Fix specialization of template member functions in the non-template class in GCC 5.0.

paddle/framework/operator.cc

@@ -209,8 +209,7 @@ const std::vector<const Tensor*> InferShapeContext::MultiInput<Tensor>(
 template <>
 Tensor* ExecutionContext::Output<Tensor>(const std::string& name) const {
   auto* var = OutputVar(name);
-  if (var == nullptr) return nullptr;
-  return GetTensorFromVar(var);
+  return var == nullptr ? nullptr : const_cast<Tensor*>(GetTensorFromVar(var));
 }
 
 template <>
@@ -222,7 +221,9 @@ std::vector<Tensor*> ExecutionContext::MultiOutput<Tensor>(
   std::transform(names.begin(), names.end(), std::back_inserter(res),
                  [&](const std::string& sub_name) {
                    auto var = scope().FindVar(sub_name);
-                   return var == nullptr ? nullptr : GetTensorFromVar(var);
+                   return var == nullptr
+                              ? nullptr
+                              : const_cast<Tensor*>(GetTensorFromVar(var));
                  });
   return res;
 }

paddle/framework/operator.h

@@ -327,13 +327,13 @@ class InferShapeContext {
     return res;
   }
 
-  Tensor* GetTensorFromVar(const Variable* var) const {
+  const Tensor* GetTensorFromVar(const Variable* var) const {
     if (var->IsType<LoDTensor>()) {
-      return const_cast<LoDTensor*>(&var->Get<LoDTensor>());
+      return &var->Get<LoDTensor>();
     }
     PADDLE_ENFORCE(var->IsType<Tensor>(),
                    "The Input(%s) must be LoDTensor or Tensor.");
-    return const_cast<Tensor*>(&var->Get<Tensor>());
+    return &var->Get<Tensor>();
   }
 
  private:
@@ -341,6 +341,13 @@ class InferShapeContext {
   const Scope& scope_;
 };
 
+template <>
+const Tensor* InferShapeContext::Input<Tensor>(const std::string& name) const;
+
+template <>
+const std::vector<const Tensor*> InferShapeContext::MultiInput<Tensor>(
+    const std::string& name) const;
+
 template <typename T>
 struct EigenDeviceConverter;
 
@@ -397,6 +404,13 @@ class ExecutionContext : public InferShapeContext {
   const platform::DeviceContext* device_context_;
 };
 
+template <>
+Tensor* ExecutionContext::Output<Tensor>(const std::string& name) const;
+
+template <>
+std::vector<Tensor*> ExecutionContext::MultiOutput<Tensor>(
+    const std::string& name) const;
+
 class OpKernel {
  public:
   /**

paddle/operators/sequence_avg_pool_op.cc

@@ -66,7 +66,7 @@ class SequenceAvgPoolGradOp : public framework::OperatorWithKernel {
     auto x_dims = ctx.Input<framework::LoDTensor>("X")->dims();
     PADDLE_ENFORCE_EQ(og_dims.size(), x_dims.size(),
                       "The rank of output grad must equal to Input(X).");
-    for (size_t i = 1; i < og_dims.size(); ++i) {
+    for (int64_t i = 1; i < og_dims.size(); ++i) {
       PADDLE_ENFORCE_EQ(og_dims[i], x_dims[i], "The dimension mismatch.");
     }
     auto* x_grad =

paddle/operators/sequence_avg_pool_op.h

@@ -38,7 +38,7 @@ class SequenceAvgPoolKernel : public framework::OpKernel {
 
     out->mutable_data<T>(context.GetPlace());
     auto place = context.GetEigenDevice<Place>();
-    for (int i = 0; i < lod[0].size() - 1; ++i) {
+    for (int i = 0; i < static_cast<int>(lod[0].size()) - 1; ++i) {
       Tensor in_t = in->Slice<T>(static_cast<int>(lod[0][i]),
                                  static_cast<int>(lod[0][i + 1]));
       Tensor out_t = out->Slice<T>(i, i + 1);
@@ -64,7 +64,7 @@ class SequenceAvgPoolGradKernel : public framework::OpKernel {
 
     in_g->mutable_data<T>(context.GetPlace());
     auto place = context.GetEigenDevice<Place>();
-    for (int i = 0; i < lod[0].size() - 1; ++i) {
+    for (int i = 0; i < static_cast<int>(lod[0].size()) - 1; ++i) {
       auto in_g_t = in_g->Slice<T>(static_cast<int>(lod[0][i]),
                                    static_cast<int>(lod[0][i + 1]));
       auto out_g_t = out_g->Slice<T>(i, i + 1);