House Price Improvement Recommender

MSCI 436 Decision Support Systems Term Project

Group 17: Annie Yuan, Edward Jeong, Nishesh Jagga, Tian Xing Zhang

Description

This tool will suggest renovations to a house based on the details a user enters.

Files

• model.py contains all the functions that deal with data processing and the Linear Regression model from SciKit-Learn.
• app.py contains is run by Streamlit and describes the UI layout and the user interaction.
• data/features.yaml contains information about the features that are used for prediction. This is used throughout the app via the FeatureInfo class.
• data/train.csv contains the training data of the Ames Housing Dataset retreived from Kaggle.
• requirements.txt lists all the Python packages that are needed for this app to run.
• Dockerfile contains the instructions to build a Docker image.

Running the program

Local

1. Install Python 3.11 (might work with 3.9 or newer)
2. Install requirements via the Terminal: pip install -r requirements.txt
3. Run the Streamlit app: streamlit run app.py
4. Access the app at the Network URL shown

Docker

1. Install Docker
2. Pull the image: docker pull ghcr.io/invisco/msci436-group17-dss-project:main
3. Run the container: docker run --rm -p 8080:8080 msci436-group17-dss-project:main
4. Access the app at localhost:8080

Development

1. Follow this guide to setup Streamlit.
2. Run pip install -r requirements.txt instead of the command in those instructions.
3. Start working.