diff --git a/mindspore/ccsrc/CMakeLists.txt b/mindspore/ccsrc/CMakeLists.txt
index 6eba48d1c7..1c94397526 100644
--- a/mindspore/ccsrc/CMakeLists.txt
+++ b/mindspore/ccsrc/CMakeLists.txt
@@ -363,7 +363,7 @@ set_target_properties(_c_expression PROPERTIES INSTALL_RPATH ${MINDSPORE_RPATH})
 if(CMAKE_SYSTEM_NAME MATCHES "Windows")
     target_link_libraries(mindspore mindspore::pybind11_module)
     target_link_libraries(mindspore mindspore_gvar)
-    target_link_libraries(_c_expression PRIVATE -Wl,--whole-archive mindspore -Wl,--no-whole-archive)
+    target_link_libraries(_c_expression PRIVATE -Wl,--whole-archive mindspore mindspore_core -Wl,--no-whole-archive)
 elseif(CMAKE_SYSTEM_NAME MATCHES "Darwin")
     target_link_libraries(mindspore mindspore::pybind11_module)
     target_link_libraries(mindspore mindspore_gvar)
diff --git a/mindspore/ccsrc/backend/optimizer/ascend/ir_fission/dynamic_rnn_grad_fission_v2.cc b/mindspore/ccsrc/backend/optimizer/ascend/ir_fission/dynamic_rnn_grad_fission_v2.cc
index 3ad53404b9..db2f833708 100644
--- a/mindspore/ccsrc/backend/optimizer/ascend/ir_fission/dynamic_rnn_grad_fission_v2.cc
+++ b/mindspore/ccsrc/backend/optimizer/ascend/ir_fission/dynamic_rnn_grad_fission_v2.cc
@@ -459,7 +459,7 @@ AnfNodePtr CreateValueNode(const FuncGraphPtr &func_graph, const CNodePtr &dynam
   std::vector<size_t> shape = {t_size, IntToSize(1), n_size};
   std::vector<int64_t> output_shape = {SizeToLong(t_size), SizeToLong(1), SizeToLong(n_size)};
   std::vector<int64_t> output_tensor = {SizeToLong(t_size) * SizeToLong(n_size)};
-  auto tensor = TensorConstructUtils::CreateOnesTensor(kNumberTypeFloat32, output_tensor);
+  auto tensor = TensorConstructUtils::CreateOnesTensor(kFloat32, output_tensor);
   auto x_abstract = std::make_shared<abstract::AbstractTensor>(kFloat32, output_shape);
   auto kernel_graph = func_graph->cast<KernelGraphPtr>();
   auto value_node = kernel_graph->NewValueNode(x_abstract, tensor);
diff --git a/mindspore/core/base/core_ops.h b/mindspore/core/base/core_ops.h
index 181195b06b..dd6b225f1f 100644
--- a/mindspore/core/base/core_ops.h
+++ b/mindspore/core/base/core_ops.h
@@ -287,6 +287,7 @@ inline const PrimitivePtr kPrimElu = std::make_shared<Primitive>("Elu");
 inline const PrimitivePtr kPrimRelu6 = std::make_shared<Primitive>("ReLU6");
 inline const PrimitivePtr kPrimReluV2 = std::make_shared<Primitive>("ReLUV2");
 inline const PrimitivePtr kPrimPRelu = std::make_shared<Primitive>("PReLU");
+inline const PrimitivePtr kPrimZeros = std::make_shared<Primitive>("Zeros");
 inline const PrimitivePtr kPrimZerosLike = std::make_shared<Primitive>("ZerosLike");
 inline const PrimitivePtr kPrimOnesLike = std::make_shared<Primitive>("OnesLike");
 inline const PrimitivePtr kPrimBpropCut = std::make_shared<Primitive>("bprop_cut");
diff --git a/mindspore/core/ops/bias_add.cc b/mindspore/core/ops/bias_add.cc
index b6660b28e0..d1bb713b27 100644
--- a/mindspore/core/ops/bias_add.cc
+++ b/mindspore/core/ops/bias_add.cc
@@ -29,7 +29,7 @@ abstract::ShapePtr BiasAddInferShape(const PrimitivePtr &primitive, const std::v
   MS_EXCEPTION_IF_NULL(primitive);
   auto prim_name = primitive->name();
   // check
-  CheckAndConvertUtils::CheckInteger("biasadd_infer", input_args.size(), kEqual, 2, prim_name);
+  CheckAndConvertUtils::CheckInteger("arg size", input_args.size(), kEqual, 2, prim_name);
   auto x_shape = CheckAndConvertUtils::ConvertShapePtrToShape("x_shape", input_args[0]->BuildShape(), prim_name);
   auto b_shape = CheckAndConvertUtils::ConvertShapePtrToShape("b_shape", input_args[1]->BuildShape(), prim_name);
   CheckAndConvertUtils::CheckInteger("x rank", x_shape.size(), kGreaterEqual, 2, prim_name);
@@ -55,7 +55,7 @@ TypePtr BiasAddInferType(const PrimitivePtr &prim, const std::vector<AbstractBas
   std::map<std::string, TypePtr> types;
   types.emplace("input_x", input_args[0]->BuildType());
   types.emplace("bias", input_args[1]->BuildType());
-  return CheckAndConvertUtils::CheckTensorTypeSame(types, common_valid_types, prim->name());
+  return CheckAndConvertUtils::CheckTensorTypeSame(types, common_valid_types, prim_name);
 }
 }  // namespace
 void BiasAdd::set_format(const Format &format) {
diff --git a/mindspore/core/ops/gather.cc b/mindspore/core/ops/gather.cc
index 31ec4947b5..47a274ebc6 100644
--- a/mindspore/core/ops/gather.cc
+++ b/mindspore/core/ops/gather.cc
@@ -1,5 +1,5 @@
 /**
- * Copyright 2020 Huawei Technologies Co., Ltd
+ * Copyright 2021 Huawei Technologies Co., Ltd
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -20,28 +20,6 @@
 
