific response.
• Stores all questions, responses, and feedback in a SQL database.

Project Structure

vertex-ai-fastAPI/
├── classify.py                 # Vertex AI classification logic
├── main.py                     # FastAPI app and API endpoints
├── db.py                       # Database models and session setup
├── test.py                     # API endpoint tests
├── requirements.txt            # Python dependencies
├── service-account-key.json    # (GCP credentials — should be in .gitignore)

Setup

1. Clone the repository

   git clone https://github.com/<your-username>/<repo-name>.git
   cd vertex-ai-fastAPI
   
   

2. Create and activate a virtual environment

    python -m venv venv
    venv\Scripts\activate   # On Windows
    # source venv/bin/activate   # On Linux/Mac

3. Install dependencies

   pip install -r requirement.txt
   

4. Configure Google Cloud credentials

• Place your service-account-key.json in the project root.
• Important: Do NOT commit this file to GitHub.

5. Set up your database

• Update DATABASE_URL in db.py with your actual database credentials.

6. Run the FastAPI server

   uvicorn main:app --reload
   

7. Test the API -Use the included test.py or tools like Postman or curl

   # Test
   /api/classify endpoint curl -X POST http://127.0.0.1:8000/api/classify \ -H "Content-Type: application/json" \ -d '{"question": "Wifi issues at mc nair?"}'
   # Test
   /api/feedback endpoint curl -X POST http://127.0.0.1:8000/api/feedback \ -H "Content-Type: application/json" \ -d '{"response_id": 1, "feedback": "Accurate classification"}' 
   # Test
   /api/feedback/{response_id} endpoint curl http://127.0.0.1:8000/api/feedback/1