AskPDF

PDF Question Answering Application

A Streamlit application that allows users to upload PDF documents and ask questions about their content using Google's Gemini AI and Vertex AI Vector Search.

Features

• 📄 PDF document upload and processing
• 🔍 Semantic search using Vertex AI Vector Search
• 💬 Question answering using Google's Gemini AI
• 🎯 Accurate answers based on document content
• 🔒 Secure handling of API keys and credentials

Prerequisites

• Python 3.8 or higher
• Google Cloud Platform account
• Gemini API key
• Vertex AI enabled in your GCP project

Installation

1. Clone the repository:

git clone https://github.com/yourusername/askPdf.git
cd askPdf

2. Create and activate a virtual environment:

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

3. Install dependencies:

pip install -r requirements.txt

4. Set up your credentials:
   • Create a .streamlit/secrets.toml file with the following structure:

   # Google Cloud Project Configuration
   GCP_PROJECT_ID = "your-project-id"
   GCP_LOCATION = "us-central1"
   VERTEX_AI_INDEX_ID = "your-index-id"
   VERTEX_AI_ENDPOINT_ID = "your-endpoint-id"
   GEMINI_API_KEY = "your-gemini-api-key"
   
   # Service Account Credentials
   [gcp_service_account]
   type = "service_account"
   project_id = "your-project-id"
   private_key_id = "your-private-key-id"
   private_key = """your-private-key"""
   client_email = "your-service-account-email"
   client_id = "your-client-id"
   auth_uri = "https://accounts.google.com/o/oauth2/auth"
   token_uri = "https://oauth2.googleapis.com/token"
   auth_provider_x509_cert_url = "https://www.googleapis.com/oauth2/v1/certs"
   client_x509_cert_url = "your-client-cert-url"
   universe_domain = "googleapis.com"
   
   # Vertex AI Index Configuration
   [vertex_ai_index]
   dimensions = 384
   algorithm_config = { "bruteForceConfig" = {} }
   distance_measure_type = "DOT_PRODUCT_DISTANCE"
   
   # Vertex AI Streaming Index Configuration
   [vertex_ai_stream_index]
   dimensions = 384
   algorithm_config = { "bruteForceConfig" = {} }
   distance_measure_type = "DOT_PRODUCT_DISTANCE"
   shard_size = "SHARD_SIZE_MEDIUM"

Usage

1. Start the Streamlit app:

streamlit run src/app.py

2. Open your browser and navigate to http://localhost:8501

3. Upload a PDF document

4. Ask questions about the document content

Project Structure

askPdf/
├── src/
│   └── app.py              # Main application code
├── .streamlit/
│   └── secrets.toml        # API keys and configuration
├── requirements.txt        # Python dependencies
└── README.md              # Project documentation

Configuration

The application uses Streamlit secrets for configuration. All sensitive information and configuration settings are stored in .streamlit/secrets.toml, including:

• Google Cloud Project settings
• Service Account credentials
• Vertex AI index configurations
• API keys

Dependencies

• streamlit: Web application framework
• PyMuPDF (fitz): PDF processing
• sentence-transformers: Text embedding generation
• google-generativeai: Gemini API integration
• google-cloud-aiplatform: Vertex AI integration

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

• Run tests:

  python -m unittest tests/test_pdf_processing.py

• Google Gemini AI for the language model
• Vertex AI for vector search capabilities
• Streamlit for the web interface framework