 namespace mindspore {
 namespace ops {
-AbstractBasePtr GatherInfer(const abstract::AnalysisEnginePtr &, const PrimitivePtr &primitive,
-                            const std::vector<AbstractBasePtr> &input_args) {
-  MS_EXCEPTION_IF_NULL(primitive);
-  auto prim_name = primitive->name();
-  CheckAndConvertUtils::CheckInteger("gather_infer", input_args.size(), kEqual, 3, prim_name);
-
-  // Infer type
-  std::set<TypePtr> valid_x_type = {kTensorType};
-  auto x_type =
-    CheckAndConvertUtils::CheckTensorTypeValid("x_type", input_args[0]->BuildType(), valid_x_type, prim_name);
-  std::set<TypePtr> valid_index_types = {kInt32, kInt64};
-  CheckAndConvertUtils::CheckTensorTypeValid("index_type", input_args[2]->BuildType(), valid_index_types, prim_name);
-  std::set<TypePtr> valid_dim_type = {kInt32, kInt64};
-  CheckAndConvertUtils::CheckSubClass("dim_type", input_args[1]->BuildType(), valid_dim_type, prim_name);
-
-  // Infer shape
-  auto x_shape = CheckAndConvertUtils::ConvertShapePtrToShape("x_shape", input_args[0]->BuildShape(), prim_name);
-  auto index_shape = CheckAndConvertUtils::ConvertShapePtrToShape("dim_shape", input_args[2]->BuildShape(), prim_name);
-  CheckAndConvertUtils::Check("x_rank", x_shape.size(), kEqual, "index_rank", index_shape.size(), prim_name);
-
-  return std::make_shared<abstract::AbstractTensor>(x_type, index_shape);
-}
 REGISTER_PRIMITIVE_C(kNameGather, Gather);
 }  // namespace ops
 }  // namespace mindspore
diff --git a/mindspore/core/ops/gather.h b/mindspore/core/ops/gather.h
index aa1e88a4f9..55c735f77c 100644
--- a/mindspore/core/ops/gather.h
+++ b/mindspore/core/ops/gather.h
@@ -1,5 +1,5 @@
 /**
- * Copyright 2020 Huawei Technologies Co., Ltd
+ * Copyright 2021 Huawei Technologies Co., Ltd
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -34,8 +34,6 @@ class Gather : public PrimitiveC {
   MS_DECLARE_PARENT(Gather, PrimitiveC);
   void Init() {}
 };
-AbstractBasePtr GatherInfer(const abstract::AnalysisEnginePtr &, const PrimitivePtr &primitive,
-                            const std::vector<AbstractBasePtr> &input_args);
 using PrimGatherPtr = std::shared_ptr<Gather>;
 }  // namespace ops
 }  // namespace mindspore
diff --git a/mindspore/core/ops/gather_d.cc b/mindspore/core/ops/gather_d.cc
new file mode 100644
index 0000000000..b195977a9c
--- /dev/null
+++ b/mindspore/core/ops/gather_d.cc
@@ -0,0 +1,75 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "ops/gather_d.h"
+#include <memory>
+#include <set>
+#include "ops/op_utils.h"
+#include "utils/check_convert_utils.h"
+#include "abstract/primitive_infer_map.h"
+
+namespace mindspore {
+namespace ops {
+// gather_d
+namespace {
+abstract::ShapePtr GatherDInferShape(const PrimitivePtr &primitive, const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(primitive);
+  auto prim_name = primitive->name();
+  // check
+  auto x_shape = CheckAndConvertUtils::ConvertShapePtrToShape("x_shape", input_args[0]->BuildShape(), prim_name);
+  auto index_shape = CheckAndConvertUtils::ConvertShapePtrToShape("dim_shape", input_args[2]->BuildShape(), prim_name);
+  int64_t x_rank = x_shape.size();
+  CheckAndConvertUtils::Check("x_rank", x_rank, kEqual, "index_rank", index_shape.size(), prim_name);
+  auto dim_v = GetValue<int64_t>(input_args[1]->BuildValue());
+  CheckAndConvertUtils::Check("dim value", dim_v, kGreaterEqual, "negative index_rank", -x_rank, prim_name);
+  CheckAndConvertUtils::Check("dim value", dim_v, kLessThan, "index_rank", x_rank, prim_name);
+
+  if (dim_v < 0) {
+    dim_v = dim_v + x_rank;
+  }
+  for (int i = 0; i < x_rank; ++i) {
+    if (i == dim_v) continue;
+    MS_LOG(INFO) << "Check " << i << "th x shape";
+    CheckAndConvertUtils::Check("x shape", x_shape[i], kEqual, "index_rank", index_shape[i], prim_name);
+  }
+  return std::make_shared<abstract::Shape>(index_shape);
+}
+
+TypePtr GatherDInferType(const PrimitivePtr &prim, const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(prim);
+  auto prim_name = prim->name();
+  // check
+  std::set<TypePtr> valid_x_type = {kTensorType};
+  auto x_type =
+    CheckAndConvertUtils::CheckTensorTypeValid("x_type", input_args[0]->BuildType(), valid_x_type, prim_name);
+  std::set<TypePtr> valid_index_types = {kInt32, kInt64};
+  CheckAndConvertUtils::CheckTensorTypeValid("index_type", input_args[2]->BuildType(), valid_index_types, prim_name);
+  std::set<TypePtr> valid_dim_type = {kInt32, kInt64};
+  CheckAndConvertUtils::CheckSubClass("dim_type", input_args[1]->BuildType(), valid_dim_type, prim_name);
+  return x_type;
+}
+}  // namespace
+AbstractBasePtr GatherDInfer(const abstract::AnalysisEnginePtr &, const PrimitivePtr &primitive,
+                             const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(primitive);
+  auto abs = std::make_shared<abstract::AbstractTensor>(GatherDInferType(primitive, input_args),
+                                                        GatherDInferShape(primitive, input_args));
+  return abs;
+}
+REGISTER_PRIMITIVE_EVAL_IMPL(GatherD, prim::kPrimGatherD, GatherDInfer, nullptr, false);
+REGISTER_PRIMITIVE_C(kNameGatherD, GatherD);
+}  // namespace ops
+}  // namespace mindspore
diff --git a/mindspore/core/ops/gather_d.h b/mindspore/core/ops/gather_d.h
new file mode 100644
index 0000000000..d39958b118
--- /dev/null
+++ b/mindspore/core/ops/gather_d.h
@@ -0,0 +1,41 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef MINDSPORE_CORE_OPS_GATHER_D_H_
+#define MINDSPORE_CORE_OPS_GATHER_D_H_
+#include <map>
+#include <vector>
+#include <string>
+#include <memory>
+#include "ops/primitive_c.h"
+#include "abstract/abstract_value.h"
+#include "utils/check_convert_utils.h"
+#include "ops/op_utils.h"
+
+namespace mindspore {
+namespace ops {
+constexpr auto kNameGatherD = "GatherD";
+class GatherD : public PrimitiveC {
+ public:
+  GatherD() : PrimitiveC(kNameGatherD) { InitIOName({"x", "dim", "index"}, {"output"}); }
+  ~GatherD() = default;
+  MS_DECLARE_PARENT(GatherD, PrimitiveC);
+  void Init() {}
+};
+}  // namespace ops
+}  // namespace mindspore
+
+#endif  // MINDSPORE_CORE_OPS_GATHER_D_H_
diff --git a/mindspore/core/ops/scalar_summary.cc b/mindspore/core/ops/scalar_summary.cc
index 342c4d4a43..03e0c87adc 100644
--- a/mindspore/core/ops/scalar_summary.cc
+++ b/mindspore/core/ops/scalar_summary.cc
@@ -1,5 +1,5 @@
 /**
- * Copyright 2020 Huawei Technologies Co., Ltd
+ * Copyright 2021 Huawei Technologies Co., Ltd
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -22,7 +22,18 @@
 
