I want to apply k-fold cross validation on the following regression models:

1. Linear Regression
2. Polynomial Regression
3. Support Vector Regression
4. Decision Tree Regression
5. Random Forest Regression

I am able to apply k-fold cross validation on all except polynomial regression which gives me this error PolynomialFeatures' object has no attribute 'predict. How to work around this issue. Also am I doing the job correctly, actually my main motive is to see which model is performing better, so is there a better way to do this job ??

# Compare Algorithms
import pandas
import matplotlib.pyplot as plt
from sklearn import model_selection
from sklearn.linear_model import LinearRegressionfrom sklearn.preprocessing import PolynomialFeatures
from sklearn.svm import SVR
from sklearn.tree import DecisionTreeRegressor
from sklearn.ensemble import RandomForestRegressor# load dataset
names = ['YearsExperience', 'Salary']
dataframe = pandas.read_csv('Salary_Data.csv', names=names)
array = dataframe.values
X = array[1:,0]
Y = array[1:,1]X = X.reshape(-1, 1)
Y = Y.reshape(-1, 1)# prepare configuration for cross validation test harness
seed = 7# prepare models
models = []
models.append(('LR', LinearRegression()))models.append(('PR', PolynomialFeatures(degree = 4)))
models.append(('SVR', SVR(kernel = 'rbf')))
models.append(('DTR', DecisionTreeRegressor()))
models.append(('RFR', RandomForestRegressor(n_estimators = 10)))# evaluate each model in turn
results = []
names = []
scoring = 'neg_mean_absolute_error'
for name, model in models:kfold = model_selection.KFold(n_splits=10, random_state=seed)cv_results = model_selection.cross_val_score(model, X, Y.ravel(), cv=kfold, scoring=scoring)results.append(cv_results)names.append(name)msg = "%s: %f (%f)" % (name, cv_results.mean(), cv_results.std())print(msg)# boxplot algorithm comparison
fig = plt.figure()
fig.suptitle('Algorithm Comparison')
ax = fig.add_subplot(111)
plt.boxplot(results)
ax.set_xticklabels(names)
plt.show()

In sklearn you get polynomial regression by:

1. generating polynomial and interaction features on your original dataset by using sklearn.preprocessing.PolynomialFeatures
2. running ordinary least squares Linear Regression on the transformed dataset by using sklearn.linear_model.LinearRegression

Toy example:

from sklearn.preprocessing import PolynomialFeatures
from sklearn import linear_model# Create linear regression object
poly = PolynomialFeatures(degree=3)X_train = poly.fit_transform(X_train)
X_test = poly.fit_transform(X_test)model = linear_model.LinearRegression()
model.fit(X_train, y_train)print(model.score(X_train, y_train))