Deploying an application using the Arm NN SDK

You must link your application with the Arm NN SDK library and the Caffe parsing library. The following example code links your application. You must use your own Arm NN library path to replace <your_armnn_library_path> in the code:

g++ -O3 -std=c++14 -I$(ARMNN_INC) mnist_caffe.cpp -o mnist_caffe -L$(ARMNN_LIB)-larmnn -larmnnCaffeParser 
export LD_LIBRARY_PATH = <your_armnn_library_path>: $LD_LIBRARY_PATH 

For convenience, the Arm NN SDK can run with a reference implementation on x86 for development and testing, but the optimized kernels are only available for Arm CPUs and GPUs. This means that you must run your profiling and optimization steps on a real Arm-powered device because x86 performance does not represent real performance.