Regression-on-Ames-Housing-Dataset

• Visualizing the univariate distribution of each continuous variable, and the distribution of the target

• Visualizing the dependency of the target on each continuous feature

• Cross-validating a linear regression model for nominal variables using one-hot encoding. And visualizing the relationship of the categorical variable of best R2 value with the target.

• Used column transformer and pipeline to encode ordinal, nominal, and discrete variables.

• Evaluate Linear Regression (OLS), Ridge, Lasso, and ElasticNet using cross-validation with the default parameters.

• Scaling the data within the pipeline using Standard Scaler.

• Tuned the parameters of the model using Gridsearch CV and visualized the dependence of the validation score on the parameters for Ridge, Lasso, and ElasticNet.

• Visualizing the coefficients of the resulting model to determine which features are important.