- Perform Shapley Additive Explanation (SHAP) interpretation to determine the impact of attributes by their SHAP values and their interaction.
- Implemented Light Gradient Boosting Model algorithm to predict house's price with 17 numeric attributes inputs.
- Technology: Python, shap, lightgbm, optuna, numpy, pandas, streamlit, Google Colab, GitHub.
Ask a customer to describe their dream house, and they probably won’t begin with the height of the basement ceiling or the proximity to an east-west railroad. Different customers will have different preference but generally, what are the most important features to cost a house? With 79 explanatory variables describing every aspect of residential homes in Ames, Iowa given in /dataset/train.csv this projects helps you predict the final price of each home.
Milestone 1: link
Set up development environment
Milestone 2: link
Perform the SHAP interpretation of the house price prediction using Linear Regression model
Milestone 3: link
Repeat Milestone 2 but with Optuna and LightGBM model, then build a website by streamlit
Milestone 4: link
DEMO TIME !