交叉验证 cross_validation

from sklearn.datasets import load_boston, load_iris
from sklearn.linear_model import LinearRegression
from sklearn.neighbors import KNeighborsClassifier
import numpy as np
from sklearn import preprocessing #标准化
from sklearn.cross_validation import cross_val_score
from sklearn.cross_validation import train_test_split
a = np.array([[10,2.7,3.6], [-100, 5, -2], [120, 20, 40]],dtype=np.float64)


iris1 = load_iris()
print(iris1)

iris_X = iris1.data
iris = preprocessing.scale(iris_X) #标准化
print(iris)
iris_y = iris1.target

knn = KNeighborsClassifier(n_neighbors=5)
#把原数据集划分五块，每次取1／5为test集，取五次，求哪一个更精确
scores = cross_val_score(knn, iris_X,iris_y,cv = 5,scoring = 'accuracy') #for classification
loss = -cross_val_score(knn, iris_X,iris_y,cv = 5,scoring = 'mean_squared_error') #for regression回归 与真实值的误差



print(scores.mean())