Integrating NVIDIA Jetson TX1 Running TensorRT into Deep Learning DataFlows with Apache MiniFi Part 1 of 4

Article

Use Case:

Ingesting sensors, images, voice and video from moving vehicles and running deep learning in the running vehicle. Transporting data and messages to remote data centers via Apache MiniFi and NiFi over Secure S2S HTTPS.

Background:

NVidia Jetson TX1 is a specialized developer kit for running a powerful GPU as an embedded device for robots, UAV and specialized platforms. I envision it's usage in field trucks for intermodal, utilities, telecommunications, delivery services, government and other industries with field vehicles.

Installation and Setup

You will need a workstation running Ubuntu 16 with enough disk space and network access. This will be to download all the software and push it over a network to your NVidia Jetson TX1. You can download Ubuntu here (https://www.ubuntu.com/download/desktop). Fortunately I had a MiniPC with 4GB of RAM that reformatted with Ubuntu to be the host PC to build my Jetson. You cannot run this from a Mac or Windows machine.

You will need a monitor, mouse and keyboard for your host machine and a set for your NVIdia Jetson.

First step boot to your NVidia Jetson and set up WiFi networking and make sure your monitor, keyboards and mouse work.

Make sure you download the latest NVidia JetPack on your host Ubuntu machine https://developer.nvidia.com/embedded/jetpack The one I used was JetPack 3.1 and that included: 64-bit Ubuntu 16.04 cuDNN 6.0 TensorRT 2.1 CUDA 8.0.

Initial login: ubuntu/ubuntu

After installation it will be nvidia/nvidia.

Please change that password, security is important and this GPU could do some serious bitcoin mining.

sudo su
apt update
apt-get install git zip unzip autoconf automake libtool curl zlib1g-dev maven swig bzip2
apt-get purge libopencv4tegra-dev libopencv4tegra
apt-get purge libopencv4tegra-repo
apt-get update
apt-get install build-essential
apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev
apt-get install libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev
apt-get install python2.7-dev
apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev
apt-get install libgtkglext1 libgtkglext1-dev
apt-get install qtbase5-dev
apt-get install libv4l-dev v4l-utils qv4l2 v4l2ucp
cd $HOME/NVIDIA-INSTALL 
./installer.sh

Downloaded and run NVidia Jetson TX1 JetPack from host Ubuntu computer ./JetPack-L4T-3.1-linux-x64.run.

This will run on the host server for probably an hour and require networking connection between the two and a few reboots.

I added a 64GB SD Card as the space on the Jetson is tiny. I would recommend adding a big SATA hard drive.

umount /dev/sdb1
mount -o umask=000 -t vfat /dev/sdb1 /media/

Turn on the fan on the Jetson, echo 255 > /sys/kernel/debug/tegra_fan/target_pwm.

Download MiniFi https://nifi.apache.org/minifi/download.html or https://hortonworks.com/downloads/#dataflow. You will need to install JDK 8.

sudo add-apt-repository ppa:webupd8team/java
sudo apt update
sudo apt install oracle-java8-installer -y
download minifi-0.2.0-bin.zip
unzip *.zip
bin/minifi.sh start

In the next part, we will classify images.

Part 2: Classifying Images with ImageNet

Part 3: Detecting Faces in Images

Part 4: Ingesting with MiniFi and NiFi

Shell Call Example

/media/nvidia/96ed93f9-7c40-4999-85ba-3eb24262d0a5/jetson-inference-master/build/aarch64/bin/detectnet-console pic-007.png outputface7.png facenet

Source Code

https://github.com/tspannhw/jetsontx1-TensorRT

References

• http://elinux.org/Jetson/Computer_Vision_Performance#Hardware_Acceleration_of_OpenCV
• https://github.com/dusty-nv/jetson-inference#system-setup
• https://github.com/dusty-nv/jetson-inference/blob/master/docs/deep-learning.md
• https://github.com/NVIDIA/DIGITS/tree/master/examples/semantic-segmentation
• https://developer.nvidia.com/tensorrt
• https://community.hortonworks.com/articles/130814/sensors-and-image-capture-and-deep-learning-analys.html
• https://github.com/dusty-nv/jetson-inference/blob/master/detectnet-console/detectnet-console.cpp
• http://elinux.org/Jetson_TX1#Jetson_TX1_Module
• http://www.jetsonhacks.com/2016/12/30/tensorflow-nvidia-jetson-tx1-development-kit/
• https://developer.nvidia.com/embedded/learn/tutorials#collapseOne
• http://www.jetsonhacks.com/2016/12/30/install-tensorflow-on-nvidia-jetson-tx1-development-kit/
• http://www.nvidia.com/object/JetsonTX1DeveloperKitSE.html?
• http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html
• https://github.com/jetsonhacks/installTensorFlowTX1/blob/master/scripts/installDependencies.sh
• https://github.com/dusty-nv/jetson-inference#system-setup
• http://docs.nvidia.com/jetpack-l4t/index.html#developertools/mobile/jetpack/l4t/3.0/jetpack_l4t_install.htm
• https://github.com/dusty-nv/jetson-inference/blob/master/data/networks/ilsvrc12_synset_words.txt
• https://github.com/tspannhw/rpi-rainbowhat/blob/master/minifi.py
• https://community.hortonworks.com/articles/130814/sensors-and-image-capture-and-deep-learning-analys.html
• https://nifi.apache.org/minifi/system-admin-guide.html