Skip to content

[submission] REVANTH -RAG app

AUVSULA REVANTH CHARY requested to merge revanthchary/ip1-icfai:main into main

🤖 HF RAG Chatbot

A Retrieval-Augmented Generation (RAG) Chatbot built using Streamlit, Hugging Face Datasets, Sentence Transformers, FAISS, and Transformers.

🚀 Live Demo

🔗 Application URL: https://mwve7uggbf7mqdfwete745.streamlit.app/

📌 Project Overview

This project implements a Retrieval-Augmented Generation (RAG) pipeline that combines semantic search and transformer-based text generation.

The chatbot:

  • Loads datasets from Hugging Face
  • Preprocesses and structures textual data
  • Creates semantic embeddings using Sentence Transformers
  • Stores embeddings using FAISS vector indexing
  • Retrieves relevant context based on user queries
  • Generates context-aware responses using FLAN-T5
  • Provides an interactive Streamlit interface

🛠️ Technologies Used

  • Python
  • Streamlit
  • Hugging Face Datasets
  • Sentence Transformers
  • FAISS
  • Transformers (FLAN-T5)
  • NumPy

⚙️ Architecture

User Query
    ↓
Semantic Retrieval (FAISS)
    ↓
Relevant Context
    ↓
FLAN-T5 Generation
    ↓
Final Response

✨ Features

  • Interactive chatbot interface
  • Retrieval-Augmented Generation (RAG)
  • Semantic similarity search
  • Hugging Face dataset integration
  • Transformer-based response generation
  • Context-aware answers
  • Fast vector search using FAISS
  • Streamlit web deployment

👥 Team Contributions

1. Revanth Chary

Role: Frontend & Streamlit Interface

Contributions

  • Designed and structured the Streamlit UI
  • Created sidebar functionality
  • Implemented chatbot layout
  • Managed chat history display
  • Implemented session state handling
  • Added clear chat functionality

Prompt Used

Build a clean Streamlit interface for a RAG chatbot with sidebar navigation, chat history display, chat input box, responsive layout, session state handling, and clear chat functionality.

2. Abhinav Reddy

Role: Dataset Integration & Preprocessing

Contributions

  • Developed Hugging Face dataset loading pipeline
  • Integrated multiple datasets
  • Implemented text extraction logic
  • Performed data cleaning and preprocessing
  • Managed dataset handling and warnings

Prompt Used

Create a Python module that loads datasets from Hugging Face, extracts meaningful text fields, handles multiple dataset formats, cleans and preprocesses text, and returns structured text data for RAG systems.

3. Manasvi

Role: Vector Database & Retrieval System

Contributions

  • Implemented text chunking
  • Generated embeddings using Sentence Transformers
  • Built FAISS vector indexing
  • Developed semantic similarity retrieval
  • Optimized top-k search functionality

Prompt Used

Implement a vector database pipeline using Sentence Transformers embeddings, FAISS indexing, text chunking with overlap, semantic similarity retrieval, and efficient top-k search functionality.

4. Meneni Srikar Rao

Role: Response Generation Pipeline

Contributions

  • Integrated Hugging Face Transformers
  • Implemented FLAN-T5 generation
  • Built context-aware prompting
  • Developed retrieval-based answer generation
  • Managed response formatting

Prompt Used

Build a response generation pipeline using Hugging Face Transformers, FLAN-T5 model, context-aware prompting, retrieval-augmented generation, and clean answer formatting.

5. Badrinath Goud

Role: Testing & Deployment

Contributions

  • Configured Python environment
  • Installed dependencies
  • Performed debugging and testing
  • Validated application execution
  • Prepared deployment setup

Prompt Used

Setup and test a Streamlit RAG application with Python virtual environment, dependency installation, Streamlit execution, debugging support, and deployment readiness checks.

📂 Installation

Clone Repository

git clone https://github.com/sAbhinavReddy/rag.git
cd rag

Install Dependencies

pip install -r requirements.txt

Run the Application

streamlit run app.py

🌐 Deployment

The application is deployed using Streamlit Community Cloud.

Live Application: https://mwve7uggbf7mqdfwete745.streamlit.app/

🎯 Project Outcome

The team successfully developed a Retrieval-Augmented Generation chatbot capable of:

  • Loading real-world datasets from Hugging Face
  • Creating semantic vector embeddings
  • Retrieving relevant contextual information
  • Generating AI-powered responses using transformer models
  • Delivering an intuitive Streamlit-based user experience

📜 License

This project was developed for educational and hackathon purposes.

