Using kernels for nonlinear SVR

Recall from our discussion at the beginning of this chapter that we can use a kernel function to fit a nonlinear epsilon-insensitive tube. In this section, we will run a nonlinear SVR with the land temperatures data that we worked with in the previous chapter. But first, we will construct a linear SVR with the same data for comparison.

We will model the average temperature for weather stations as a function of latitude and elevation. Follow these steps:

1. We will begin by loading the familiar libraries. The only new class is SVR from scikit-learn:

   import pandas as pd
   import numpy as np
   from sklearn.model_selection import train_test_split
   from sklearn.preprocessing import StandardScaler
   from sklearn.svm import LinearSVR, SVR
   from scipy.stats import uniform
   from sklearn.impute import SimpleImputer
   from sklearn.pipeline import make_pipeline
   from sklearn.compose import TransformedTargetRegressor
   from sklearn.impute import KNNImputer
   from sklearn...