Shortlist Promising Models

In this section, you will fit and compare three regression models to your transformed data, both with and without the G$1/$G$2$ columns, using cross$-$validation. Follow the steps below, using the specified variable names.

Initialize Three Regression Models

Linear Regression

Support Vector Machine $($SVM$)$ Regression

Lasso Regression

Compare Models with Cross$-$Validation

Using the above models, perform cross$-$validation on each model using both sets of transformed data $($with and without G$1/$G$2$ columns$).$

Instructions for Submission

Initialize the Models: Instantiate a Linear Regression, SVM Regression, and Lasso Regression model.

Use the specified variable names for the respective models:

lin$\_$reg

svm$\_$reg

lasso$\_$reg

Cross$-$Validation: Using both sets of transformed data $($with and without G$1/$G$2$ columns$),$ perform $3-$fold cross$-$validation for each model using RMSE as the metric.

You will run cross$-$validation six times $($e$.$g$.,$ cross$-$validation of the linear regression model with the G$1/$G$2$ data, cross$-$validation of the linear regression model without the G$1/$G$2$ data, etc.$)$

Use the specified variable names to save each respective array of scores:

cv$\_$scores$\_$lin$\_$reg$\_$with$\_$grades

cv$\_$scores$\_$lin$\_$reg$\_$without$\_$grades

cv$\_$scores$\_$svm$\_$with$\_$grades

cv$\_$scores$\_$svm$\_$without$\_$grades

cv$\_$scores$\_$lasso$\_$with$\_$grades

cv$\_$scores$\_$lasso$\_$without$\_$grades

Use the specified variable names to save the mean of each cross$-$validation array and print it to view your mean scores:

rmse$\_$lin$\_$reg$\_$with$\_$grades

rmse$\_$lin$\_$reg$\_$without$\_$grades

rmse$\_$svm$\_$with$\_$grades

rmse$\_$svm$\_$without$\_$grades

rmse$\_$lasso$\_$with$\_$grades

rmse$\_$lasso$\_$without$\_$grades