LCE_compute_engine #779
-
Hello I am trying to integrate to my c++ code the larq compute engine using a cmake file, and I am new to this so I am failing to do that, I cant find any docs on that, I already managed to add tensorflow lite and I am doing normal inference with my c++ code, so my question is how can I add LCE using cmake file and should I remove the tensorflow lite that I added since tensorflow lite is part of the LCE build ? |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 2 replies
-
Hello, add_subdirectory( <path_to_tensorflow_lite> "${CMAKE_CURRENT_BINARY_DIR}/tensorflow-lite" EXCLUDE_FROM_ALL )
...
target_link_libraries( <your_binary> tensorflow-lite ) To use LCE's binary layers, do the following instead: add_subdirectory( <path_to_larq_compute_engine> )
...
target_link_libraries( <your_binary> larq-compute-engine ) (remove the old Note that you have to update compute-engine to the latest |
Beta Was this translation helpful? Give feedback.
Hello,
If you are using tensorflow-lite with cmake, then you probably have something like this:
To use LCE's binary layers, do the following instead:
(remove the old
add_subdirectory
for tensorflow-lite, that is done inside the larq-compute-engine cmake file).Note that you have to update compute-engine to the latest
main
branch because I just merged a fix for this (#780).