diff --git a/python/paddle/audio/functional/window.py b/python/paddle/audio/functional/window.py
index 4836afbb61d03f81422b8102efbd63a00b7b794e..52ca3e477324c33c9f98073f4d1ae503c8176f25 100644
--- a/python/paddle/audio/functional/window.py
+++ b/python/paddle/audio/functional/window.py
@@ -210,7 +210,7 @@ def _tukey(
     if alpha <= 0:
         return paddle.ones((M,), dtype=dtype)
     elif alpha >= 1.0:
-        return hann(M, sym=sym)
+        return _hann(M, sym=sym)
 
     M, needs_trunc = _extend(M, sym)
 
diff --git a/python/paddle/distributed/auto_parallel/cost/base_cost.py b/python/paddle/distributed/auto_parallel/cost/base_cost.py
index 249f64b7a09c2a6a58e5afd04ed6090f2a03eb7c..f9dc6b6fc273217c99f07a3e78fe0212e89beef7 100644
--- a/python/paddle/distributed/auto_parallel/cost/base_cost.py
+++ b/python/paddle/distributed/auto_parallel/cost/base_cost.py
@@ -833,7 +833,7 @@ class CommOpCost(OpCost):
             if self.op_desc is not None:
                 self._group_ranks = self.op_desc["group_ranks"]
             elif self.op is not None:
-                ring_id = op.attrs("ring_id")
+                ring_id = self.op.attrs("ring_id")
                 process_group = get_process_group(ring_id)
                 if process_group is None:
                     raise ValueError(
diff --git a/python/paddle/distributed/auto_parallel/operators/dist_reshape.py b/python/paddle/distributed/auto_parallel/operators/dist_reshape.py
index dcb85bf5966b412e82704596a026ee0db67827a6..b305d88d7df1b270b352b74c1d58198c2f47d25b 100644
--- a/python/paddle/distributed/auto_parallel/operators/dist_reshape.py
+++ b/python/paddle/distributed/auto_parallel/operators/dist_reshape.py
@@ -23,6 +23,7 @@ from .dist_default import DistributedDefaultImpl0
 from ..cost import build_comp_desc_from_dist_op, build_comp_costs_from_descs
 from ..cost import Reshape2OpCost
 from ..cost import Reshape2GradOpCost
+from ..cost import build_dp_costs
 from paddle.distributed.fleet.meta_optimizers.common import OpRole
 
 
diff --git a/python/paddle/distributed/auto_parallel/planner.py b/python/paddle/distributed/auto_parallel/planner.py
index 8148392f0adb628e5acdae191c799a7a5ddcfb7c..15bf8058f739614335c966b92c00026da2d913bd 100755
--- a/python/paddle/distributed/auto_parallel/planner.py
+++ b/python/paddle/distributed/auto_parallel/planner.py
@@ -437,7 +437,7 @@ class SearchAlgorithm:
 
     @property
     def name(self):
-        self.name = name
+        self.name = self._name
 
