ENH Change fit_transform of MissingIndicator class to get mask only once (scikit-learn#14356)

Author: harsh020

doc/whats_new/v0.22.rst

@@ -56,7 +56,7 @@ Changelog
 :mod:`sklearn.datasets`
 .......................
 
-- |Feature| :func:`datasets.fetch_openml` now supports heterogeneous data using pandas 
+- |Feature| :func:`datasets.fetch_openml` now supports heterogeneous data using pandas
   by setting `as_frame=True`. :pr:`13902` by `Thomas Fan`_.
 
 - |Enhancement| The parameter `return_X_y` was added to
@@ -107,6 +107,11 @@ Changelog
   preserve the class balance of the original training set. :pr:`14194`
   by :user:`Johann Faouzi <johannfaouzi>`.
 
+- |Efficiency| :func:`ensemble.MissingIndicator.fit_transform` the
+  _get_missing_features_info function is now called once when calling
+  fit_transform for MissingIndicator class. :pr:`14356` by :user:
+  `Harsh Soni <harsh020>`.
+
 - |Fix| :class:`ensemble.AdaBoostClassifier` computes probabilities based on
   the decision function as in the literature. Thus, `predict` and
   `predict_proba` give consistent results.
@@ -139,7 +144,7 @@ Changelog
 - |Feature| Added multiclass support to :func:`metrics.roc_auc_score`.
   :issue:`12789` by :user:`Kathy Chen <kathyxchen>`,
   :user:`Mohamed Maskani <maskani-moh>`, and :user:`Thomas Fan <thomasjpfan>`.
-  
+
 - |Feature| Add :class:`metrics.mean_tweedie_deviance` measuring the
   Tweedie deviance for a power parameter ``p``. Also add mean Poisson deviance
   :class:`metrics.mean_poisson_deviance` and mean Gamma deviance
@@ -190,7 +195,7 @@ Changelog
 :mod:`sklearn.cluster`
 ......................
 
-- |Enhancement| :class:`cluster.SpectralClustering` now accepts a ``n_components`` 
+- |Enhancement| :class:`cluster.SpectralClustering` now accepts a ``n_components``
   parameter. This parameter extends `SpectralClustering` class functionality to
   match `spectral_clustering`.
   :pr:`13726` by :user:`Shuzhe Xiao <fdas3213>`.

sklearn/impute/_base.py

@@ -586,7 +586,7 @@ def _validate_input(self, X):
 
         return X
 
-    def fit(self, X, y=None):
+    def _fit(self, X, y=None):
         """Fit the transformer on X.
 
         Parameters
@@ -597,8 +597,10 @@ def fit(self, X, y=None):
 
         Returns
         -------
-        self : object
-            Returns self.
+        imputer_mask : {ndarray or sparse matrix}, shape (n_samples, \
+        n_features)
+            The imputer mask of the original data.
+
         """
         X = self._validate_input(X)
         self._n_features = X.shape[1]
@@ -612,7 +614,26 @@ def fit(self, X, y=None):
             raise ValueError("'sparse' has to be a boolean or 'auto'. "
                              "Got {!r} instead.".format(self.sparse))
 
-        self.features_ = self._get_missing_features_info(X)[1]
+        missing_features_info = self._get_missing_features_info(X)
+        self.features_ = missing_features_info[1]
+
+        return missing_features_info[0]
+
+    def fit(self, X, y=None):
+        """Fit the transformer on X.
+
+        Parameters
+        ----------
+        X : {array-like, sparse matrix}, shape (n_samples, n_features)
+            Input data, where ``n_samples`` is the number of samples and
+            ``n_features`` is the number of features.
+
+        Returns
+        -------
+        self : object
+            Returns self.
+        """
+        self._fit(X, y)
 
         return self
 
@@ -667,7 +688,12 @@ def fit_transform(self, X, y=None):
             will be boolean.
 
         """
-        return self.fit(X, y).transform(X)
+        imputer_mask = self._fit(X, y)
+
+        if self.features_.size < self._n_features:
+            imputer_mask = imputer_mask[:, self.features_]
+
+        return imputer_mask
 
     def _more_tags(self):
         return {'allow_nan': True,