👨‍💻 Team

  • Revanth Chary
  • Abhinav Reddy
  • Manasvi
  • Meneni Srikar Rao
  • Badrinath Goud# 🤖 HF RAG Chatbot

A Retrieval-Augmented Generation (RAG) Chatbot built using Streamlit, Hugging Face Datasets, Sentence Transformers, FAISS, and Transformers.

🚀 Live Demo

🔗 Application URL: https://mwve7uggbf7mqdfwete745.streamlit.app/

📌 Project Overview

This project implements a Retrieval-Augmented Generation (RAG) pipeline that combines semantic search and transformer-based text generation.

The chatbot:

  • Loads datasets from Hugging Face
  • Preprocesses and structures textual data
  • Creates semantic embeddings using Sentence Transformers
  • Stores embeddings using FAISS vector indexing
  • Retrieves relevant context based on user queries
  • Generates context-aware responses using FLAN-T5
  • Provides an interactive Streamlit interface

🛠️ Technologies Used

  • Python
  • Streamlit
  • Hugging Face Datasets
  • Sentence Transformers
  • FAISS
  • Transformers (FLAN-T5)
  • NumPy

⚙️ Architecture

User Query
    ↓
Semantic Retrieval (FAISS)
    ↓
Relevant Context
    ↓
FLAN-T5 Generation
    ↓
Final Response

✨ Features

  • Interactive chatbot interface
  • Retrieval-Augmented Generation (RAG)
  • Semantic similarity search
  • Hugging Face dataset integration
  • Transformer-based response generation
  • Context-aware answers
  • Fast vector search using FAISS
  • Streamlit web deployment

👥 Team Contributions

1. Revanth Chary

Role: Frontend & Streamlit Interface

Contributions

  • Designed and structured the Streamlit UI
  • Created sidebar functionality
  • Implemented chatbot layout
  • Managed chat history display
  • Implemented session state handling
  • Added clear chat functionality

Prompt Used

Build a clean Streamlit interface for a RAG chatbot with sidebar navigation, chat history display, chat input box, responsive layout, session state handling, and clear chat functionality.

2. Abhinav Reddy

Role: Dataset Integration & Preprocessing

Contributions

  • Developed Hugging Face dataset loading pipeline
  • Integrated multiple datasets
  • Implemented text extraction logic
  • Performed data cleaning and preprocessing
  • Managed dataset handling and warnings

Prompt Used

Create a Python module that loads datasets from Hugging Face, extracts meaningful text fields, handles multiple dataset formats, cleans and preprocesses text, and returns structured text data for RAG systems.

3. Manasvi

Role: Vector Database & Retrieval System

Contributions

  • Implemented text chunking
  • Generated embeddings using Sentence Transformers
  • Built FAISS vector indexing
  • Developed semantic similarity retrieval
  • Optimized top-k search functionality

Prompt Used

Implement a vector database pipeline using Sentence Transformers embeddings, FAISS indexing, text chunking with overlap, semantic similarity retrieval, and efficient top-k search functionality.

4. Meneni Srikar Rao

Role: Response Generation Pipeline

Contributions

  • Integrated Hugging Face Transformers
  • Implemented FLAN-T5 generation
  • Built context-aware prompting
  • Developed retrieval-based answer generation
  • Managed response formatting

Prompt Used

Build a response generation pipeline using Hugging Face Transformers, FLAN-T5 model, context-aware prompting, retrieval-augmented generation, and clean answer formatting.

5. Badrinath Goud

Role: Testing & Deployment

Contributions

  • Configured Python environment
  • Installed dependencies
  • Performed debugging and testing
  • Validated application execution
  • Prepared deployment setup

Prompt Used

Setup and test a Streamlit RAG application with Python virtual environment, dependency installation, Streamlit execution, debugging support, and deployment readiness checks.

📂 Installation

Clone Repository

git clone https://github.com/sAbhinavReddy/rag.git
cd rag

Install Dependencies

pip install -r requirements.txt

Run the Application

streamlit run app.py

🌐 Deployment

The application is deployed using Streamlit Community Cloud.

Live Application: https://mwve7uggbf7mqdfwete745.streamlit.app/

🎯 Project Outcome

The team successfully developed a Retrieval-Augmented Generation chatbot capable of:

  • Loading real-world datasets from Hugging Face
  • Creating semantic vector embeddings
  • Retrieving relevant contextual information
  • Generating AI-powered responses using transformer models
  • Delivering an intuitive Streamlit-based user experience

📜 License

This project was developed for educational and hackathon purposes.