 namespace mindspore {
 namespace ops {
-
+// scalar_summary
+namespace {
+abstract::ShapePtr ScalarSummaryInferShape(const PrimitivePtr &primitive,
+                                           const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(primitive);
+  auto prim_name = primitive->name();
+  // check
+  auto v_shape = CheckAndConvertUtils::ConvertShapePtrToShape("v_shape", input_args[1]->BuildShape(), prim_name);
+  CheckAndConvertUtils::CheckInteger("v rank", v_shape.size(), kLessEqual, 1, prim_name);
+  return std::make_shared<abstract::Shape>(ShapeVector(1));
+}
+}  // namespace
 void ScalarSummary::set_side_effect_io() { this->AddAttr(kSideEffectIO, MakeValue(true)); }
 
 bool ScalarSummary::get_side_effect_io() const {
@@ -35,12 +46,9 @@ void ScalarSummary::Init() { this->set_side_effect_io(); }
 AbstractBasePtr ScalarSummaryInfer(const abstract::AnalysisEnginePtr &, const PrimitivePtr &primitive,
                                    const std::vector<AbstractBasePtr> &input_args) {
   MS_EXCEPTION_IF_NULL(primitive);
-  auto prim_name = primitive->name();
   // check
-  CheckAndConvertUtils::CheckSummaryParam(input_args[0], input_args[1], prim_name);
-  auto v_shape = CheckAndConvertUtils::ConvertShapePtrToShape("v_shape", input_args[1]->BuildShape(), prim_name);
-  CheckAndConvertUtils::CheckInteger("v rank", v_shape.size(), kLessEqual, 1, prim_name);
-  return std::make_shared<abstract::AbstractTensor>(kInt32, std::make_shared<abstract::Shape>(ShapeVector(1)));
+  CheckAndConvertUtils::CheckSummaryParam(input_args[0], input_args[1], primitive->name());
+  return std::make_shared<abstract::AbstractTensor>(kInt32, ScalarSummaryInferShape(primitive, input_args));
 }
 REGISTER_PRIMITIVE_EVAL_IMPL(ScalarSummary, prim::kPrimScalarSummary, ScalarSummaryInfer, nullptr, true);
 REGISTER_PRIMITIVE_C(kNameScalarSummary, ScalarSummary);
diff --git a/mindspore/core/ops/scalar_summary.h b/mindspore/core/ops/scalar_summary.h
index 332adef2fc..e89c85aca1 100644
--- a/mindspore/core/ops/scalar_summary.h
+++ b/mindspore/core/ops/scalar_summary.h
@@ -1,5 +1,5 @@
 /**
- * Copyright 2020 Huawei Technologies Co., Ltd
+ * Copyright 2021 Huawei Technologies Co., Ltd
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
diff --git a/mindspore/core/ops/tensor_summary.cc b/mindspore/core/ops/tensor_summary.cc
index 4b17b528e8..efe23d4e43 100644
--- a/mindspore/core/ops/tensor_summary.cc
+++ b/mindspore/core/ops/tensor_summary.cc
@@ -1,5 +1,5 @@
 /**
- * Copyright 2020 Huawei Technologies Co., Ltd
+ * Copyright 2021 Huawei Technologies Co., Ltd
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -22,7 +22,18 @@
 
 namespace mindspore {
 namespace ops {
-
+// scalar_summary
+namespace {
+abstract::ShapePtr TensorSummaryInferShape(const PrimitivePtr &primitive,
+                                           const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(primitive);
+  auto prim_name = primitive->name();
+  // check
+  auto v_shape = CheckAndConvertUtils::ConvertShapePtrToShape("v_shape", input_args[1]->BuildShape(), prim_name);
+  CheckAndConvertUtils::CheckInteger("v rank", v_shape.size(), kGreaterEqual, 1, prim_name);
+  return std::make_shared<abstract::Shape>(ShapeVector(1));
+}
+}  // namespace
 void TensorSummary::set_side_effect_io() { this->AddAttr(kSideEffectIO, MakeValue(true)); }
 
