ENH do not allocate memory for temporary array of 1s

jnothman · jnothman · commit 301061ff2bd5 · 2013-12-12T11:48:49.000+11:00
diff --git a/sklearn/preprocessing/label.py b/sklearn/preprocessing/label.py
@@ -10,6 +10,7 @@
 import array
 
 import numpy as np
+from numpy.lib.stride_tricks import as_strided
 import scipy.sparse as sp
 
 from ..base import BaseEstimator, TransformerMixin
@@ -553,7 +554,9 @@ def _transform(self, y, class_mapping):
         for labels in y:
             indices.extend(set(class_mapping[label] for label in labels))
             indptr.append(len(indices))
-        data = np.ones(len(indices), dtype=int)
+        # virtual array of len(indices) 1s:
+        data = as_strided(np.array([1], dtype=int), strides=(0,),
+                          shape=(len(indices),))
         return sp.csr_matrix((data, indices, indptr),
                              shape=(len(indptr) - 1, len(class_mapping)))
 
