Generate the necessary features to fit polynomial regressions up to the 20th degree (up to and including the x20 term) on the training data. Hint: You will be fitting multi-variate linear regression models with polynomial features of x. Familiarize yourself with sklearn. preprocessing. PolynomialFeatures . Here, we're just asking you to practice generating the features. You'll pass one of them into the autograder for a quick check (although the autograder will not be very strict, so if you end up failing the next test case definitely make sure your work here is correct!)
Now, let's check to make sure your highest-degree polynomial features are correct; namely the set of features that includes X20, or PolynomialFeatures (degree 20). Please set polynomial_features_test_df as this dataframe. = If you do not receive full points, that means that either you have the wrong number of columns or some column values aren't correct!
Calculate the training MSE and the test MSE for 20 polynomial models up to degree 20. Store these as lists named mse_train and mse_test respectively. Hint: Familiarize yourself with the sklearn. metrics. mean_squared_error package and try to automate the process. ]: mse_train mse test [ ]