     def search(self):
         raise NotImplementedError("Please Implement this method in subclass.")
diff --git a/python/paddle/distributed/fleet/base/orthogonal_strategy.py b/python/paddle/distributed/fleet/base/orthogonal_strategy.py
index d0fec2cfdb2d574a45b2e2c6c72118892f558b60..36af85d415d556709ca50a4be09f3d1d47cede47 100644
--- a/python/paddle/distributed/fleet/base/orthogonal_strategy.py
+++ b/python/paddle/distributed/fleet/base/orthogonal_strategy.py
@@ -129,7 +129,9 @@ class OrthogonalStrategy:
     def _check_valid_strategy(self):
         assert len(self._list_of_strategy_name) == len(
             set(self._list_of_strategy_name)
-        ), "Defined duplicated strategies: {}".format(list_of_strategy)
+        ), "Defined duplicated strategies: {}".format(
+            self._list_of_strategy_name
+        )
         num_of_ranks = functools.reduce(
             lambda x, y: x * y, self._list_of_degree
         )
diff --git a/python/paddle/distributed/fleet/data_generator/data_generator.py b/python/paddle/distributed/fleet/data_generator/data_generator.py
index abf8f5f49fa1f7b911a7d377627dda0b06b2ceeb..297a2cf003ecb4e7d6a5609a6d152475504b6c98 100644
--- a/python/paddle/distributed/fleet/data_generator/data_generator.py
+++ b/python/paddle/distributed/fleet/data_generator/data_generator.py
@@ -342,9 +342,7 @@ class MultiSlotDataGenerator(DataGenerator):
                 for elem in elements:
                     if isinstance(elem, float):
                         self._proto_info[-1] = (name, "float")
-                    elif not isinstance(elem, int) and not isinstance(
-                        elem, long
-                    ):
+                    elif not isinstance(elem, int):
                         raise ValueError(
                             "the type of element%s must be in int or float"
                             % type(elem)
@@ -379,9 +377,7 @@ class MultiSlotDataGenerator(DataGenerator):
                     if self._proto_info[index][1] != "float":
                         if isinstance(elem, float):
                             self._proto_info[index] = (name, "float")
-                        elif not isinstance(elem, int) and not isinstance(
-                            elem, long
-                        ):
+                        elif not isinstance(elem, int):
                             raise ValueError(
                                 "the type of element%s must be in int or float"
                                 % type(elem)
diff --git a/python/paddle/distributed/fleet/layers/mpu/mp_ops.py b/python/paddle/distributed/fleet/layers/mpu/mp_ops.py
index 07fc4e7172b9f00ded095df45a5bb47c171eebe4..83ba760c9e0a788a51133c49f7004f99053bc7d1 100644
--- a/python/paddle/distributed/fleet/layers/mpu/mp_ops.py
+++ b/python/paddle/distributed/fleet/layers/mpu/mp_ops.py
@@ -18,6 +18,7 @@ from paddle.fluid import core
 from paddle.fluid.framework import _non_static_mode
 from paddle.fluid.framework import _in_legacy_dygraph
 from paddle.fluid.framework import in_dygraph_mode
+from paddle.fluid.framework import _varbase_creator
 from paddle.fluid.layer_helper import LayerHelper
 from paddle.fluid.data_feeder import check_variable_and_dtype
 from paddle.fluid.dygraph import layers
diff --git a/python/paddle/distributed/fleet/runtime/the_one_ps.py b/python/paddle/distributed/fleet/runtime/the_one_ps.py
index 51765845507a62688f511c1d1fe3130546af3b21..98e06299789135fac41b9868e18febf146ea17b5 100644
--- a/python/paddle/distributed/fleet/runtime/the_one_ps.py
+++ b/python/paddle/distributed/fleet/runtime/the_one_ps.py
@@ -902,7 +902,9 @@ class TheOnePSRuntime(RuntimeBase):
                 heter_device_type = self.role_maker._heter_device_type().upper()
                 if heter_device_type not in ["GPU", "XPU", "CPU"]:
                     raise ValueError(
-                        "Heter Worker Not Support Device {}".format(device_type)
+                        "Heter Worker Not Support Device {}".format(
+                            heter_device_type
+                        )
                     )
                 if heter_device_type == "GPU":
                     executor = Executor(
diff --git a/python/paddle/distributed/fleet/utils/fs.py b/python/paddle/distributed/fleet/utils/fs.py
index 667752e668a96f753a81013128344b77e101cbb2..fd2338e36cc25aa0d67348157a748094628dfaa6 100644
--- a/python/paddle/distributed/fleet/utils/fs.py
+++ b/python/paddle/distributed/fleet/utils/fs.py
@@ -19,6 +19,7 @@ import re
 import time
 import abc
 from paddle.fluid import core
+from .log_util import logger
 import functools
 
 import shutil
diff --git a/python/paddle/distributed/launch/job/pod.py b/python/paddle/distributed/launch/job/pod.py
index a322bcdccfe6d7428d1719daebde031e3850183f..ef72263dd8758dda2b4b6ee6467982be42a4bc56 100644
--- a/python/paddle/distributed/launch/job/pod.py
+++ b/python/paddle/distributed/launch/job/pod.py
@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import annotations
+
 from .container import Container
 
 from .status import Status
@@ -27,8 +29,8 @@ class PodSepc:
         )
 
         # by controller
-        self._init_containers: List[Container] = []
-        self._containers: List[Container] = []
+        self._init_containers: list[Container] = []
+        self._containers: list[Container] = []
 
         # self.resource: Resource = None
         # self.status: Status = None
diff --git a/python/paddle/distributed/launch/utils/kv_client.py b/python/paddle/distributed/launch/utils/kv_client.py
index b60970382aee807bb5a26dab1f762e13bd9c0a53..e0482203357c72ff5d40d6024c4a44ecd7ad182b 100644
--- a/python/paddle/distributed/launch/utils/kv_client.py
+++ b/python/paddle/distributed/launch/utils/kv_client.py
@@ -79,7 +79,7 @@ class KVClient:
 
 
 if __name__ == '__main__':
-    cli = PKVClient("http://localhost:8090")
+    cli = KVClient("http://localhost:8090")
     data = {"/workers/1": "rank1", "/workers/2": "rank2"}
     for k, v in data.items():
         cli.put(k, v)
diff --git a/python/paddle/distributed/passes/auto_parallel_data_parallel_optimization.py b/python/paddle/distributed/passes/auto_parallel_data_parallel_optimization.py
index cbc9170a1e49c089c24d019d24fef0aebe5090be..601cd31948b3fc754ed6417fce9969f22f1212b6 100644
--- a/python/paddle/distributed/passes/auto_parallel_data_parallel_optimization.py
+++ b/python/paddle/distributed/passes/auto_parallel_data_parallel_optimization.py
@@ -494,7 +494,9 @@ class DataParallelOptimizationPass(PassBase):
             for idx in sorted(remove_op_indices, reverse=True):
                 assert (
                     block.ops[idx].type in remove_op_types
-                ), "Unexception: try to remove op {}".format(str(op))
+                ), "Unexception: try to remove op {}".format(
+                    str(block.ops[idx])
+                )
                 block._remove_op(idx)
 
