scikit-learn · amueller · Aug 7, 2019 · Jul 30, 2019 · Jul 31, 2019 · Jul 31, 2019
diff --git a/doc/whats_new/v0.22.rst b/doc/whats_new/v0.22.rst
@@ -45,6 +45,7 @@ Changelog
     :pr:`123456` by :user:`Joe Bloggs <joeongithub>`.
     where 123456 is the *pull request* number, not the issue number.
 
+
 :mod:`sklearn.base`
 ...................
 
@@ -143,12 +144,17 @@ Changelog
   :pr:`14114` by :user:`Guillaume Lemaitre <glemaitre>`.
 
 :mod:`sklearn.feature_extraction`
-.......................
+.................................
 
 - |Fix| Functions created by build_preprocessor and build_analyzer of
   :class:`feature_extraction.text.VectorizerMixin` can now be pickled.
   :pr:`14430` by :user:`Dillon Niederhut <deniederhut>`.
 
+- |API| Deprecated unused `copy` param for
+  :meth: `feature_extraction.text.TfidfVectorizer.transform` it will be
+  removed in v0.24. :pr:`14520` by
+  :user:`Guillem G. Subies <guillemgsubies>`.
+
 :mod:`sklearn.gaussian_process`
 ...............................
 

diff --git a/sklearn/feature_extraction/tests/test_text.py b/sklearn/feature_extraction/tests/test_text.py
@@ -509,6 +509,18 @@ def test_tfidf_vectorizer_setters():
     assert tv._tfidf.sublinear_tf
 
 
+# FIXME Remove copy parameter support in 0.24
+def test_tfidf_vectorizer_deprecationwarning():
+    msg = ("'copy' param is unused and has been deprecated since "
+           "version 0.22. Backward compatibility for 'copy' will "
+           "be removed in 0.24.")
+    with pytest.warns(DeprecationWarning, match=msg):
+        tv = TfidfVectorizer()
+        train_data = JUNK_FOOD_DOCS
+        tv.fit(train_data)
+        tv.transform(train_data, copy=True)
+
+
 @fails_if_pypy
 def test_hashing_vectorizer():
     v = HashingVectorizer()

diff --git a/sklearn/feature_extraction/text.py b/sklearn/feature_extraction/text.py
@@ -1729,7 +1729,7 @@ def fit_transform(self, raw_documents, y=None):
         # we set copy to False
         return self._tfidf.transform(X, copy=False)
 
-    def transform(self, raw_documents, copy=True):
+    def transform(self, raw_documents, copy="deprecated"):
         """Transform documents to document-term matrix.
 
         Uses the vocabulary and document frequencies (df) learned by fit (or
@@ -1744,13 +1744,24 @@ def transform(self, raw_documents, copy=True):
             Whether to copy X and operate on the copy or perform in-place
             operations.
 
+            .. deprecated:: 0.22
+               The `copy` parameter is unused and was deprecated in version
+               0.22 and will be removed in 0.24. This parameter will be
+               ignored.
+
         Returns
         -------
         X : sparse matrix, [n_samples, n_features]
             Tf-idf-weighted document-term matrix.
         """
         check_is_fitted(self, '_tfidf', 'The tfidf vector is not fitted')
 
+        # FIXME Remove copy parameter support in 0.24
+        if copy != "deprecated":
+            msg = ("'copy' param is unused and has been deprecated since "
+                   "version 0.22. Backward compatibility for 'copy' will "
+                   "be removed in 0.24.")
+            warnings.warn(msg, DeprecationWarning)
         X = super().transform(raw_documents)
         return self._tfidf.transform(X, copy=False)