Let's now see how the three models compare using the root mean squared error $($RMSE$)$ and R$-$squared metrics.

Exercise $7$:

Using your ols$\_$model, sgd$\_$model, and ridge$\_$model make predictions using your X$\_$test data and save these predictions as ols$\_$pred, sgd$\_$pred, and ridge$\_$pred respectively.

Find the RMSE for each model using Scikit$-$Learn's mean$\_$squared$\_$error function passing it your y$\_$test and predictions from step $1$ above. Round these RMSE scores to $4$ decimal places. Save these scores as rmse$\_$ols, rmse$\_$sgd$,$ and rmse$\_$ridge respectively. Hint: Remember that we are asking for RMSE and not MSE. Check Scikit$-$Learn's documentation for how to calculate the RMSE value.

Find the R$-$squared values for each model using Scikit$-$Learn's r$2\_$score function passing it your y$\_$test and predictions from step $1$ above. Round these R$-$squared scores to $4$ decimal places. Save these scores as r$2\_$ols, r$2\_$sgd$,$ and r$2\_$ridge respectively.

Print the MSE and R$-$squared scores for each model.