Merge 297fbe6 into 98fe01f

rasbt · Oct 11, 2019 · 1c3c024 · 1c3c024
2 parents 98fe01f + 297fbe6
commit 1c3c024
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 4 deletions.
diff --git a/docs/sources/CHANGELOG.md b/docs/sources/CHANGELOG.md
@@ -29,7 +29,7 @@ The CHANGELOG for the current development version is available at
 - `fpmax` returns an empty data frame now instead of raising an error if the frequent itemset set is empty. ([#573](https://github.com/rasbt/mlxtend/pull/573) via [Steve Harenberg](https://github.com/harenbergsd))
 - Fixes and issue in `mlxtend.plotting.plot_confusion_matrix`, where the font-color choice for medium-dark cells was not ideal and hard to read. [#588](https://github.com/rasbt/mlxtend/pull/588) via [sohrabtowfighi](https://github.com/sohrabtowfighi))
 - The `svd` mode of `mlxtend.feature_extraction.PrincipalComponentAnalysis` now also *n-1* degrees of freedom instead of *n* d.o.f. when computing the eigenvalues to match the behavior of `eigen`. [#595](https://github.com/rasbt/mlxtend/pull/595)
-
+- Disable input validation for `StackingCVClassifier` because it causes issues if pipelines are used as input. [#606](https://github.com/rasbt/mlxtend/pull/606)
 
 
 

diff --git a/mlxtend/classifier/stacking_cv_classification.py b/mlxtend/classifier/stacking_cv_classification.py
@@ -19,7 +19,7 @@
 from sklearn.base import clone
 from sklearn.model_selection import cross_val_predict
 from sklearn.model_selection._split import check_cv
-from sklearn.utils import check_X_y
+# from sklearn.utils import check_X_y
 
 
 class StackingCVClassifier(_BaseXComposition, ClassifierMixin,
@@ -207,8 +207,11 @@ def fit(self, X, y, groups=None, sample_weight=None):
             final_cv.shuffle = self.shuffle
             final_cv.random_state = self.random_state
 
-        # Input validation.
-        X, y = check_X_y(X, y, accept_sparse=['csc', 'csr'], dtype=None)
+        # Disable global input validation, because it causes issue when
+        # pipelines are used that perform preprocessing on X. I.e., X may
+        # not be directly passed to the classifiers, which is why this code
+        # would raise unecessary errors at this point.
+        # X, y = check_X_y(X, y, accept_sparse=['csc', 'csr'], dtype=None)
 
         if sample_weight is None:
             fit_params = None