ENH add deprecation for projected gradient

TomDLT · TomDLT · commit 3269e6cee89d · 2015-06-12T17:07:00.000+02:00
diff --git a/sklearn/decomposition/nmf.py b/sklearn/decomposition/nmf.py
@@ -706,6 +706,8 @@ def non_negative_matrix_factorization(X, W=None, H=None, n_components=None,
                                random_state=random_state)
 
     if solver == 'proj-grad':
+        warnings.warn("'proj-grad' solver will be removed in release 0.19."
+                      " Use 'coordinate' solver instead.", DeprecationWarning)
         if update_H:  # fit_transform
             W, H, n_iter = _fit_projected_gradient(X, W, H, tol,
                                                    max_iter,
@@ -946,6 +948,8 @@ def transform(self, X):
         return W
 
 
+@deprecated("It will be removed in release 0.19. Use NMF instead."
+            "'proj-grad' solver is still available until release 0.19.")
 class ProjectedGradientNMF(NMF):
     __doc__ = NMF.__doc__
 
diff --git a/sklearn/decomposition/tests/test_nmf.py b/sklearn/decomposition/tests/test_nmf.py
@@ -23,6 +23,7 @@ def test_initialize_nn_output():
         assert_false((W < 0).any() or (H < 0).any())
 
 
+@ignore_warnings
 def test_parameter_checking():
     A = np.ones((2, 2))
     name = 'spam'
@@ -65,6 +66,7 @@ def test_initialize_variants():
         assert_true(np.allclose(evl[ref != 0], ref[ref != 0]))
 
 
+@ignore_warnings
 def test_nmf_fit_nn_output():
     # Test that the decomposition does not contain negative values
     A = np.c_[5 * np.ones(5) - np.arange(1, 6),
@@ -78,6 +80,7 @@ def test_nmf_fit_nn_output():
                          (transf < 0).any())
 
 
+@ignore_warnings
 def test_nmf_fit_close():
     # Test that the fit is not too far away
     for solver in ('proj-grad', 'coordinate'):
@@ -101,6 +104,7 @@ def test_nls_close():
     assert_true((np.abs(Ap - A) < 0.01).all())
 
 
+@ignore_warnings
 def test_nmf_transform():
     # Test that NMF.transform returns close values
     A = np.abs(random_state.randn(6, 5))
@@ -139,6 +143,7 @@ def test_nmf_sparseness():
     assert_greater(comp_sp, m.comp_sparseness_)
 
 
+@ignore_warnings
 def test_sparse_input():
     # Test that sparse matrices are accepted as input
     from scipy.sparse import csc_matrix
@@ -164,6 +169,7 @@ def test_sparse_input():
         assert_array_almost_equal(H1, H2)
 
 
+@ignore_warnings
 def test_sparse_transform():
     # Test that transform works on sparse data.  Issue #2124
     from scipy.sparse import csc_matrix
@@ -179,6 +185,7 @@ def test_sparse_transform():
         assert_array_almost_equal(A_fit_tr, A_tr, decimal=2)
 
 
+@ignore_warnings
 def test_non_negative_matrix_factorization_path():
     # Test path consistency between the class and the public function
     A = np.abs(random_state.randn(10, 10))
@@ -197,6 +204,7 @@ def test_non_negative_matrix_factorization_path():
         assert_array_almost_equal(W_nmf_bis, W_cls_bis, decimal=10)
 
 
+@ignore_warnings
 def test_non_negative_matrix_factorization_checking():
     A = np.ones((2, 2))
     # Test parameters checking is public function
