In earlier times buyers approached real estate agents to buy their dream property. Every buyer has different requirements making it difficult for the agent to predict the House Price. This is where Machine Learning plays an important role, Data Science enthusiasts have tried to fasten and ease this process. To increase an agent's productivity, we implemented several advanced regression techniques to predict the sale price of the house. The model consists of various preprocessing techniques, visualizations, and detailed analysis. Trained several machine learning models like Random Forest regressor, XGBoost regressor and Stacking and Blending techniques, ultimately getting best RMSE score of 0.053 with a stack of CatBoost and Bayesian Regressors. After various permutation and combinations, we arrived at a model that gives a low RMSE value. (Lower RMSE value = Good model)
- Download the .ipynb file
- Run the python notebook in colab or jupyter notebook