Support 'nuc' norm in dpnp.linalg.cond

CHANGELOG.md

@@ -23,6 +23,7 @@ This release is compatible with NumPy 2.5.
 * Added the `ndmax` keyword to `dpnp.array` for compatibility with NumPy [#3044](https://github.com/IntelPython/dpnp/pull/3044)
 * Added `UsmNDArray_RemoveQueueRef` C API function to release a queue reference obtained from `UsmNDArray_GetQueueRef` [#3042](https://github.com/IntelPython/dpnp/pull/3042)
 * Added a `rattler-build`-compatible conda recipe (`conda-recipe/rattler_recipe.yaml`) alongside the existing `conda-build` recipe [#3031](https://github.com/IntelPython/dpnp/pull/3031)
+* Added support for the `"nuc"` (nuclear) norm in `dpnp.linalg.cond` [#3084](https://github.com/IntelPython/dpnp/pull/3084)
 
 ### Changed
 

dpnp/linalg/dpnp_iface_linalg.py

@@ -171,14 +171,15 @@ def cond(x, p=None):
     ----------
     x : {dpnp.ndarray, usm_ndarray}
         The matrix whose condition number is sought.
-    p : {None, 1, -1, 2, -2, inf, -inf, "fro"}, optional
+    p : {None, 1, -1, 2, -2, inf, -inf, "fro", "nuc"}, optional
         Order of the norm used in the condition number computation:
 
         =====  ============================
         p      norm for matrices
         =====  ============================
         None   2-norm
         'fro'  Frobenius norm
+        'nuc'  nuclear norm
         inf    max(sum(abs(x), axis=1))
         -inf   min(sum(abs(x), axis=1))
         1      max(sum(abs(x), axis=0))
@@ -188,7 +189,8 @@ def cond(x, p=None):
         =====  ============================
 
         ``inf`` means the :obj:`dpnp.inf` object, and the Frobenius norm is
-        the root-of-sum-of-squares norm.
+        the root-of-sum-of-squares norm. The nuclear norm is the sum of the
+        singular values.
 
         Default: ``None``.
 
@@ -204,9 +206,9 @@ def cond(x, p=None):
     Notes
     -----
     This function will raise :class:`dpnp.linalg.LinAlgError` on singular input
-    when using any of the norm: ``1``, ``-1``, ``inf``, ``-inf``, or ``'fro'``.
-    In contrast, :obj:`numpy.linalg.cond` will fill the result array with
-    ``inf`` values for each 2D batch in the input array that is singular
+    when using any of the norm: ``1``, ``-1``, ``inf``, ``-inf``, ``'fro'``, or
+    ``'nuc'``. In contrast, :obj:`numpy.linalg.cond` will fill the result array
+    with ``inf`` values for each 2D batch in the input array that is singular
     when using these norms.
 
     Examples
@@ -221,6 +223,8 @@ def cond(x, p=None):
     array(1.41421356)
     >>> np.linalg.cond(a, 'fro')
     array(3.16227766)
+    >>> np.linalg.cond(a, 'nuc')
+    array(9.24264069)
     >>> np.linalg.cond(a, np.inf)
     array(2.)
     >>> np.linalg.cond(a, -np.inf)

dpnp/tests/test_linalg.py

@@ -278,7 +278,7 @@ def test_cholesky_errors(self):
 
 
 class TestCond:
-    _norms = [None, -dpnp.inf, -2, -1, 1, 2, dpnp.inf, "fro"]
+    _norms = [None, -dpnp.inf, -2, -1, 1, 2, dpnp.inf, "fro", "nuc"]
 
     @pytest.mark.parametrize(
         "shape", [(0, 4, 4), (4, 0, 3, 3)], ids=["(0, 4, 4)", "(4, 0, 3, 3)"]
@@ -325,7 +325,7 @@ def test_nan_to_inf(self, p):
         # NumPy does not raise LinAlgError on singular matrices.
         # It returns `inf`, `0`, or large/small finite values
         # depending on the norm and the matrix content.
-        # DPNP raises LinAlgError for 1, -1, inf, -inf, and 'fro'
+        # DPNP raises LinAlgError for 1, -1, inf, -inf, 'fro', and 'nuc'
         # due to use of gesv in the 2D case.
         # For [None, 2, -2], DPNP does not raise.
         if p in [None, 2, -2]:

dpnp/tests/test_sycl_queue.py

@@ -1619,7 +1619,7 @@ def test_cholesky(self, data, is_empty, device):
         assert_sycl_queue_equal(result.sycl_queue, x.sycl_queue)
 
     @pytest.mark.parametrize(
-        "p", [None, -dpnp.inf, -2, -1, 1, 2, dpnp.inf, "fro"]
+        "p", [None, -dpnp.inf, -2, -1, 1, 2, dpnp.inf, "fro", "nuc"]
     )
     def test_cond(self, device, p):
         a = generate_random_numpy_array((2, 4, 4))

dpnp/tests/test_usm_type.py

@@ -1447,7 +1447,7 @@ def test_cholesky(self, data, is_empty, usm_type):
         assert x.usm_type == result.usm_type
 
     @pytest.mark.parametrize(
-        "p", [None, -dpnp.inf, -2, -1, 1, 2, dpnp.inf, "fro"]
+        "p", [None, -dpnp.inf, -2, -1, 1, 2, dpnp.inf, "fro", "nuc"]
     )
     def test_cond(self, usm_type, p):
         a = generate_random_numpy_array((2, 4, 4), seed_value=42)