fix build errors

src/ngraph/runtime/interpreter/int_executable.cpp

@@ -39,8 +39,6 @@ using namespace ngraph;
 
 using descriptor::layout::DenseTensorLayout;
 
-atomic<size_t> runtime::interpreter::INTExecutable::m_next_group_id;
-
 runtime::interpreter::INTExecutable::INTExecutable(const shared_ptr<Function>& function,
                                                    bool enable_performance_collection)
     : m_is_compiled{true}
@@ -345,7 +343,6 @@ vector<shared_ptr<runtime::Tensor>>
 {
     vector<shared_ptr<runtime::HostTensor>> tensors;
     shared_ptr<op::Parameter> parameter = get_parameter(input_index);
-    size_t group_id = m_next_group_id.fetch_add(1);
     for (size_t i = 0; i < pipeline_depth; i++)
     {
         shared_ptr<runtime::HostTensor> tensor;
@@ -368,7 +365,6 @@ vector<shared_ptr<runtime::Tensor>>
 {
     vector<shared_ptr<runtime::HostTensor>> tensors;
     shared_ptr<op::Result> result = get_result(output_index);
-    size_t group_id = m_next_group_id.fetch_add(1);
     for (size_t i = 0; i < pipeline_depth; i++)
     {
         shared_ptr<runtime::HostTensor> tensor;

src/ngraph/runtime/interpreter/int_executable.hpp

@@ -210,7 +210,6 @@ private:
     std::vector<NodeWrapper> m_wrapped_nodes;
     std::unordered_map<const Node*, std::shared_ptr<RNGState>> m_states;
     std::set<std::string> m_unsupported_op_name_list;
-    static std::atomic<size_t> m_next_group_id;
 
     static void perform_nan_check(const std::vector<std::shared_ptr<HostTensor>>&,
                                   const Node* op = nullptr);