开源软件名称（OpenSource Name）：

开源软件地址(OpenSource Url)：

开源编程语言(OpenSource Language)：

开源软件介绍(OpenSource Introduction)：

Damascene

GPU gPb image segmentation.

Basic install task list on ubuntu 16.04:

• Install cuda 8.0
• Install gcc-5
• Install Lapack
• Run cmake and build libdamascene.so
• Build the python bindings
• ... start GPU segmenting

Install CUDA

This assumes you have installed the proprietary nvidia drivers for ubuntu. (although I think the cuda install process may take care of that).

wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_8.0.61-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1604_8.0.61-1_amd64.deb
sudo apt-get update
sudo sudo apt-get install cuda

The above download link is for cuda 8.0 on ubuntu 16.04. Other versions are available here: https://developer.nvidia.com/cuda-downloads

Install gcc-5

sudo apt-get install gcc-5 g++-5

Note. Ubuntu 16.04 installs gcc-6 as its default gcc compiler. Cuda 8.0 requires that a max version of 5 is used. Later in the process we will set an environment variable to instruct cmake is use gcc-5

Install Lapack

sudo apt-get install liblapack-dev

Run cmake and install libdamascene

Start from the root of the checked out directory.

mkdir build
cd build
CC=/usr/bin/gcc-5 cmake .. -DACMLPATH=/usr/local/acml
make -j`cat /proc/cpuinfo | grep MHz | wc -l`
sudo make install

Build and install the python bindings

Start from the root of the checked out directory

sudo apt-get install python-pip python-numpy
sudo pip install cython
cd bindings
sudo python setup.py install

...start GPU segmenting

sudo apt-get install python-imaging
sudo apt-get install python-tk
sudo pip install matplotlib
python example.py