formatting/error fixes

lobpcg · lobpcg · commit 17c9196b8596 · 2018-10-08T10:46:36.000-04:00
diff --git a/sklearn/decomposition/tests/test_pca.py b/sklearn/decomposition/tests/test_pca.py
@@ -358,7 +358,8 @@ def test_pca_validation(solver):
     # parameter raise errors
     X = np.array([[0, 1, 0], [1, 0, 0]])
     smallest_d = 2  # The smallest dimension
-    lower_limit = {'randomized': 1, 'arpack': 1, 'full': 0, 'auto': 0}
+    lower_limit = {'randomized': 1, 'arpack': 1, 'lobpcg': 1,
+                   'full': 0, 'auto': 0}
 
     # We conduct the same test on X.T so that it is invariant to axis.
     for data in [X, X.T]:
diff --git a/sklearn/utils/lobpcg.py b/sklearn/utils/lobpcg.py
@@ -1,16 +1,14 @@
 """
-Pure SciPy implementation of Locally Optimal Block Preconditioned Conjugate
-Gradient Method (LOBPCG), see
-https://bitbucket.org/joseroman/blopex
-
-The original code is at scipy\sparse\linalg\eigen\lobpcg\lobpcg.py
+Locally Optimal Block Preconditioned Conjugate Gradient Method (LOBPCG)
+"""
 
-License: BSD
+# sk-learn copy of scipy\sparse\linalg\eigen\lobpcg\lobpcg.py
 
-Authors: Robert Cimrman, Andrew Knyazev
+# Author: Robert Cimrman,
+#         Andrew Knyazev
+# Examples in tests directory contributed by Nils Wagner.
+# License: BSD
 
-Examples in tests directory contributed by Nils Wagner.
-"""
 import numpy as np
 
 from numpy.testing import assert_allclose
@@ -472,9 +470,9 @@ def lobpcg(A, X,
                             [xbp.T.conj(), wbp.T.conj(), ident]])
         else:
             gramA = np.bmat([[np.diag(_lambda), xaw],
-                              [xaw.T.conj(), waw]])
+                            [xaw.T.conj(), waw]])
             gramB = np.bmat([[ident0, xbw],
-                              [xbw.T.conj(), ident]])
+                            [xbw.T.conj(), ident]])
 
         if verbosityLevel > 10:
             save(gramA, 'gramA')
