MinDutch03 / file_based_chatbot

Gemini File Chatbot

(https://geminip-chatbot.streamlit.app/) Gemini File Chatbot is a Streamlit-based application that allows users to chat with a conversational AI model trained on documents. The chatbot extracts information from uploaded files and answers user questions based on the provided context.

Features

• PDF, Text, Image Upload: Users can upload multiple files.
• Text Extraction: Extracts text from uploaded files.
• Conversational AI: Uses the Gemini conversational AI model to answer user questions.
• Chat Interface: Provides a chat interface to interact with the chatbot.

Getting Started

If you have docker installed, you can run the application using the following command:

• Obtain a Google API key and set it in the .env file.

  GOOGLE_API_KEY=your_api_key_here

docker-compose up --build

Your application will be available at http://localhost:8501.

Deploying your application to the cloud

First, build your image, e.g.: docker build -t myapp .. If your cloud uses a different CPU architecture than your development machine (e.g., you are on a Mac M1 and your cloud provider is amd64), you'll want to build the image for that platform, e.g.: docker build --platform=linux/amd64 -t myapp ..

Then, push it to your registry, e.g. docker push myregistry.com/myapp.

Consult Docker's getting started docs for more detail on building and pushing.

References

• Docker's Python guide

Local Development

Follow these instructions to set up and run this project on your local machine.

Note: This project requires Python 3.10 or higher.

1. Clone the Repository:

   git clone https://github.com/MinDutch03/file_based_chatbot.git

2. Install Dependencies:

   pip install -r requirements.txt
   pip install pytesseract pillow

3. Set up Tesseract: On Windows:

   • Download Tesseract here: https://github.com/UB-Mannheim/tesseract/wiki
   • Add Path to the system environment: /path/to/Tesseract-OCR

4. Set up Google API Key:

   • Obtain a Google API key and set it in the .env file.

   GOOGLE_API_KEY=your_api_key_here

5. Run the Application:

   streamlit run main.py

6. Upload Files:

   • Use the sidebar to upload files.
   • Click on "Submit & Process" to extract text and generate embeddings.

7. Chat Interface:

   • Chat with the AI in the main interface.

Project Structure

• app.py: Main application script.
• .env: file which will contain your environment variable.
• requirements.txt: Python packages required for working of the app.
• README.md: Project documentation.

Dependencies

• PyPDF2
• langchain
• Streamlit
• google.generativeai
• dotenv
• tesseract

Acknowledgments

• Google Gemini: For providing the underlying language model.
• Streamlit: For the user interface framework.