If you are using Tizen ML API from a .RPM platform binary (you are building with GBS, not Tizen Studio), not from a .TPK application, you can bypass privilege controls; thus, you may access it right now, today with Tizen:Unified binaries.
Add nnstreamer-tizen-internal-capi-devel in your GBS project,
Include "nnstreamer-tizen-internal.h" (at /usr/include/nnstreamer/)
Use ml_pipeline_construct_internal instead of ml_pipeline_constrct.
Then, you may use EVERY features without privilege control in Tizen.
Note that you are not allowed to use this interface in a .TPK application, which is not executed in platform space.
