是否可以在 cross_val_predict 中使用与 cross_val_score 中相同的 k 折叠？

Question

您好，如果我们执行以下操作来计算交叉验证的准确性：

cv_acc = cross_val_score(model, X_train, y_train, cv=5, scoring='accuracy')    

是否可以使用cross_val_score 中 k 折叠的相同输入来估计 y 预测并创建混淆矩阵（如下所示）？

y_pred = cross_val_predict(model, X_train, y_train, cv=5)
conf_mat = confusion_matrix(y_test, y_pred)

有没有办法存储 k 折叠在 cross_val_score 中的精确拆分方式以确保可比较的混淆矩阵？

干杯:)

Answer 1

以下应该有效：

from sklearn.model_selection import KFold, cross_val_score, cross_val_predict
k_folds = KFold(n_splits=5)
splits = list(k_folds.split(X_train, y_train)) # note list here as k_folds.split is a one-off generator
cv_acc = cross_val_score(model, X_train, y_train, cv=splits, scoring='accuracy')
y_pred = cross_val_predict(model, X_train, y_train, cv=splits)

有关更多信息，请参阅 cross_val_score 和 cross_val_predict 的文档。

一个完整的工作示例：

from sklearn.model_selection import KFold, cross_val_score, cross_val_predict
from sklearn.datasets import make_classification
from sklearn.linear_model import LogisticRegression

k_folds = KFold(n_splits=5)
X_train, y_train = make_classification(1000)
splits = list(k_folds.split(X_train, y_train))
model = LogisticRegression()
cv_acc = cross_val_score(model, X_train, y_train, cv=splits, scoring='accuracy')
y_pred = cross_val_predict(model, X_train, y_train, cv=splits)