fix: frontend torch linalg cholesky_ex (#28773)

Co-authored-by: Jin Wang <[email protected]>

ivy/functional/frontends/torch/linalg.py

@@ -16,18 +16,23 @@ def cholesky(input, *, upper=False, out=None):
 
 
 @to_ivy_arrays_and_back
+@with_supported_dtypes(
+    {"2.2 and below": ("float32", "float64", "complex32", "complex64")}, "torch"
+)
 def cholesky_ex(input, *, upper=False, check_errors=False, out=None):
     try:
+        results = namedtuple("cholesky_ex", ['L', 'info'])
         matrix = ivy.cholesky(input, upper=upper, out=out)
         info = ivy.zeros(input.shape[:-2], dtype=ivy.int32)
-        return matrix, info
+        return results(matrix, info)
     except RuntimeError as e:
         if check_errors:
             raise RuntimeError(e) from e
         else:
+            results = namedtuple("cholesky_ex", ['L', 'info'])
             matrix = input * math.nan
             info = ivy.ones(input.shape[:-2], dtype=ivy.int32)
-            return matrix, info
+            return results(matrix, info)
 
 
 @to_ivy_arrays_and_back

ivy_tests/test_ivy/test_frontends/test_torch/test_linalg.py

@@ -336,7 +336,7 @@ def test_torch_cholesky(
 
 @handle_frontend_test(
     fn_tree="torch.linalg.cholesky_ex",
-    dtype_and_x=_get_dtype_and_matrix(square=True, batch=True),
+    dtype_and_x=_get_dtype_and_matrix(square=True),
     upper=st.booleans(),
 )
 def test_torch_cholesky_ex(
@@ -350,8 +350,9 @@ def test_torch_cholesky_ex(
     backend_fw,
 ):
     dtype, x = dtype_and_x
-    x = np.matmul(x.T, x) + np.identity(x.shape[0])  # make symmetric positive-definite
-
+    x = np.asarray(x[0], dtype=dtype[0])
+    x = np.matmul(np.conjugate(x.T), x) + np.identity(x.shape[0], dtype=dtype[0])
+    # make symmetric positive-definite
     helpers.test_frontend_function(
         input_dtypes=dtype,
         backend_to_test=backend_fw,