             # insert coalecse op
diff --git a/python/paddle/distributed/ps/utils/public.py b/python/paddle/distributed/ps/utils/public.py
index 522bc5f36c16361845b5f68f608cafddbfad06ea..b15edf62d1e553dc312f8bd4912eb944632e6f00 100755
--- a/python/paddle/distributed/ps/utils/public.py
+++ b/python/paddle/distributed/ps/utils/public.py
@@ -482,7 +482,7 @@ def get_geo_trainer_send_context(attrs):
     if attrs['ps_mode'] != DistributedMode.GEO:
         raise ValueError(
             "ps mode: {} not matched {}",
-            format(ps_mode, "get_geo_trainer_send_context"),
+            format(attrs['ps_mode'], "get_geo_trainer_send_context"),
         )
     send_ctx = {}
     trainer_id = get_role_id(attrs['role_maker'])
diff --git a/python/paddle/distribution/variable.py b/python/paddle/distribution/variable.py
index 99cafc5ea788ec29a3d09ba0d7ffba8bc89fe2e7..58d4a7521542eaf5c65cba355bab530ece0c54c6 100644
--- a/python/paddle/distribution/variable.py
+++ b/python/paddle/distribution/variable.py
@@ -14,6 +14,8 @@
 
 from paddle.distribution import constraint
 
+import paddle
+
 
 class Variable:
     """Random variable of probability distribution.
diff --git a/python/paddle/fluid/tests/unittests/ir/inference/test_trt_convert_equal.py b/python/paddle/fluid/tests/unittests/ir/inference/test_trt_convert_equal.py
index ee73586e6e96a4a228af351b3867f789bee34770..ab873b4b3f09e7487ea97d93c9d81017706bfac1 100644
--- a/python/paddle/fluid/tests/unittests/ir/inference/test_trt_convert_equal.py
+++ b/python/paddle/fluid/tests/unittests/ir/inference/test_trt_convert_equal.py
@@ -27,7 +27,7 @@ class TrtConvertElementwiseTest_one_input_corner_case(TrtLayerAutoScanTest):
             program_config.ops[i].attrs for i in range(len(program_config.ops))
         ]
         if attrs[0]['axis'] == 0:
-            return false
+            return False
         ver = paddle_infer.get_trt_compile_version()
         if ver[0] * 1000 + ver[1] * 100 + ver[2] * 10 < 8415:
             return False
diff --git a/python/paddle/fluid/tests/unittests/test_dropout_nd_op.py b/python/paddle/fluid/tests/unittests/test_dropout_nd_op.py
index a21ec9625f927873a5cfe6c3120288b37642ffb4..e19a0104de38d9d339874f2136808b0000612705 100644
--- a/python/paddle/fluid/tests/unittests/test_dropout_nd_op.py
+++ b/python/paddle/fluid/tests/unittests/test_dropout_nd_op.py
@@ -22,6 +22,7 @@ from paddle.fluid.layer_helper import LayerHelper
 from paddle.fluid.framework import _non_static_mode
 from paddle import _legacy_C_ops
 from paddle.static import default_main_program
+from paddle.fluid.data_feeder import check_variable_and_dtype
 
 
 def dropout_nd(
diff --git a/python/paddle/fluid/tests/unittests/test_mean_op.py b/python/paddle/fluid/tests/unittests/test_mean_op.py
index 0c52d7596c129e27864d9630ccb19983c82974e8..fb52745c7593da5bec3f26ffed56083f8e01f749 100644
--- a/python/paddle/fluid/tests/unittests/test_mean_op.py
+++ b/python/paddle/fluid/tests/unittests/test_mean_op.py
@@ -140,7 +140,7 @@ def ref_reduce_mean(x, axis=None, keepdim=False, reduce_all=False):
     return np.mean(x, axis=axis, keepdims=keepdim)
 
 
-def ref_reduce_mean_grad(x, axis, dtype):
+def ref_reduce_mean_grad(x, axis, dtype, reduce_all):
     if reduce_all:
         axis = list(range(x.ndim))
 
@@ -191,7 +191,6 @@ class TestReduceMeanOp(OpTest):
         if self.dtype != 'float16':
             self.check_grad(['X'], ['Out'], check_eager=True)
         else:
-            return
             if not core.is_compiled_with_cuda():
                 return
             place = paddle.CUDAPlace(0)
@@ -204,7 +203,10 @@ class TestReduceMeanOp(OpTest):
                 )
                 dx = paddle.grad(y, x)[0].numpy()
                 dx_expected = ref_reduce_mean_grad(
-                    self.inputs['X'], self.attrs['dim'], self.dtype
+                    self.inputs['X'],
+                    self.attrs['dim'],
+                    self.dtype,
+                    self.attrs['reduce_all'],
                 )
                 np.testing.assert_array_equal(dx, dx_expected)