Question

我写了一个KNN插补实现，我想让StratifiedKFold检查要使用的K和距离矩阵。我遇到了一个错误：似乎无法将我的估算器识别为回归器（“评分”功能用于回归）。

我的代码：

skf = StratifiedKFold(n_splits=10, shuffle=False, random_state=12)
NN = KnnImputation() # my own function
gridSearchNN = GridSearchCV(NN, param_grid=params, scoring='mean_squared_error', n_jobs=numIter,
                            cv=skf.split(xTrain, yTrain), verbose=verbose)
gridSearchNN.fit(xTrain, yTrain)

错误：

  File "........\dataImputation.py", line 63, in knnImputationMethod
    gridSearchNN.fit(xTrain, yTrain)
  File "C:\Users\...\Anaconda3\lib\site-packages\sklearn\model_selection\_search.py", line 651, in fit
    cv = check_cv(self.cv, y, classifier=is_classifier(estimator))
  File "C:\Users\....\Anaconda3\lib\site-packages\sklearn\model_selection\_split.py", line 2068, in check_cv
    return _CVIterableWrapper(cv)
  File "C:\Users\....\Anaconda3\lib\site-packages\sklearn\model_selection\_split.py", line 1966, in __init__
    self.cv = list(cv)
  File "C:\Users\...\Anaconda3\lib\site-packages\sklearn\model_selection\_split.py", line 331, in split
    for train, test in super(_BaseKFold, self).split(X, y, groups):
  File "C:\Users\...\Anaconda3\lib\site-packages\sklearn\model_selection\_split.py", line 100, in split
    for test_index in self._iter_test_masks(X, y, groups):
  File "C:\Users\...\Anaconda3\lib\site-packages\sklearn\model_selection\_split.py", line 681, in _iter_test_masks
    test_folds = self._make_test_folds(X, y)
  File "C:\Users\...\Anaconda3\lib\site-packages\sklearn\model_selection\_split.py", line 636, in _make_test_folds
    allowed_target_types, type_of_target_y))
ValueError: Supported target types are: ('binary', 'multiclass'). Got 'continuous' instead.

在“ GridSearchCV”过程中，我看到它进入了“ is_classifier”而不是“ is_regressor”。

有什么想法吗？

Answer 1

StratifiedKFold

将组信息考虑在内以避免构建   折叠具有不平衡的类分布（对于二进制或多类   分类任务）。

StratifiedKFold仅适用于分类数据，而不能用于回归。

https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.KFold.html

将StratifiedKFold替换为KFold

您可以在此处查看来源：

https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/model_selection/_split.py#L570

在自己的估算器上进行网格搜索，并在python中使用连续目标

1 个答案: