minor fix

client_example/23_elementwise_transpose/elementwise_transpose_3d.cpp

@@ -63,17 +63,17 @@ int main()
     std::array<void*, 1> output      = {b_dev_buf.GetDeviceBuffer()};
 
     using DeviceElementwisePermuteInstance =
-        ck::tensor_operation::device::DeviceElementwise<ck::Tuple<ADataType>,
-                                                        ck::Tuple<BDataType>,
-                                                        PassThrough,
-                                                        2,
-                                                        2,
-                                                        1,
-                                                        8,
-                                                        8,
-                                                        8,
-                                                        ck::Sequence<8>,
-                                                        ck::Sequence<1>>;
+        ck::tensor_operation::device::DeviceElementwise3dImpl<ck::Tuple<ADataType>,
+                                                              ck::Tuple<BDataType>,
+                                                              PassThrough,
+                                                              2,
+                                                              2,
+                                                              1,
+                                                              8,
+                                                              8,
+                                                              8,
+                                                              ck::Sequence<8>,
+                                                              ck::Sequence<1>>;
 
     // get device op instances
     const auto op_ptrs = ck::tensor_operation::device::instance::DeviceOperationInstanceFactory<