Question

我想对以下回归模型应用k倍交叉验证：

线性回归
多项式回归
支持向量回归
决策树回归
随机森林回归

除了多项式回归之外，我都可以对所有错误应用k倍交叉验证，这给了我这个错误PolynomialFeatures' object has no attribute 'predict。如何解决此问题。我也是在正确地完成这项工作吗，实际上我的主要动机是看哪种模型表现更好，所以有更好的方法来完成这项工作吗？

# Compare Algorithms
import pandas
import matplotlib.pyplot as plt
from sklearn import model_selection
from sklearn.linear_model import LinearRegression

from 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()

Answer 1

在sklearn中，您可以通过以下方式获得多项式回归：

使用sklearn.preprocessing.PolynomialFeatures
使用sklearn.linear_model.LinearRegression在转换后的数据集上运行普通最小二乘线性回归

玩具示例：

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))

Answer 2

如果有人要参考，这里是代码的更改部分：

# prepare models
models = []
models.append(('LR', LinearRegression()))

models.append(('PR', LinearRegression()))
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)
    if name == 'PR':
        poly_reg = PolynomialFeatures(degree = 4)
        X_poly = poly_reg.fit_transform(X)
        cv_results = model_selection.cross_val_score(model, X_poly, Y.ravel(), cv=kfold, scoring=scoring)
    else:
        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())

'PolynomialFeatures'对象没有属性'predict'

2 个答案: