HızlıBot🐇

This application was developed as part of the Fintern Future Talent Hackathon. It's a RAG-LLM-GenAI based chatbot designed to answer questions on customers' minds. You send your question and the chatbot generates the most appropriate answer from vectordb or analyzes data from PandasAI.

Technologies Used

• LangChain - LangChain version: 0.2.5
• PandasAI - PandasAI version: 2.1.1
• Streamlit - Streamlit version: 1.36.0

Architecture

---

Requirements

Environment

Ensure that your Python version is set to 3.10.12 (pip version is 24.0):

python --version

• Setting up Virtualenv:

pip install virtualenv

• Creating a Virtual Environment:

virtualenv venv

• Activating the Virtual Environment:

source venv/bin/activate

• Installing the necessary libraries:

pip install -r requirements.txt

Configuration

• Set up your .env file:
• Set up your .streamlit/secrets.toml file:
• Set up your app.py file:

cd <project-directory>

- Create the .env file and add your OPENAI_API_KEY:

    OPENAI_API_KEY='key' # .env file
    openai_api_key='Your OpenAI API Key' # secrets.toml
    OPENAI_API_KEY = 'OPENAI_API_KEY' # app.py 

Create VectorDB

python3 create_database.py

Run

• Launch the Streamlit app in terminal:

streamlit run app.py

---

HizliBotDemo.mp4