fix bug in Parameter init_data

mindspore/ccsrc/pipeline/jit/pipeline.cc

@@ -751,19 +751,17 @@ py::object ExecutorPy::Run(const py::tuple &args, const py::object &phase) {
     return ExecDFGraph(info_, args, phase_s);
   }
 #else
-  if (backend == "ms" || backend == "ge") {
-    auto ret_val = std::make_shared<py::object>();
-    if (info_.count(phase_s) != 0 && info_[phase_s]->func_graph != nullptr) {
-      if (IsGraphOutputValueNodeOrParameter(info_[phase_s]->func_graph->output(), args, ret_val)) {
-        return *ret_val;
-      }
+  auto ret_val = std::make_shared<py::object>();
+  if (info_.count(phase_s) != 0 && info_[phase_s]->func_graph != nullptr) {
+    if (IsGraphOutputValueNodeOrParameter(info_[phase_s]->func_graph->output(), args, ret_val)) {
+      return *ret_val;
     }
-    if (backend == "ge") {
-      if (args.size() > 0) {
-        return args[0];
-      }
-      return args;
+  }
+  if (backend == "ge") {
+    if (args.size() > 0) {
+      return args[0];
     }
+    return args;
   }
 #endif
   std::size_t full_arg_size = ArgListSize(phase_s);

mindspore/common/parameter.py

@@ -389,6 +389,8 @@ class Parameter(MetaTensor):
                 raise RuntimeError("Must set or change parallel mode before any Initializer created.")
         if self.init_mode is None:
             return self
+        if self.inited_param is not None:
+            return self.inited_param
         if layout is not None:
             if not isinstance(layout, list):
                 raise TypeError("The layout should be list! layout is {}.".format(layout))

mindspore/core/ir/meta_tensor_extends.cc

@@ -36,8 +36,8 @@ abstract::AbstractBasePtr MetaTensor::ToAbstract() {
   auto abs_tensor = std::make_shared<abstract::AbstractTensor>(dtype, tensor_shape);
 
   // if is parameter always no value.
-  if (is_parameter()) {
-    auto param_name = param_info()->name();
+  if (is_parameter_) {
+    auto param_name = param_info_->name();
     auto ref_key = std::make_shared<RefKey>(param_name);
     auto abs_ref_key = ref_key->ToAbstract();
     abs_tensor = std::make_shared<abstract::AbstractRef>(abs_ref_key, abs_tensor);

mindspore/core/ir/tensor.cc

@@ -476,8 +476,8 @@ abstract::AbstractBasePtr Tensor::ToAbstract() {
   auto tensor_shape = tens->shape();
   auto abs_tensor = std::make_shared<abstract::AbstractTensor>(dtype, tensor_shape);
   // if is parameter always no value.
-  if (is_parameter()) {
-    auto param_name = param_info()->name();
+  if (is_parameter_) {
+    auto param_name = param_info_->name();
     auto ref_key = std::make_shared<RefKey>(param_name);
     auto abs_ref_key = ref_key->ToAbstract();
     abs_tensor = std::make_shared<abstract::AbstractRef>(abs_ref_key, abs_tensor);

tests/ut/python/nn/test_parameter.py

@@ -17,7 +17,7 @@
 import numpy as np
 import pytest
 
-from mindspore import context, Tensor, Parameter, ParameterTuple
+from mindspore import context, Tensor, Parameter, ParameterTuple, nn
 from mindspore._checkparam import _check_str_by_regular
 from mindspore.common import dtype as mstype
 from mindspore.common.initializer import initializer
@@ -229,3 +229,25 @@ def test_parameter_lazy_init():
     para.set_parameter_data(initializer('ones', [1, 2], mstype.float32), slice_shape=True)
     assert np.array_equal(para.default_input.asnumpy(), np.ones((1, 2)))
     context.reset_auto_parallel_context()
+
+
+def test_parameter_as_output():
+    context.reset_auto_parallel_context()
+    context.set_auto_parallel_context(parallel_mode="semi_auto_parallel")
+    initial_input = initializer('One', shape=(2,), dtype=mstype.int32)
+    updated_input = Tensor([2, 2], mstype.int32)
+    class Net(nn.Cell):
+        def __init__(self, initial, updated):
+            super().__init__()
+            self.initial = initial
+            self.updated = updated
+            self.p = Parameter(self.initial, name="weight")
+            self.new_p = self.p.init_data()
+            self.new_p.set_parameter_data(self.updated)
+        def construct(self):
+            return self.new_p
+
+    net = Net(initial_input, updated_input)
+    output = net()
+    assert np.array_equal(output.asnumpy(), np.array([2, 2], np.int32))
+    context.reset_auto_parallel_context()