Arm NN is an inference engine for CPUs, GPUs and NPUs

 Arm NN is a set of tools that enables machine learning workloads on
 any hardware. It provides a bridge between existing neural network
 frameworks and whatever hardware is available and supported. On arm
 architectures (arm64 and armhf) it utilizes the Arm Compute Library
 to target Cortex-A CPUs, Mali GPUs and Ethos NPUs as efficiently as
 possible. On other architectures/hardware it falls back to unoptimised
 functions.
 .
 This release supports Caffe, TensorFlow, TensorFlow Lite, and ONNX.
 Arm NN takes networks from these frameworks, translates them
 to the internal Arm NN format and then through the Arm Compute Library,
 deploys them efficiently on Cortex-A CPUs, and, if present, Mali GPUs.
 .
 This is the shared library package.