https://github.com/rohanchanana2/context-aware-religious-chatbot

A Streamlit chatbot using RAG for context-aware responses from the Bhagavad Gita and Patanjali Yog Sutras.
https://github.com/rohanchanana2/context-aware-religious-chatbot

faiss langchain python rag streamlit

Last synced: 19 days ago
JSON representation

A Streamlit chatbot using RAG for context-aware responses from the Bhagavad Gita and Patanjali Yog Sutras.

• Host: GitHub
• URL: https://github.com/rohanchanana2/context-aware-religious-chatbot
• Owner: rohanchanana2
• Created: 2025-01-10T03:47:06.000Z (about 1 month ago)
• Default Branch: main
• Last Pushed: 2025-01-18T11:25:16.000Z (25 days ago)
• Last Synced: 2025-01-18T12:27:42.029Z (25 days ago)
• Topics: faiss, langchain, python, rag, streamlit
• Language: Python
• Homepage:
• Size: 14.6 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# 📖 Context Aware Religious Chatbot

A Streamlit-based application that answers queries using the sacred texts of the **Bhagavad Gita** and **Patanjali Yog Sutras**. This chatbot combines advanced AI with hybrid retrieval techniques to deliver precise answers and relevant verses.

## 🌟 Features

- **AI-Powered Answers**: Generates contextual responses to your queries.

- **Verse Retrieval**: Displays related verses with translations.

- **Hybrid Search**: Combines dense (vector) and sparse (keyword) retrieval for accuracy.

- **Sub-Query Processing**: Decomposes queries for improved understanding.

- **Download Results**: Export answers and verses as a JSON file.

## 📂 Project Structure

- **`app.py`**: Main application script.

- **`indexing_retriever.py`**: Handles document splitting and retrieval mechanisms.

- **`decomposition.py`**: Decomposes user queries into sub-queries for better comprehension.

- **`dataset_preprocessing.py`**: Prepares the dataset for processing and optimization.

- **`retriever.py`**: Fetches query-related context and sub-query results.

- **`generator.py`**: Generates answers based on the retrieved context.

- **`final_generation.py`**: Combines answers into a final, cohesive response.

- **`config.py`**: Configures the AI models and embeddings (API key setup instructions are in this file).

## 📥 Dataset

Download the datasets:

- Bhagavad Gita dataset: [Bhagwad Gita Verses](https://github.com/atmabodha/Vedanta_Datasets/blob/main/Bhagwad_Gita/Bhagwad_Gita_Verses_English_Questions.csv)

- Patanjali Yog Sutras dataset: [Patanjali Yoga Sutras Verses](https://github.com/atmabodha/Vedanta_Datasets/blob/main/Patanjali_Yoga_Sutras/Patanjali_Yoga_Sutras_Verses_English_Questions.csv)