 bool TensorSummary::get_side_effect_io() const {
@@ -35,12 +46,9 @@ void TensorSummary::Init() { this->set_side_effect_io(); }
 AbstractBasePtr TensorSummaryInfer(const abstract::AnalysisEnginePtr &, const PrimitivePtr &primitive,
                                    const std::vector<AbstractBasePtr> &input_args) {
   MS_EXCEPTION_IF_NULL(primitive);
-  auto prim_name = primitive->name();
   // check
-  CheckAndConvertUtils::CheckSummaryParam(input_args[0], input_args[1], prim_name);
-  auto v_shape = CheckAndConvertUtils::ConvertShapePtrToShape("v_shape", input_args[1]->BuildShape(), prim_name);
-  CheckAndConvertUtils::CheckInteger("v rank", v_shape.size(), kGreaterEqual, 1, prim_name);
-  return std::make_shared<abstract::AbstractTensor>(kInt32, std::make_shared<abstract::Shape>(ShapeVector(1)));
+  CheckAndConvertUtils::CheckSummaryParam(input_args[0], input_args[1], primitive->name());
+  return std::make_shared<abstract::AbstractTensor>(kInt32, TensorSummaryInferShape(primitive, input_args));
 }
 REGISTER_PRIMITIVE_EVAL_IMPL(TensorSummary, prim::kPrimTensorSummary, TensorSummaryInfer, nullptr, true);
 REGISTER_PRIMITIVE_C(kNameTensorSummary, TensorSummary);
diff --git a/mindspore/core/ops/tensor_summary.h b/mindspore/core/ops/tensor_summary.h
index ec8d5b676a..61badeadb2 100644
--- a/mindspore/core/ops/tensor_summary.h
+++ b/mindspore/core/ops/tensor_summary.h
@@ -1,5 +1,5 @@
 /**
- * Copyright 2020 Huawei Technologies Co., Ltd
+ * Copyright 2021 Huawei Technologies Co., Ltd
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
diff --git a/mindspore/core/ops/zeros.cc b/mindspore/core/ops/zeros.cc
new file mode 100644
index 0000000000..e39a0a58a2
--- /dev/null
+++ b/mindspore/core/ops/zeros.cc
@@ -0,0 +1,75 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "ops/zeros.h"
+#include <memory>
+#include <set>
+#include "ops/op_utils.h"
+#include "utils/check_convert_utils.h"
+#include "utils/tensor_construct_utils.h"
+#include "abstract/primitive_infer_map.h"
+
+namespace mindspore {
+namespace ops {
+// zeros
+namespace {
+abstract::ShapePtr ZerosInferShape(const PrimitivePtr &primitive, const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(primitive);
+  auto prim_name = primitive->name();
+  // check
+  auto shape_value = input_args[0]->BuildValue();
+  std::vector<int64_t> out_shape = CheckAndConvertUtils::CheckAttrIntOrTupleInt("shape", shape_value, prim_name);
+  CheckAndConvertUtils::CheckPositiveVector("shape", out_shape, prim_name);
+  return std::make_shared<abstract::Shape>(out_shape);
+}
+
+TypePtr ZerosInferType(const PrimitivePtr &prim, const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(prim);
+  auto prim_name = prim->name();
+  // check
+  auto dtype_value = input_args[1]->BuildValue();
+  if (!dtype_value->isa<Type>()) {
+    MS_EXCEPTION(TypeError) << "The dtype of Zeros is invalid!";
+  }
+  auto output_type = dtype_value->cast<TypePtr>();
+  const std::set<TypePtr> valid_types = {kBool,   kInt8,   kInt16,  kInt32,   kInt64,   kUInt8,
+                                         kUInt16, kUInt32, kUInt64, kFloat16, kFloat32, kFloat64};
+  return CheckAndConvertUtils::CheckSubClass("dtype", output_type, valid_types, prim_name);
+}
+ValuePtr ZerosInferValue(const PrimitivePtr &prim, const std::vector<AbstractBasePtr> &input_args,
+                         const abstract::AbstractBasePtr &abs) {
+  MS_EXCEPTION_IF_NULL(prim);
+  auto prim_name = prim->name();
+  // check
+  auto out_shape = CheckAndConvertUtils::ConvertShapePtrToShape("output shape", abs->BuildShape(), prim_name);
+  auto out_type = abs->BuildType();
+  MS_EXCEPTION_IF_NULL(out_type);
+  return TensorConstructUtils::CreateZerosTensor(out_type, out_shape);
+}
+}  // namespace
+
+AbstractBasePtr ZerosInfer(const abstract::AnalysisEnginePtr &, const PrimitivePtr &primitive,
+                           const std::vector<AbstractBasePtr> &input_args) {
+  MS_EXCEPTION_IF_NULL(primitive);
+  auto abs = std::make_shared<abstract::AbstractTensor>(ZerosInferType(primitive, input_args),
