using rowvar flag in dpnp.cov

dpnp/dpnp_iface_statistics.py

@@ -244,8 +244,8 @@ def cov(x1, y=None, rowvar=True, bias=False, ddof=None, fweights=None, aweights=
 
     Limitations
     -----------
-    Input array ``m`` is supported as :obj:`dpnp.ndarray`.
-    Dimension of input array ``m`` is limited by ``m.ndim > 2``.
+    Input array ``x1`` is supported as :obj:`dpnp.ndarray`.
+    Dimension of input array ``x1`` is limited by ``x1.ndim > 2``.
     Size and shape of input arrays are supported to be equal.
     Prameters ``y`` is supported only with default value ``None``.
     Prameters ``rowvar`` is supported only with default value ``True``.
@@ -274,14 +274,14 @@ def cov(x1, y=None, rowvar=True, bias=False, ddof=None, fweights=None, aweights=
 
     """
 
+    if not rowvar and x1.shape[0] != 1:
+        x1 = dpnp.transpose(x1)
     x1_desc = dpnp.get_dpnp_descriptor(x1, copy_when_nondefault_queue=False)
     if x1_desc:
         if x1_desc.ndim > 2:
             pass
         elif y is not None:
             pass
-        elif not rowvar:
-            pass
         elif bias:
             pass
         elif ddof is not None:

tests/test_statistics.py

@@ -114,3 +114,16 @@ def test_bincount_weights(self, array, weights):
         expected = numpy.bincount(np_a, weights=weights)
         result = dpnp.bincount(dpnp_a, weights=weights)
         numpy.testing.assert_array_equal(expected, result)
+
+def test_cov_rowvar():
+    a = dpnp.array([[0, 2], [1, 1], [2, 0]],device='cpu')
+    b = numpy.array([[0, 2], [1, 1], [2, 0]])
+    numpy.testing.assert_array_equal(dpnp.cov(a.T), dpnp.cov(a,rowvar=False))
+    numpy.testing.assert_array_equal(numpy.cov(b,rowvar=False), dpnp.cov(a,rowvar=False))
+
+def test_cov_rowvar():
+    a = dpnp.array([[0, 1, 2]],device='cpu')
+    b = numpy.array([[0, 1, 2]])
+    numpy.testing.assert_array_equal(numpy.cov(b,rowvar=False), dpnp.cov(a,rowvar=False))
+
+