Check out the live project here!
Welcome to the Harry Potter Query Bot! This project is a LLM-powered RAG application designed to answer any query related to the Harry Potter series.
 |
The Harry Potter In-Depth Query Bot is a powerful application that combines Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to deliver accurate answers to queries about the Harry Potter series. Utilizing LangChain for seamless integration, the bot employs a Chroma vector database built from the books' text and is accessible via a user-friendly Flask interface. The entire project is containerized with Docker and deployed on AWS Lambda and API Gateway for scalable, serverless execution.
- LLM (GPT-4o-mini model): The core model that understands and generates responses to user queries, specifically tuned for the Harry Potter series.
- RAG (Retrieval-Augmented Generation): Enhances the LLM's capabilities by retrieving relevant information from a vector database, ensuring that the answers are both accurate and contextually aligned with the source material.
- LangChain: Serves as the framework that integrates the LLM, RAG mechanism, and vector database, coordinating their interaction to provide seamless query responses.
- Vector Database (Chroma with ChatGPT embeddings): Stores embeddings from all seven Harry Potter books, enabling efficient retrieval of the most relevant text to enrich the LLM’s responses.
- Dockerized: Containerizes the entire application, making it easy to deploy across various environments with consistent behavior.
- AWS Deployment: The application is deployed on AWS Lambda and API Gateway, providing a scalable, serverless architecture.
├── LICENSE
├── README.md <- The top-level README for developers using this project.
├── data
│ ├── processed <- Contains the document.pkl file which holds the chunks of text.
│ ├── vector_db <- Stores the vector database for quick retrieval.
│ └── raw <- The original seven Harry Potter books as text files.
├── requirements.txt <- The requirements file for reproducing the environment.
├── src <- Source code used in this project.
│ ├── __init__.py <- Makes src a Python module.
│ ├── base.py <- Contains utility functions.
│ ├── data_preprocessing.py <- Code to create the vector database using Chroma and OpenAIEmbeddings.
│ └── data_ingestion.py <- Handles the loading, chunking, and saving of text data.
├── templates
│ └── index.html <- HTML template for the web app.
├── .gitignore <- Specifies which files to ignore in the version control.
├── .dvc <- DVC configuration file for data version control.
├── app.py <- Main Flask application file.
├── dvc.lock <- DVC lock file ensuring reproducibility.
├── Dockerfile <- Dockerfile for containerizing the app.
├── dvc.yaml <- DVC pipeline configuration.
├── template.yaml <- AWS Lambda and API Gateway configuration template.
To get a copy of the project up and running on your local machine, follow these steps:
-
Clone the repository:
git clone https://github.com/TarunSingh2002/Harry-Potter-In-Depth-Query-Bot.git cd Harry-Potter-In-Depth-Query-Bot -
Set up a virtual environment:
conda create --name harry-potter-bot python=3.12 conda activate harry-potter-bot
-
Install the required dependencies:
pip install -r requirements.txt
-
Build and run the Docker container:
docker build -t harry-potter-bot . docker run -p 5000:5000 harry-potter-bot
-
Run the Flask application:
python app.py
-
Open your browser and go to:
http://localhost:5000 -
Interact with the bot by asking your Harry Potter-related questions.
-
Ensure you have AWS CLI installed and configured:
aws configure
-
Build and push the Docker image to Amazon ECR:
aws ecr get-login-password --region us-east-1 | docker login --username AWS --password-stdin <your-account-id>.dkr.ecr.us-east-1.amazonaws.com docker tag harry-potter-bot:latest <your-account-id>.dkr.ecr.us-east-1.amazonaws.com/harry-potter-bot:latest docker push <your-account-id>.dkr.ecr.us-east-1.amazonaws.com/harry-potter-bot:latest
-
Deploy the application using AWS SAM:
sam build --template-file template.yaml sam deploy --guided
-
Follow the prompts to complete the deployment.
MIT License
Copyright (c) 2024 Tarun Singh
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.This project would not have been possible without the works of J.K. Rowling and her creation of the Harry Potter series. The text from the seven books serves as the foundation for the vector database used in this application. Special thanks to the entire Harry Potter fandom for keeping the magic alive and inspiring projects like this one.