+                                                        ZerosInferShape(primitive, input_args));
+  abs->set_value(ZerosInferValue(primitive, input_args, abs));
+  return abs;
+}
+REGISTER_PRIMITIVE_EVAL_IMPL(Zeros, prim::kPrimZeros, ZerosInfer, ZerosInferValue, false);
+REGISTER_PRIMITIVE_C(kNameZeros, Zeros);
+}  // namespace ops
+}  // namespace mindspore
diff --git a/mindspore/core/ops/zeros.h b/mindspore/core/ops/zeros.h
new file mode 100644
index 0000000000..b9afc5d8e2
--- /dev/null
+++ b/mindspore/core/ops/zeros.h
@@ -0,0 +1,41 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef MINDSPORE_CORE_OPS_ZEROS_H_
+#define MINDSPORE_CORE_OPS_ZEROS_H_
+#include <map>
+#include <vector>
+#include <string>
+#include <memory>
+#include "ops/primitive_c.h"
+#include "abstract/abstract_value.h"
+#include "utils/check_convert_utils.h"
+#include "ops/op_utils.h"
+
+namespace mindspore {
+namespace ops {
+constexpr auto kNameZeros = "Zeros";
+class Zeros : public PrimitiveC {
+ public:
+  Zeros() : PrimitiveC(kNameZeros) {}
+  ~Zeros() = default;
+  MS_DECLARE_PARENT(Zeros, PrimitiveC);
+  void Init() {}
+};
+}  // namespace ops
+}  // namespace mindspore
+
+#endif  // MINDSPORE_CORE_OPS_ZEROS_H_
diff --git a/mindspore/core/utils/check_convert_utils.cc b/mindspore/core/utils/check_convert_utils.cc
index 0aa3d3d706..8e2594165e 100644
--- a/mindspore/core/utils/check_convert_utils.cc
+++ b/mindspore/core/utils/check_convert_utils.cc
@@ -442,8 +442,8 @@ TypePtr CheckAndConvertUtils::CheckTensorTypeSame(const std::map<std::string, Ty
   auto type = types.begin()->second;
   MS_EXCEPTION_IF_NULL(type);
   if (!type->isa<TensorType>()) {
-    MS_EXCEPTION(TypeError) << "The " << prim_name << "'s" << types.begin()->first << " input must be a tensor but got "
-                            << type->ToString();
+    MS_EXCEPTION(TypeError) << "The " << prim_name << "'s " << types.begin()->first
+                            << " input must be a tensor but got " << type->ToString();
   }
   TypePtr check_type = _CheckTypeSame(types, prim_name, false);
   return CheckTypeValid(types.begin()->first, check_type, check_list, prim_name);
@@ -599,4 +599,27 @@ void CheckAndConvertUtils::CheckMode(const std::string &class_name) {
     MS_EXCEPTION(NotSupportError) << class_name << "operator does not support PyNative mode.";
   }
 }
+
+std::vector<int64_t> CheckAndConvertUtils::CheckAttrIntOrTupleInt(const std::string &arg_name, const ValuePtr &attr,
+                                                                  const std::string &prim_name) {
+  std::vector<int64_t> result;
+  MS_EXCEPTION_IF_NULL(attr);
+  if (attr->isa<ValueTuple>()) {
+    std::vector<ValuePtr> attr_vec = attr->cast<ValueTuplePtr>()->value();
+    (void)std::transform(
+      attr_vec.begin(), attr_vec.end(), std::back_inserter(result), [=](const ValuePtr &e) -> int64_t {
+        if (!e->isa<Int64Imm>()) {
+          MS_EXCEPTION(TypeError) << "For " << prim_name << ", the type of" << arg_name << " must be Int64";
+        }
+        return GetValue<int64_t>(e);
+      });
+  } else {
+    if (!attr->isa<Int64Imm>()) {
+      MS_EXCEPTION(TypeError) << "For " << prim_name << ", the type of" << arg_name << " must be Int64";
+    }
+    int64_t attr_val = attr->cast<Int64ImmPtr>()->value();
+    result.push_back(attr_val);
+  }
+  return result;
+}
 }  // namespace mindspore
diff --git a/mindspore/core/utils/check_convert_utils.h b/mindspore/core/utils/check_convert_utils.h
index 03bf225164..c3e76d13ee 100644
--- a/mindspore/core/utils/check_convert_utils.h
+++ b/mindspore/core/utils/check_convert_utils.h
@@ -321,6 +321,8 @@ class CheckAndConvertUtils {
   static void CheckSummaryParam(const AbstractBasePtr &name, const AbstractBasePtr &value,
                                 const std::string &class_name);
   static void CheckMode(const std::string &class_name);
+  static std::vector<int64_t> CheckAttrIntOrTupleInt(const std::string &prim_name, const ValuePtr &attr,
+                                                     const std::string &arg_name);
 
