def Call_SV_Reg(X_train, y_train, X_test, y_test):
"""
SVM Regression
"""
clf = SVR()
parameters = [{
'C': [1, 10, 100, 1000],
'gamma': [1e-1, 1, 1e1],
'kernel': ['rbf', 'linear', 'poly', 'sigmoid'],
'degree': [3],
'epsilon': [0.9]
}]
scoring_function = make_scorer(r2_score, greater_is_better=True)
# Make the GridSearchCV object
clf = GridSearchCV(clf, parameters, scoring=scoring_function, cv=10)
clf.fit(X_train, y_train)
Predicted = clf.predict(X_test)
print("SVR Score = ", clf.score(X_test, y_test))
MSE = mean_squared_error(y_test, Predicted)
R2 = r2_score(y_test, Predicted)
plot_regression(y_test, 'SV Reg', Predicted)
return "SVR Mean Square Error =", MSE, "SVR R2 =", R2
def Call_KNN_Reg(X_train, y_train, X_test, y_test):
"""
KNN Regression
"""
clf = KNeighborsRegressor()
clf.fit(X_train, y_train)
Predicted = clf.predict(X_test)
print("KNeighborsRegressor Score = ", clf.score(X_test, y_test))
MSE = mean_squared_error(y_test, Predicted)
R2 = r2_score(y_test, Predicted)
plot_regression(y_test, 'KNN Reg', Predicted)
return "KNeighborsRegressor MSE =", MSE, "KNeighborsRegressor R2 =", R2
def Call_GradientBoosting_Reg(X_train, y_train, X_test, y_test):
"""
Gradient Boosting Regression
"""
clf = ensemble.GradientBoostingRegressor()
clf.fit(X_train, y_train)
Predicted = clf.predict(X_test)
print("GradientBoostingRegressor Score = ", clf.score(X_test, y_test))
MSE = mean_squared_error(y_test, Predicted)
R2 = r2_score(y_test, Predicted)
plot_regression(y_test, 'GradientBoosting Reg', Predicted)
return "GradientBoosting MSE =", MSE, "GradientBoosting R2 =", R2
def Call_AdaBoost_Reg(X_train, y_train, X_test, y_test):
"""
Ada Boost Regression
"""
clf = ensemble.AdaBoostRegressor()
clf.fit(X_train, y_train)
Predicted = clf.predict(X_test)
print("AdaBoostRegressor Score = ", clf.score(X_test, y_test))
MSE = mean_squared_error(y_test, Predicted)
R2 = r2_score(y_test, Predicted)
plot_regression(y_test, 'AdaBoost Reg', Predicted)
return "AdaBoost MSE =", MSE, "AdaBoost R2 =", R2
def Call_Bagging_Reg(X_train, y_train, X_test, y_test):
"""
Bagging Regression
"""
clf = ensemble.BaggingRegressor()
clf.fit(X_train, y_train)
Predicted = clf.predict(X_test)
print("BaggingRegressor Score = ", clf.score(X_test, y_test))
MSE = mean_squared_error(y_test, Predicted)
R2 = r2_score(y_test, Predicted)
plot_regression(y_test, 'Bagging Reg', Predicted)
return "BaggingRegressor MSE =", MSE, "BaggingRegressor R2 =", R2
def Call_DecisionTree_Reg(X_train, y_train, X_test, y_test):
""" Tunes a decision tree regressor model using GridSearchCV on the input data X
and target labels y and returns this optimal model. """
# Create a decision tree regressor object
regressor = DecisionTreeRegressor(max_depth=2,
min_samples_leaf=1,
min_samples_split=2,
splitter='best')
# Set up the parameters we wish to tune
parameters = [{
'max_depth': (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
17, 18, 19, 20),
'presort': ['True']
}]
# Make an appropriate scoring function
scoring_function = None
#scoring_function = make_scorer(performance_metric(), greater_is_better=False)
scoring_function = make_scorer(r2_score, greater_is_better=True)
# Make the GridSearchCV object
reg = None
reg = GridSearchCV(regressor, parameters, scoring=scoring_function, cv=10)
# Fit the learner to the data to obtain the optimal model with tuned parameters
reg.fit(X_train, y_train)
Predicted = reg.predict(X_test)
print("DecisionTreeRegressor = ", reg.score(X_test, y_test))
#print "Best model parameter: " + str(reg.best_params_)
#print "Best model estimator: " + str(reg.best_estimator_)
# Return the optimal model
Best_Estimator = reg.best_estimator_
MSE = mean_squared_error(y_test, Predicted)
R2 = r2_score(y_test, Predicted)
# print("y_test :",y_test.head(),y_test[1])
# print("y_test_1 :",y_test.shape)
plot_regression(y_test, 'DecisionTree Reg', Predicted)
learning_curves(X_train, y_train, X_test, y_test)
model_complexity(X_train, y_train, X_test, y_test)
return "DecisionTreeRegressor Best Estimator ", Best_Estimator, "DecisionTreeRegressor MSE =", MSE, "DecisionTreeRegressor R2 =", R2
def Call_RandomForest_Reg(X_train, y_train, X_test, y_test):
"""
Random Forest Regression
"""
clf = ensemble.RandomForestRegressor(n_estimators=100, n_jobs=-1)
parameters = [{
'n_estimators': [20],
'criterion': ['mse'],
'min_weight_fraction_leaf': [0.25],
'n_jobs': [-1]
}]
scoring_function = make_scorer(r2_score, greater_is_better=True)
# Make the GridSearchCV object
clf = GridSearchCV(clf, parameters, scoring=scoring_function, cv=10)
clf = clf.fit(X_train, y_train)
Predicted = clf.predict(X_test)
print("RandomForestRegressor Score = ", clf.score(X_test, y_test))
MSE = mean_squared_error(y_test, Predicted)
R2 = r2_score(y_test, Predicted)
plot_regression(y_test, 'RandomForest Reg', Predicted)
return "Random Forest Predicted Mean Square Error =", MSE, "and Random Forest R-Square =", R2