Before you begin
To work through this guide, you will need two hardware devices and some installed software.
You will need these devices:
A HiKey 960 development board
The HiKey 960 runs the final demo, using Autoware and Arm NN to run a real-time object detection system.
Note: This guide has been developed and tested on the HiKey 960 device. If you choose to use another platform, you will need to modify the various scripts used in this tutorial.
A Linux or Apple Mac host machine
The host machine is used to prepare an Ubuntu filesystem and flash the base firmware and OS onto the Hikey960 device. The host machine is also used for some of the more computationally intensive tasks, like converting the neural network to the TensorFlow protobuf format.
Before starting the instructions in this guide, you must install Ubuntu Linux and build Arm NN on the HiKey 960. This installation process is described in our guide Running and profiling Arm NN on the HiKey 960 .
Specifically, you must complete the following steps in Running and profiling Arm NN on the HiKey 960:
- Before you begin
- Run Ubuntu Linux on the HiKey 960
- Build an Ubuntu filesystem
- Flash the base firmware and OS
- Flash the base firmware and OS - recovery mode
- Flash the base firmware and OS - fastboot mode
- Boot Linux
- Add more diskspace
You can stop after performing the Add more diskspace step, which includes building Arm NN.