Skip to content

Latest commit

 

History

History
42 lines (41 loc) · 2.45 KB

build-host-dnndk.md

File metadata and controls

42 lines (41 loc) · 2.45 KB
Raw

Edge AI Tutorials

Zynq 7000 DPU TRD

This Tutorial will help you build your dnndk kit in host

  • Your linux version should be 14.04 or 16.04, and you should have python2.7 with pip installed if you want to use caffe.

  • Download dnndk and extract. This guide will use dnndk_v3.0. Download our new install.sh, replace the install.sh in xilinx_dnndk_v3.0/host_x86/ with ours. Then you can run ./install PynqZ2, please pay attention that if you use Anaconda environment, you should run this command after conda activate decent.

  • Installing the GPU Platform Software

    The current DNNDK release can be used on the X86 host machine with or without GPU. With GPU support, DECENT is able to run faster.
    If GPU is available in the X86 host machine, install the necessary GPU platform software in accordance with your GPU product documentation. Ensure all versions are compatible with the version of DNNDK.
    For version information, please refer to this

  • Using Caffe

    1. Installing Dependent Libraries:
    apt-get install -y --force-yes build-essential autoconf libtool libopenblasdev libgflags-dev libgoogle-glog-dev libopencv-dev protobuf-compiler libleveldbdev liblmdb-dev libhdf5-dev libsnappy-dev libboost-all-dev libssl-dev
    1. Install Caffe, please refer to Caffe official website
    2. Change the $HOME/.bashrc:
      Add two lines:
    export PYTHONPATH=/home/(your name)/caffe/python:$PYTHONPATH  
export LD_LIBRARY_PATH=/home/(your name))/caffe/.build_release/lib:$LD_LIBRARY_PATH
    Then source .bashrc

  • Using tensorflow

    1. Install Anaconda Refer to this to install conda.
    2. Install tensorflow
    conda create -n decent pip python=3.6
source activate decent
(decent)$ cd $YourDir/xilinx_dnndk_v3.0/host_x86/decent-tf/ubuntu$YourLinuxVersion/
(decent)$ pip install ./tensorflow* # Select the right installation package for your environment
(decent)$ pip install numpy opencv-python sklearn scipy progressbar2 keras==2.2.0