RuntimeError: The classifier does not expose "coef_" or "feature_importances_" attributes

Question

In my code, it throws a runtime error. Here, I am trying to fit RFE for regression data.

from sklearn.feature_selection import RFE
from sklearn.svm import SVR           
from sklearn.feature_selection import SelectKBest
from sklearn.preprocessing import *

scaler = StandardScaler().fit(trainFeatures)
xscaled = scaler.transform(trainFeatures)
estimator = SVR()
selector = RFE(estimator, dimension, step=1)
selector = selector.fit(xscaled, trainOutput.ravel())
selectedFeatures = selector.transform(xscaled)

Answer 1

As per this link, RFE only works with SVR when the kernel is linear.

By default, it is "rbf", so do:

estimator = SVR(kernel="linear")