fix issues

python/cudf/cudf/testing/dataset_generator.py

@@ -377,7 +377,6 @@ def rand_dataframe(
                     ),
                     is_sorted=False,
                     dtype=dtype,
-                    rng=rng,
                 )
             )
         elif dtype == "struct":
@@ -794,6 +793,7 @@ def get_nested_lists(dtype, size, nesting_depth, lists_max_length, rng):
                 lists_max_length=lists_max_length,
                 nesting_depth=nesting_depth,
                 top_level_list=[],
+                rng=rng,
             )
         )
 
@@ -870,7 +870,7 @@ def create_nested_struct_type(max_types_at_each_level, nesting_level, rng):
 
 
 def _generate_string(str_seq: str, rng, length: int = 10) -> str:
-    return "".join(rng.choices(str_seq, k=length))
+    return "".join(rng.choice(list(str_seq), size=length))
 
 
 def _unique_string() -> str:

python/cudf/cudf/tests/test_factorize.py

@@ -19,7 +19,10 @@ def test_factorize_series_obj(ncats, nelem):
     df["cats"] = arr = rng.integers(2, size=10, dtype=np.int32)
 
     uvals, labels = df["cats"].factorize()
-    np.testing.assert_array_equal(labels.to_numpy(), sorted(set(arr)))
+    unique_values, indices = np.unique(arr, return_index=True)
+    expected_values = unique_values[np.argsort(indices)]
+
+    np.testing.assert_array_equal(labels.to_numpy(), expected_values)
     assert isinstance(uvals, cp.ndarray)
     assert isinstance(labels, Index)
 
@@ -38,7 +41,10 @@ def test_factorize_index_obj(ncats, nelem):
     df = df.set_index("cats")
 
     uvals, labels = df.index.factorize()
-    np.testing.assert_array_equal(labels.values.get(), sorted(set(arr)))
+    unique_values, indices = np.unique(arr, return_index=True)
+    expected_values = unique_values[np.argsort(indices)]
+
+    np.testing.assert_array_equal(labels.values.get(), expected_values)
     assert isinstance(uvals, cp.ndarray)
     assert isinstance(labels, Index)