Building for raspberry pi or Jetson Nano
bash pi_build.sh using this helper script one can cross build libnd4j and dl4j with arm COMPUTE LIBRARY . it will download cross compiler and arm compute library.
options
value
description
-a or --arch
arm32
cross compiles for pi/linux 32bit
-a or --arch
arm64
cross compiles for pi/linux 64bit
-a or --arch
android-arm
cross compiles for android 32bit
-a or --arch
android-arm64
cross compiles for android 64bit
-a or --arch
jetson-arm64
cross compiles for jetson nano 64bit
-m or --mvn
if provided will build dl4j using maven
example: bash pi_build.sh --arch android-arm64 --mvn
to change version of the arm COMPUTE LIBRARY modify this line in the script
1
ARMCOMPUTE_TAG=v20.05
Copied!

old one

Please follow following instructions to build nd4j for raspberry PI:
    1.
    download cross compilation tools for Raspberry PI
    1
    $ apt-get/yum install git cmake
    2
    (You may substitute any path you prefer instead of $HOME/raspberrypi in the following two steps)
    3
    $ mkdir $HOME/raspberrypi
    4
    $ export RPI_HOME=$HOME/raspberrypi
    5
    $ cd $RPI_HOME
    6
    $ git clone git://github.com/raspberrypi/tools.git
    7
    $ export PATH=$PATH:$RPI_HOME/tools/arm-bcm2708/arm-rpi-4.9.3-linux-gnueabihf/bin
    Copied!
    2.
    download deeplearning4j:
    1
    $ cd $HOME
    2
    $ git clone https://github.com/eclipse/deeplearning4j.git
    Copied!
    3.
    build libnd4j:
    1
    $ cd deeplearning4j/libnd4j
    2
    $ ./buildnativeoperations.sh -o linux-armhf
    Copied!
    4.
    build nd4j
    1
    $ export LIBND4J_HOME=<pathTond4JNI>
    2
    $ cd $HOME/deeplearning4j/nd4j
    3
    $ mvn clean install -Djavacpp.platform=linux-armhf -Djavacpp.platform.compiler=$HOME/raspberrypi/tools/arm-bcm2708/arm-rpi-4.9.3-linux-gnueabihf/bin/arm-linux-gnueabihf-g++ -DskipTests -Dmaven.javadoc.skip=true -pl '!:nd4j-cuda-9.1,!:nd4j-cuda-9.1-platform,!:nd4j-tests'
    Copied!
Export as PDF
Copy link
Contents
old one