https://github.com/yehengchen/ncs1-rpi

Intel® Neural Compute Stick Neural Compute Stick1 + Raspberry pi 3B NCSDK/NCAPPZOO
https://github.com/yehengchen/ncs1-rpi

intel ncs ncsdk neural-compute-stick rpi3

Last synced: about 1 month ago
JSON representation

Intel® Neural Compute Stick Neural Compute Stick1 + Raspberry pi 3B NCSDK/NCAPPZOO

• Host: GitHub
• URL: https://github.com/yehengchen/ncs1-rpi
• Owner: yehengchen
• Created: 2019-04-11T04:30:21.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2019-04-16T04:35:04.000Z (about 7 years ago)
• Last Synced: 2025-09-07T08:43:42.082Z (9 months ago)
• Topics: intel, ncs, ncsdk, neural-compute-stick, rpi3
• Language: Shell
• Homepage:
• Size: 33.2 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# RPI3-NCSDK

#### Intel® Neural Compute Stick - [[Get Started]](https://software.intel.com/en-us/movidius-ncs-get-started)

#### Intel® Movidius™ Neural Network Community - [[New Discussion]](https://ncsforum.movidius.com/)

# Get Started 

### Step 1: Gather Your Equipment

__Before you start, make sure you have the following:__

* Ubuntu* 16.04 operating system, RPI3 Model B or Ubuntu VirtualBox instance

* Intel® Movidius™ Neural Compute Stick

* Internet connection to download and install the Intel® Movidius™ Software Development Kit (SDK)

 

### Step 2: Install the SDK - (Pick one Version)

__To install NCSDK 1, run these commands in a terminal window:__

    

    mkdir -p ~/workspace 

    cd ~/workspace

    git clone https://github.com/movidius/ncsdk.git

    cd ncsdk

    make install

__To install NCSDK 2.x you can use the following command to clone the ncsdk2__

	

	mkdir -p ~/workspace

	cd ~/workspace

	git clone -b ncsdk2 https://github.com/movidius/ncsdk.git

    cd ncsdk

    make install

#### if you've installed tensorflow - “INSTALL_TENSORFLOW=no”

	

	gedit ncsdk/ncsdk.conf 

	...

	MAKE_PROCS=1

	SETUPDIR=/opt/movidius

	VERBOSE=yes

	SYSTEM_INSTALL=yes

	CAFFE_FLAVOR=ssd

	CAFFE_USE_CUDA=no	

	INSTALL_TENSORFLOW=no

	INSTALL_TOOLKIT=yes

### Step 3: Test the Installation

__Run the built-in examples to test your installation. To do this, plug the compute stick into the USB port, and then run the following commands in a new terminal window:__

	  

    cd ~/workspace/ncsdk/examples/apps/hello_ncs_py

	make run

	...

	making run

	python3 hello_ncs.py;

	Hello NCS! Device opened normally.

	Goodbye NCS! Device closed normally.

	NCS device working.

	

	cd ~/workspace/ncsdk

	make examples

# How to run tensorflow and opencv on RPI 3B

#### get shell script to install package - [[install.sh]](https://github.com/yehengchen/RPI-NCS/blob/master/install.sh)[[Tensorflow - 1.8.0]](https://github.com/lhelontra/tensorflow-on-arm/releases)

	

	$ chmod 777 ./install.sh

	$ ./install.sh

	...

	$ python3

	>>> import cv2

	>>> import tensorflow

	

# NCAPPZOO Examples

### The ncappzoo is a valuable resource for NCS users and includes community developed applications and neural networks for the NCS - [[ncappzoo]](https://github.com/movidius/ncappzoo)

	git clone https://github.com/movidius/ncappzoo.git

#### NC App Zoo Repository Layout

See the README file in each of these directory or just click on the links below to explore the contents of the NC App Zoo.

* [apps](https://github.com/movidius/ncappzoo/blob/master/apps/README.md) : Applications built to use the Intel Movidius NCS. This is a great place to start in the NC App Zoo!

* [caffe](https://github.com/movidius/ncappzoo/blob/master/caffe/README.md) : Scripts to download caffe models and compile graphs for use with the NCS

* [tensorflow](https://github.com/movidius/ncappzoo/blob/master/tensorflow/README.md) : Scripts to download TensorFlow™ models and compile graphs for use with the NCS

* data : Data and scripts to download data for use with models and applications that use the NCS

### Makefile Guidance - [[MAKEFILE_GUIDANCE]](https://github.com/movidius/ncappzoo/blob/master/MAKEFILE_GUIDANCE.md)

#### Makefile required targets for ncappzoo/apps directory:

    [make help] : Display make targets and descriptions.

    [make data] : Download data (images, etc.) If no data download required create empty target.

    [make deps] : Download/Prepare/compile networks. If not needed create empty target.

    [make all] : Prepares everything needed to run the application including other projects in the repository. Should not run application, should not popup GUI.

    [make run] : Runs the application with default parameters if needed.

    [make clean] : Removes all the files in this project directory that may get created when making or running this project. Should not clean other projects in the repository.

#### Makefile required targets for ncappzoo/caffe and ncappzoo/tensorflow directories:

    [make help] : Display make targets and descriptions.

    [make all] : creates a graph file for the network and does one of compile, check, or profile on it. Should not bring up GUI, or run lengthy program.

    [make deps] : Download/Prepare networks. If not needed create empty target

    [make compile] : Run the NCSDK compiler to create a graph file.

    [make profile] : Run the NCSDK profiler to display a profile of the network

    [make check] : Run the NCSDK checker to validate the network

    [make run] : Run a simple program demonstrating use of the compiled network

    [make clean] : Removes all the files in this project directory that may get created when making or running this project. Should not clean other projects in the repository.

### Contributing to the Neural Compute App Zoo (NC App Zoo) - [[CONTRIBUTING]](https://github.com/movidius/ncappzoo/blob/master/CONTRIBUTING.md)