cross_val_score不考虑评分参数

Question

是否可以帮助我解释为什么来自scikit的cross_val_score学习不考虑scoring参数？我一定做错了，我真的找不到。

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split, cross_val_score
from sklearn import svm

data = load_iris()

precisions = cross_val_score(svm.SVC(), data['data'], data['target'], cv=4, scoring="precision_micro") 
recalls = cross_val_score(svm.SVC(), data['data'], data['target'], cv=4, scoring="recall_micro") 
accuracies = cross_val_score(svm.SVC(), data['data'], data['target'], cv=4, scoring="accuracy") 

print(precisions)
print(recalls)
print(accuracies)

返回：

[ 0.97435897  1.          0.94444444  0.97222222]
[ 0.97435897  1.          0.94444444  0.97222222]
[ 0.97435897  1.          0.94444444  0.97222222]