y_train_prob = clf.predict_proba(X_train)[:, 1]
print(clf.predict(x_test))
# 返回预测标签,就是预测值是哪些,比如[0,1]
print(clf.predict_proba(x_test))
# 返回预测属于某标签的概率
就是属于某类预测值的概率是多少,比如为0的概率0.66
而对于[:,1] 是取二维数组中第二维的所有数据
X_train
[[1,0 ...,1]
...
]
clf.predict_proba(X_train)
[[0.94005474 ,0.05994526]
...
]
含义是:预测X_rain[1. 0. 1. ... 0. 1. 0.]的标签是0的概率为0.94005474 ,为1的概率为0.05994526