add classes_

.github/workflows/python-package-conda.yml

@@ -25,14 +25,15 @@ jobs:
     - name: Install dependencies
       run: |
         conda env update --file environment.yml --name base
+        conda install python=3.9 -y -q
     - name: Test with pytest
       run: |
-        conda update conda -y -q
         conda config --add channels conda-forge
         conda update conda -y -q
-        conda install scikit-learn=1.0.2 pandas=1.3.5  -y -q
-        conda install tensorflow joblib pytest -y -q
-        conda install imageio scikit-image -y -q
+        conda install libcblas -y -q
+        pip install scikit-learn==1.0.2 pandas==1.3.5
+        pip install tensorflow joblib pytest
+        pip install imageio scikit-image
         conda install dlib -y -q
         pip install markdown
         pip install coverage

docs/sources/CHANGELOG.md

@@ -19,6 +19,7 @@ The CHANGELOG for the current development version is available at
 
 ##### Changes
 
+- Adds `classes_` attribute to `StackingClassifier`, `StackingCVClassifier` and `EnsembleVoteClassifier` to make these models compatible with scikit-learn 2.3.0 utility functions. ([#1060](https://github.com/rasbt/mlxtend/issues/1060))
 - Address NumPy deprecations to make mlxtend compatible to NumPy 1.24 
 - Changed the signature of the `LinearRegression` model of sklearn in the test removing the `normalize` parameter as it is deprecated. ([#1036](https://github.com/rasbt/mlxtend/issues/1036))
 

environment.yml

@@ -7,7 +7,7 @@ dependencies:
   - pandas>=1.3.4
   - pip>=21.3.1
   - pytest>=6.2.5
-  - scikit-learn>=1.0.1
+  - scikit-learn>=1.3.0
   - scipy>=1.7.3
   - setuptools>=59.4.0
   - pip:

mlxtend/classifier/ensemble_vote.py

@@ -165,7 +165,7 @@ def fit(self, X, y, sample_weight=None):
             )
 
         self.le_ = LabelEncoder()
-        self.le_.fit(y)
+        y_transform = self.le_.fit(y)
         self.classes_ = self.le_.classes_
 
         if not self.fit_base_estimators and self.use_clones:
@@ -199,9 +199,9 @@ def fit(self, X, y, sample_weight=None):
                     print(_name_estimators((clf,))[0][1])
 
                 if sample_weight is None:
-                    clf.fit(X, self.le_.transform(y))
+                    clf.fit(X, y_transform)
                 else:
-                    clf.fit(X, self.le_.transform(y), sample_weight=sample_weight)
+                    clf.fit(X, y_transform, sample_weight=sample_weight)
         return self
 
     def predict(self, X):

mlxtend/classifier/stacking_classification.py

@@ -13,6 +13,7 @@
 import numpy as np
 from scipy import sparse
 from sklearn.base import TransformerMixin, clone
+from sklearn.preprocessing import LabelEncoder
 
 from ..externals.estimator_checks import check_is_fitted
 from ..externals.name_estimators import _name_estimators
@@ -163,6 +164,11 @@ def fit(self, X, y, sample_weight=None):
         self : object
 
         """
+
+        self.le_ = LabelEncoder()
+        y_transform = self.le_.fit_transform(y)
+        self.classes_ = self.le_.classes_
+
         if not self.fit_base_estimators:
             warnings.warn(
                 "fit_base_estimators=False " "enforces use_clones to be `False`"
@@ -195,9 +201,9 @@ def fit(self, X, y, sample_weight=None):
                 if self.verbose > 1:
                     print(_name_estimators((clf,))[0][1])
                 if sample_weight is None:
-                    clf.fit(X, y)
+                    clf.fit(X, y_transform)
                 else:
-                    clf.fit(X, y, sample_weight=sample_weight)
+                    clf.fit(X, y_transform, sample_weight=sample_weight)
 
         meta_features = self.predict_meta_features(X)
 
@@ -212,9 +218,9 @@ def fit(self, X, y, sample_weight=None):
             meta_features = np.hstack((X, meta_features))
 
         if sample_weight is None:
-            self.meta_clf_.fit(meta_features, y)
+            self.meta_clf_.fit(meta_features, y_transform)
         else:
-            self.meta_clf_.fit(meta_features, y, sample_weight=sample_weight)
+            self.meta_clf_.fit(meta_features, y_transform, sample_weight=sample_weight)
 
         return self
 

mlxtend/classifier/stacking_cv_classification.py

@@ -14,6 +14,7 @@
 from sklearn.base import TransformerMixin, clone
 from sklearn.model_selection import cross_val_predict
 from sklearn.model_selection._split import check_cv
+from sklearn.preprocessing import LabelEncoder
 
 from ..externals.estimator_checks import check_is_fitted
 from ..externals.name_estimators import _name_estimators
@@ -130,11 +131,17 @@ class StackingCVClassifier(
         Fitted classifiers (clones of the original classifiers)
     meta_clf_ : estimator
         Fitted meta-classifier (clone of the original meta-estimator)
+    le_ : `sklearn.preprocessing.LabelEncoder`
+        Transformer used to encode the labels during fit and decode during
+        prediction.
+    classes_ : ndarray of shape (n_classes,)
+        The classes labels.
     train_meta_features : numpy array, shape = [n_samples, n_classifiers]
         meta-features for training data, where n_samples is the
         number of samples
         in training data and n_classifiers is the number of classfiers.
 
+
     Examples
     -----------
     For usage examples, please see
@@ -212,6 +219,11 @@ def fit(self, X, y, groups=None, sample_weight=None):
         self : object
 
         """
+
+        self.le_ = LabelEncoder()
+        y_transform = self.le_.fit_transform(y)
+        self.classes_ = self.le_.classes_
+
         if self.use_clones:
             self.clfs_ = clone(self.classifiers)
             self.meta_clf_ = clone(self.meta_classifier)
@@ -221,7 +233,7 @@ def fit(self, X, y, groups=None, sample_weight=None):
         if self.verbose > 0:
             print("Fitting %d classifiers..." % (len(self.classifiers)))
 
-        final_cv = check_cv(self.cv, y, classifier=self.stratify)
+        final_cv = check_cv(self.cv, y_transform, classifier=self.stratify)
         if isinstance(self.cv, int):
             # Override shuffle parameter in case of self generated
             # cross-validation strategy
@@ -287,18 +299,18 @@ def fit(self, X, y, groups=None, sample_weight=None):
         # Fit the base models correctly this time using ALL the training set
         for model in self.clfs_:
             if sample_weight is None:
-                model.fit(X, y)
+                model.fit(X, y_transform)
             else:
-                model.fit(X, y, sample_weight=sample_weight)
+                model.fit(X, y_transform, sample_weight=sample_weight)
 
         # Fit the secondary model
         if self.use_features_in_secondary:
             meta_features = self._stack_first_level_features(X, meta_features)
 
         if sample_weight is None:
-            self.meta_clf_.fit(meta_features, y)
+            self.meta_clf_.fit(meta_features, y_transform)
         else:
-            self.meta_clf_.fit(meta_features, y, sample_weight=sample_weight)
+            self.meta_clf_.fit(meta_features, y_transform, sample_weight=sample_weight)
 
         return self
 

requirements.txt

@@ -1,6 +1,6 @@
 scipy>=1.2.1
 numpy>=1.16.2
 pandas>=0.24.2
-scikit-learn>=1.0.2
+scikit-learn>=1.3.0
 matplotlib>=3.0.0
-joblib>=0.13.2
+joblib>=0.13.2