Canada Real Estate Price Predictor

BUILT BY

The Model

XGBoost with Bayesian optimization via Optuna. Trained on Canadian housing data to predict property values based on location, size, and amenities.

The Data

Trained on the Canada Housing dataset from Kaggle with thousands of listings across all provinces.

How It Works

Input Processing — Your property details are validated and encoded using a ColumnTransformer with one-hot encoding for categorical features and standard scaling for numerics.

Prediction — The XGBoost model (trained with log1p target transform) generates a prediction, which is then exponentiated to get the final CAD price.

Visualization — Property descriptions are sent to Imagen 4.0 Ultra which generates an AI image of your potential property.

About Me

Built by Mohammad Waliduddin, a 4th year Software Engineering student at York University.

I'm passionate about open source software and believe technology should be accessible to everyone. This project is built entirely with open tools and libraries – from XGBoost to Flask – because I believe in democratizing tech and empowering others to learn, build, and contribute.

Currently exploring AI/ML pipelines, agentic systems, and full-stack development. When not coding, you'll find me playing FIFA, tinkering with Linux, or on the football field.

GitHub LinkedIn

Location

Basics

Amenities

Results

Your Selection

Selected Location

View & edit amenities

Garage

Basement

Fireplace

Pool

Garden

Flooring

Quick Steps:

Click "Predict Price" to see your estimate

Click "Visualize" to see an AI-generated image