fix shape of tile_grad op (#29289) (#29324)

paddle/fluid/operators/tile_op.cc

@@ -167,6 +167,7 @@ class TileGradOp : public framework::OperatorWithKernel {
                    framework::GradVarName("Out"), "TileGrad");
 
     auto x_dims = ctx->GetInputDim("X");
+
     std::vector<int> repeat_times =
         ctx->Attrs().Get<std::vector<int>>("repeat_times");
     if (repeat_times.size() == 0) {

paddle/fluid/operators/tile_op.h

@@ -186,9 +186,9 @@ template <typename DeviceContext, typename T>
 class TileGradKernel : public framework::OpKernel<T> {
  public:
   void Compute(const framework::ExecutionContext& context) const override {
-    auto* in0 = context.Input<Tensor>("X");
+    auto* x = context.Input<Tensor>("X");
     auto repeat_times = get_repeat_times(context);
-    auto x_dims = in0->dims();
+    auto x_dims = x->dims();
     auto vec_in_dims = framework::vectorize<int>(x_dims);
     if (repeat_times.size() < vec_in_dims.size()) {
       int diff = vec_in_dims.size() - repeat_times.size();
@@ -220,11 +220,13 @@ class TileGradKernel : public framework::OpKernel<T> {
     }
     // no need reduce, just copy
     if (just_copy) {
-      auto* in0 = context.Input<Tensor>(framework::GradVarName("Out"));
-      auto* out0 = context.Output<Tensor>(framework::GradVarName("X"));
-      out0->mutable_data<T>(context.GetPlace());
-      framework::TensorCopy(*in0, context.GetPlace(), context.device_context(),
-                            out0);
+      auto* dout = context.Input<Tensor>(framework::GradVarName("Out"));
+      auto* dx = context.Output<Tensor>(framework::GradVarName("X"));
+      dx->mutable_data<T>(context.GetPlace());
+      framework::TensorCopy(*dout, context.GetPlace(), context.device_context(),
+                            dx);
+      // TensorCopy may change the dims of dx
+      dx->Resize(x_dims);
     } else {
       PADDLE_ENFORCE_GE(dims, 1,
                         platform::errors::InvalidArgument(
@@ -261,6 +263,7 @@ class TileGradKernel : public framework::OpKernel<T> {
     for (size_t i = 0; i < reduce_size; ++i) {
       reduce_dims[i] = reduce_dims_vec[i];
     }
+
     auto out_grad = EigenVector<T>::Flatten(*in0);
     x_grad.device(
         *context.template device_context<DeviceContext>().eigen_device()) =