What is Arm NN?
Arm NN is an inference engine for Arm CPUs, GPUs, and NPUs. Arm NN executes machine learning models on Arm-based devices to make predictions based on input data.
Arm NN bridges the gap between existing NN frameworks and the underlying IP, as shown in the following diagram:
Arm NN enables efficient translation of existing neural network frameworks, like TensorFlow Lite, TensorFlow, ONNX, and Caffe. This translation process allows neural networks to run efficiently, without modification, across Arm Cortex-A CPUs, Arm Mali GPUs, and Arm Ethos NPUs.
Arm NN includes the TensorFlow parser
armnnTfParser. This parser is a library for loading neural networks that are
defined by TensorFlow into the Arm NN runtime. This guide uses the TensorFlow
parser to parse our object detection model.