scikit-learn
diff --git a/‎sklearn/utils/sparsefuncs_fast.pyx
Lines changed: 4 additions & 3 deletions b/‎sklearn/utils/sparsefuncs_fast.pyx
Lines changed: 4 additions & 3 deletions
diff --git a/‎sklearn/utils/tests/test_sparsefuncs.py
Lines changed: 15 additions & 14 deletions b/‎sklearn/utils/tests/test_sparsefuncs.py
Lines changed: 15 additions & 14 deletions
@@ -11,6 +11,7 @@ cimport numpy as np
 import numpy as np
 import scipy.sparse as sp
 cimport cython
+from cython cimport floating
 
 np.import_array()
 
@@ -360,11 +361,11 @@ cdef void add_row_csr(np.ndarray[np.float64_t, ndim=1] data,
 def assign_rows_csr(X,
                     np.ndarray[np.npy_intp, ndim=1] X_rows,
                     np.ndarray[np.npy_intp, ndim=1] out_rows,
-                    np.ndarray[np.float64_t, ndim=2, mode="c"] out):
+                    np.ndarray[floating, ndim=2, mode="c"] out):
     """Densify selected rows of a CSR matrix into a preallocated array.
 
     Like out[out_rows] = X[X_rows].toarray() but without copying.
-    Only supported for dtype=np.float64.
+    No-copy supported for both dtype=np.float32 and dtype=np.float64.
 
     Parameters
     ----------
@@ -378,7 +379,7 @@ def assign_rows_csr(X,
         # but int is what scipy.sparse uses.
         int i, ind, j
         np.npy_intp rX
-        np.ndarray[DOUBLE, ndim=1] data = X.data
+        np.ndarray[floating, ndim=1] data = X.data
         np.ndarray[int, ndim=1] indices = X.indices, indptr = X.indptr
 
     if X_rows.shape[0] != out_rows.shape[0]:
 
@@ -179,20 +179,21 @@ def test_mean_variance_illegal_axis():
 
 
 def test_densify_rows():
-    X = sp.csr_matrix([[0, 3, 0],
-                       [2, 4, 0],
-                       [0, 0, 0],
-                       [9, 8, 7],
-                       [4, 0, 5]], dtype=np.float64)
-    X_rows = np.array([0, 2, 3], dtype=np.intp)
-    out = np.ones((6, X.shape[1]), dtype=np.float64)
-    out_rows = np.array([1, 3, 4], dtype=np.intp)
-
-    expect = np.ones_like(out)
-    expect[out_rows] = X[X_rows, :].toarray()
-
-    assign_rows_csr(X, X_rows, out_rows, out)
-    assert_array_equal(out, expect)
+    for dtype in (np.float32, np.float64):
+        X = sp.csr_matrix([[0, 3, 0],
+                        [2, 4, 0],
+                        [0, 0, 0],
+                        [9, 8, 7],
+                        [4, 0, 5]], dtype=dtype)
+        X_rows = np.array([0, 2, 3], dtype=np.intp)
+        out = np.ones((6, X.shape[1]), dtype=dtype)
+        out_rows = np.array([1, 3, 4], dtype=np.intp)
+
+        expect = np.ones_like(out)
+        expect[out_rows] = X[X_rows, :].toarray()
+
+        assign_rows_csr(X, X_rows, out_rows, out)
+        assert_array_equal(out, expect)
 
 
 def test_inplace_column_scale():