👨‍💻 Team

  • Revanth Chary
  • Abhinav Reddy
  • Manasvi
  • Meneni Srikar Rao
  • Badrinath Goud# 🤖 HF RAG Chatbot

A Retrieval-Augmented Generation (RAG) Chatbot built using Streamlit, Hugging Face Datasets, Sentence Transformers, FAISS, and Transformers.

🚀 Live Demo

🔗 Application URL: https://mwve7uggbf7mqdfwete745.streamlit.app/

📌 Project Overview

This project implements a Retrieval-Augmented Generation (RAG) pipeline that combines semantic search and transformer-based text generation.

The chatbot:

  • Loads datasets from Hugging Face
  • Preprocesses and structures textual data
  • Creates semantic embeddings using Sentence Transformers
  • Stores embeddings using FAISS vector indexing
  • Retrieves relevant context based on user queries
  • Generates context-aware responses using FLAN-T5
  • Provides an interactive Streamlit interface

🛠️ Technologies Used

  • Python
  • Streamlit
  • Hugging Face Datasets
  • Sentence Transformers
  • FAISS
  • Transformers (FLAN-T5)
  • NumPy

⚙️ Architecture

User Query
    ↓
Semantic Retrieval (FAISS)
    ↓
Relevant Context
    ↓
FLAN-T5 Generation
    ↓
Final Response

✨ Features

  • Interactive chatbot interface
  • Retrieval-Augmented Generation (RAG)
  • Semantic similarity search
  • Hugging Face dataset integration
  • Transformer-based response generation
  • Context-aware answers
  • Fast vector search using FAISS
  • Streamlit web deployment

👥 Team Contributions

1. Revanth Chary

Role: Frontend & Streamlit Interface

Contributions

  • Designed and structured the Streamlit UI
  • Created sidebar functionality
  • Implemented chatbot layout
  • Managed chat history display
  • Implemented session state handling
  • Added clear chat functionality

Prompt Used

Build a clean Streamlit interface for a RAG chatbot with sidebar navigation, chat history display, chat input box, responsive layout, session state handling, and clear chat functionality.

2. Abhinav Reddy

Role: Dataset Integration & Preprocessing

Contributions

  • Developed Hugging Face dataset loading pipeline
  • Integrated multiple datasets
  • Implemented text extraction logic
  • Performed data cleaning and preprocessing
  • Managed dataset handling and warnings

Prompt Used

Create a Python module that loads datasets from Hugging Face, extracts meaningful text fields, handles multiple dataset formats, cleans and preprocesses text, and returns structured text data for RAG systems.

3. Manasvi

Role: Vector Database & Retrieval System

Contributions

  • Implemented text chunking
  • Generated embeddings using Sentence Transformers
  • Built FAISS vector indexing
  • Developed semantic similarity retrieval
  • Optimized top-k search functionality

Prompt Used

Implement a vector database pipeline using Sentence Transformers embeddings, FAISS indexing, text chunking with overlap, semantic similarity retrieval, and efficient top-k search functionality.

4. Meneni Srikar Rao

Role: Response Generation Pipeline

Contributions

  • Integrated Hugging Face Transformers
  • Implemented FLAN-T5 generation
  • Built context-aware prompting
  • Developed retrieval-based answer generation
  • Managed response formatting

Prompt Used

Build a response generation pipeline using Hugging Face Transformers, FLAN-T5 model, context-aware prompting, retrieval-augmented generation, and clean answer formatting.

5. Badrinath Goud

Role: Testing & Deployment

Contributions

  • Configured Python environment
  • Installed dependencies
  • Performed debugging and testing
  • Validated application execution
  • Prepared deployment setup

Prompt Used

Setup and test a Streamlit RAG application with Python virtual environment, dependency installation, Streamlit execution, debugging support, and deployment readiness checks.

📂 Installation

Clone Repository

git clone https://github.com/sAbhinavReddy/rag.git
cd rag

Install Dependencies

pip install -r requirements.txt

Run the Application

streamlit run app.py

🌐 Deployment

The application is deployed using Streamlit Community Cloud.

Live Application: https://mwve7uggbf7mqdfwete745.streamlit.app/

🎯 Project Outcome

The team successfully developed a Retrieval-Augmented Generation chatbot capable of:

  • Loading real-world datasets from Hugging Face
  • Creating semantic vector embeddings
  • Retrieving relevant contextual information
  • Generating AI-powered responses using transformer models
  • Delivering an intuitive Streamlit-based user experience

📜 License

This project was developed for educational and hackathon purposes.

👨‍💻 Team

  • Revanth Chary
  • Abhinav Reddy
  • Manasvi
  • Meneni Srikar Rao
  • Badrinath Goud
Edited by AUVSULA REVANTH CHARY

Merge request reports