  private:
   static bool IsEqualVector(const std::vector<int64_t> &vec_1, const std::vector<int64_t> &vec_2);
diff --git a/mindspore/core/utils/tensor_construct_utils.cc b/mindspore/core/utils/tensor_construct_utils.cc
index 1563ecc074..a34b205b28 100644
--- a/mindspore/core/utils/tensor_construct_utils.cc
+++ b/mindspore/core/utils/tensor_construct_utils.cc
@@ -17,8 +17,10 @@
 #include <vector>
 #include <memory>
 namespace mindspore {
-tensor::TensorPtr TensorConstructUtils::CreateZerosTensor(TypeId type, const std::vector<int64_t> &shape) {
-  tensor::TensorPtr tensor = std::make_shared<tensor::Tensor>(type, shape);
+tensor::TensorPtr TensorConstructUtils::CreateZerosTensor(const TypePtr type_ptr, const std::vector<int64_t> &shape) {
+  MS_EXCEPTION_IF_NULL(type_ptr);
+  auto type_id = ExtractTypeId(type_ptr);
+  tensor::TensorPtr tensor = std::make_shared<tensor::Tensor>(type_id, shape);
   size_t mem_size = IntToSize(tensor->ElementsNum());
   auto tensor_data = tensor->data_c();
   char *data = reinterpret_cast<char *>(tensor_data);
@@ -28,8 +30,10 @@ tensor::TensorPtr TensorConstructUtils::CreateZerosTensor(TypeId type, const std
   return tensor;
 }
 
