This project implements a Retrieval-Augmented Generation (RAG) pipeline to answer queries based on user-provided text files. Users can add documents, set up the Google Gemini API, and run the application to ask questions based on the content of the added files.
Add your own .txt/.pdf files to the data/folder to create a custom knowledge base. Uses the Google Gemini API for query resolution. Answers queries based on the content of your text files. Easy-to-run Python application.
Follow these three simple steps to start querying your documents:
# 1. Clone the repository
git clone <repo-url>
cd <repo-directory>
# 2. Make a virtual environment and install all the required dependencies
pip intall -r requirements.txt
# 3. Add your .txt/.pdf files to the 'data/' folder
# 4. Set your Google Gemini API key in .env file as shown in .env_example
export GOOGLE_GEMINI_API_KEY="your_api_key_here" # Linux/macOS
setx GOOGLE_GEMINI_API_KEY "your_api_key_here" # Windows PowerShell
# 5. Run the app
python src/app.py