[Auto Parallel] Clear some fluid APIs (#49793)

* [Auto Parallel] Rename methods of ProcessMesh

* [Auto Parallel] Impl the python process_mesh by the c++ one

* [Auto Parallel] Add some minor modifications

* [Auto Parallel] Rename some methods

* [Auto Parallel] Remove unnecessary codes

* [Auto Parallel] Add back some removed files

* [Auto Parallel] Fix bugs

* [Auto Parallel] Fix a bug

* Update process_mesh.cc

* [Auto Parallel] Merge dist attrs of Python into C++

* [Auto Parallel] Add back deleted importing

* [Auto Parallel] Add back removed unittest

* [Auto Parallel] Remove type qualifiers of return types

* [Auto Parallel] Fix some bugs

* [Auto Parallel] Fix a bug of the quant pass

* [Auto Parallel] Fix the code style

* [Auto Parallel] Clear some fluid APIs

python/paddle/distributed/auto_parallel/interface.py

@@ -256,7 +256,7 @@ def add_to_collection(collection_name, value, name=None):
 
 
 def fetch(tensor, name=None, logging=False):
-    if isinstance(tensor, paddle.fluid.framework.Variable):
+    if isinstance(tensor, paddle.static.Variable):
         tensor = tensor.name
     elif isinstance(tensor, str):
         tensor = tensor

python/paddle/distributed/auto_parallel/operators/dist_embedding.py

@@ -12,12 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License
 
+from paddle.common_ops_import import check_dtype, check_variable_and_dtype
 from paddle.distributed.auto_parallel.cost.comm_op_cost import (
     AllreduceSumOpCost,
     IdentityOpCost,
 )
 from paddle.distributed.fleet.meta_optimizers.common import OP_ROLE_KEY, OpRole
-from paddle.fluid.data_feeder import check_dtype, check_variable_and_dtype
 from paddle.framework import core
 from paddle.utils import unique_name
 

python/paddle/distributed/auto_parallel/operators/dist_matmul.py

@@ -14,12 +14,12 @@
 
 import copy
 
+from paddle.common_ops_import import check_dtype, check_variable_and_dtype
 from paddle.distributed.auto_parallel.cost.comm_op_cost import (
     AllreduceSumOpCost,
     IdentityOpCost,
 )
 from paddle.distributed.fleet.meta_optimizers.common import OP_ROLE_KEY, OpRole
-from paddle.fluid.data_feeder import check_dtype, check_variable_and_dtype
 from paddle.framework import core
 from paddle.utils import unique_name
 

python/paddle/distributed/auto_parallel/operators/dist_pnorm.py

@@ -14,7 +14,7 @@
 
 import copy
 
-from paddle.fluid.data_feeder import check_dtype, check_variable_and_dtype
+from paddle.common_ops_import import check_dtype, check_variable_and_dtype
 from paddle.framework import core
 from paddle.static import Operator
 

python/paddle/distributed/auto_parallel/process_mesh.py

@@ -17,7 +17,7 @@ import copy
 import numpy as np
 
 import paddle
-from paddle.fluid import core
+from paddle.framework import core
 
 # Use to store the previous and current process mesh
 _g_previous_process_mesh = None

python/paddle/distributed/passes/auto_parallel_bf16.py

@@ -25,9 +25,7 @@ from paddle.distributed.auto_parallel.utils import (
 )
 from paddle.distributed.fleet.meta_optimizers.common import OpRole
 from paddle.distributed.passes.pass_base import PassBase, register_pass
-from paddle.fluid import unique_name
-from paddle.fluid.framework import Block
-from paddle.framework import core
+from paddle.framework import Block, core
 from paddle.static.amp.bf16 import AutoMixedPrecisionListsBF16
 from paddle.static.amp.bf16.amp_utils import (
     _dtype_to_str,
@@ -40,6 +38,7 @@ from paddle.static.amp.fp16_utils import (
     find_op_index,
     find_true_prev_op,
 )
+from paddle.utils import unique_name
 
 from ..auto_parallel.utils import is_backward_op, is_forward_op, is_loss_op
 
@@ -499,7 +498,7 @@ def _update_backward_cast_ops(params_grads, dist_context):
             # add new op in the python and cpp at the same time
             new_op_desc = main_block.desc.append_op()
             new_op_desc.copy_from(op.desc)
-            new_op = paddle.fluid.framework.Operator(
+            new_op = paddle.static.Operator(
                 block=main_block,
                 desc=new_op_desc,
                 type=None,

python/paddle/distributed/passes/auto_parallel_fp16.py

@@ -15,6 +15,7 @@
 from collections import defaultdict
 
 import paddle
+from paddle.common_ops_import import check_type, check_variable_and_dtype
 from paddle.distributed.auto_parallel.dist_attribute import OperatorDistAttr
 from paddle.distributed.auto_parallel.process_group import (
     get_world_process_group,
@@ -26,7 +27,6 @@ from paddle.distributed.auto_parallel.utils import (
     set_var_dist_attr,
 )
 from paddle.distributed.fleet.meta_optimizers.common import OP_ROLE_KEY, OpRole
-from paddle.fluid.data_feeder import check_type, check_variable_and_dtype
 from paddle.framework import core
 from paddle.static import default_main_program, default_startup_program
 from paddle.static.amp.fp16_utils import (