-tensor::TensorPtr TensorConstructUtils::CreateOnesTensor(TypeId type, const std::vector<int64_t> &shape) {
-  tensor::TensorPtr tensor = std::make_shared<tensor::Tensor>(type, shape);
+tensor::TensorPtr TensorConstructUtils::CreateOnesTensor(const TypePtr type_ptr, const std::vector<int64_t> &shape) {
+  MS_EXCEPTION_IF_NULL(type_ptr);
+  auto type_id = ExtractTypeId(type_ptr);
+  tensor::TensorPtr tensor = std::make_shared<tensor::Tensor>(type_id, shape);
   size_t mem_size = IntToSize(tensor->ElementsNum());
   if (tensor->data_type() == kNumberTypeFloat32) {
     SetTensorData<float>(tensor->data_c(), 1.0, mem_size);
@@ -39,8 +43,18 @@ tensor::TensorPtr TensorConstructUtils::CreateOnesTensor(TypeId type, const std:
   return tensor;
 }
 
-tensor::TensorPtr TensorConstructUtils::CreateTensor(TypeId type, const std::vector<int64_t> &shape, void *data) {
-  tensor::TensorPtr tensor = std::make_shared<tensor::Tensor>(type, shape, data, type);
+tensor::TensorPtr TensorConstructUtils::CreateTensor(const TypePtr type_ptr, const std::vector<int64_t> &shape,
+                                                     void *data) {
+  MS_EXCEPTION_IF_NULL(type_ptr);
+  auto type_id = ExtractTypeId(type_ptr);
+  tensor::TensorPtr tensor = std::make_shared<tensor::Tensor>(type_id, shape, data, type_id);
   return tensor;
 }
+
+TypeId TensorConstructUtils::ExtractTypeId(const TypePtr type_ptr) {
+  MS_EXCEPTION_IF_NULL(type_ptr);
+  auto tensor_type = type_ptr->cast<TensorTypePtr>();
+  auto type_id = tensor_type->element()->type_id();
+  return type_id;
+}
 }  // namespace mindspore
diff --git a/mindspore/core/utils/tensor_construct_utils.h b/mindspore/core/utils/tensor_construct_utils.h
index 4bb87ab27b..72dac8f4cc 100644
--- a/mindspore/core/utils/tensor_construct_utils.h
+++ b/mindspore/core/utils/tensor_construct_utils.h
@@ -30,9 +30,10 @@ void SetTensorData(void *data, T num, size_t data_length) {
 }
 class TensorConstructUtils {
  public:
-  static tensor::TensorPtr CreateZerosTensor(TypeId type, const std::vector<int64_t> &shape);
-  static tensor::TensorPtr CreateOnesTensor(TypeId type, const std::vector<int64_t> &shape);
-  static tensor::TensorPtr CreateTensor(TypeId type, const std::vector<int64_t> &shape, void *data);
+  static tensor::TensorPtr CreateZerosTensor(const TypePtr type, const std::vector<int64_t> &shape);
+  static tensor::TensorPtr CreateOnesTensor(const TypePtr type, const std::vector<int64_t> &shape);
+  static tensor::TensorPtr CreateTensor(const TypePtr type, const std::vector<int64_t> &shape, void *data);
+  static TypeId ExtractTypeId(const TypePtr type);
 };
 }  // namespace mindspore
 #endif  // MINDSPORE_CORE_UTILS_TENSOR_CONSTRUCT_UTILS_H_
diff --git a/mindspore/ops/operations/array_ops.py b/mindspore/ops/operations/array_ops.py
index 0ad7cf83f6..f379ac50ee 100644
--- a/mindspore/ops/operations/array_ops.py
+++ b/mindspore/ops/operations/array_ops.py
@@ -1342,27 +1342,6 @@ class Zeros(PrimitiveWithInfer):
     def __init__(self):
         """Initialize Zeros"""
 
-    def __infer__(self, dims, dtype):
-        if isinstance(dims['value'], int):
-            shape = (dims['value'],)
-        else:
-            shape = dims['value']
-        validator.check_value_type("shape", shape, [tuple], self.name)
-        for i, item in enumerate(shape):
-            validator.check_non_negative_int(item, shape[i], self.name)
-        valid_types = [mstype.bool_, mstype.int8, mstype.int16, mstype.int32, mstype.int64,
-                       mstype.uint8, mstype.uint16, mstype.uint32, mstype.uint64,
-                       mstype.float16, mstype.float32, mstype.float64]
-        validator.check_types_same_and_valid({"value": dtype['value']}, valid_types, self.name)
-        x_nptype = mstype.dtype_to_nptype(dtype['value'])
-        ret = np.zeros(shape, x_nptype)
-        out = {
-            'value': Tensor(ret),
-            'shape': shape,
-            'dtype': x_nptype,
-        }
-        return out
-
 
 class OnesLike(PrimitiveWithInfer):
     """
@@ -5193,30 +5172,6 @@ class GatherD(PrimitiveWithInfer):
         """Initialize GatherD"""
         self.init_prim_io_names(inputs=['x', 'dim', 'index'], outputs=['output'])
 
-    def __infer__(self, x, dim, index):
-        validator.check_subclass("x", x['dtype'], mstype.tensor, self.name)
-        validator.check_tensor_dtype_valid("index", index['dtype'], [mstype.int32, mstype.int64], self.name)
-        validator.check_subclass("dim", dim['dtype'], [mstype.int32, mstype.int64], self.name)
-        x_shp = x['shape']
-        idx_shp = index['shape']
-        x_rank = len(x_shp)
-        idx_rank = len(idx_shp)
-        validator.check("x_rank, idx_rank", x_rank, "expected", idx_rank, Rel.EQ, self.name)
-        dim_v = dim['value']
-        validator.check("dim value", dim_v, "expected", -x_rank, Rel.GE, self.name)
-        validator.check("dim value", dim_v, "expected", x_rank, Rel.LT, self.name)
-        if dim_v < 0:
-            dim['value'] = dim_v + x_rank
-        for i in range(x_rank):
-            if i == dim['value']:
-                continue
-            validator.check("x_shp[{0}], idx_shp[{0}]".format(i), x_shp[i], "expected", idx_shp[i], Rel.EQ, self.name)
-
-        out = {'shape': index['shape'],
-               'dtype': x['dtype'],
-               'value': None}
-        return out
-
 
 class Identity(PrimitiveWithInfer):
     """
diff --git a/mindspore/ops/operations/debug_ops.py b/mindspore/ops/operations/debug_ops.py
index 1c2bddb37a..06b0bb2f60 100644
--- a/mindspore/ops/operations/debug_ops.py
+++ b/mindspore/ops/operations/debug_ops.py
@@ -89,17 +89,6 @@ class ScalarSummary(PrimitiveWithInfer):
         """init"""
         self.add_prim_attr("side_effect_io", True)
 
-    def __infer__(self, name, value):
-        _check_summary_param(name, value, self.__class__.__name__)
-
-        v_shape = value['shape']
-        # In the summary, the value whose shape is [1] is also considered as a scalar.
-        if v_shape and v_shape != [1]:
-            raise ValueError(f"For 'value' the type should be scalar, "
-                             f"shape should be [] or [1] in {self.__class__.__name__}, but got {v_shape}.")
-
-        return SUMMARY_RETURN_VALUE
-
 
 class ImageSummary(PrimitiveWithInfer):
     """
@@ -191,17 +180,6 @@ class TensorSummary(PrimitiveWithInfer):
         """init"""
         self.add_prim_attr("side_effect_io", True)
 
-    def __infer__(self, name, value):
-        _check_summary_param(name, value, self.__class__.__name__)
-
-        v_shape = value['shape']
-        # In the summary, the value whose shape is [] is not considered as a tensor.
-        if not v_shape:
-            raise ValueError(f"For 'value' the type should be tensor in {self.__class__.__name__}, "
-                             f"shape should not be [].")
-
-        return SUMMARY_RETURN_VALUE
-
 
 class HistogramSummary(PrimitiveWithInfer):
     """