Skip to content

BRAINSia/NAMICExternalProjects

Repository files navigation

NAMICExternalProjects

A superbuild infrastructure to build all superbuild structures

Ash nazg durbatulûk, 
ash nazg gimbatul,
Ash nazg thrakatulûk 
agh burzum-ishi krimpatul.
                - Sauron, the Abhorred

This project acts as a template to model building all other SuperBuild projects around.


Directory layout

NAMICExternalProjects/

  • CMakeLists.txt

    A file that acts as a top level switch, it runs the building of pre-requisite ExternalProjects defined in SuperBuild.cmake (where ${MyProject} is last ExternalProject to be built). This file is processed by CMake 2 times: the first time processes SuperBuild.cmake and the second time processes ${MyProject}.cmake

  • Common.cmake

    Included in SuperBuild.cmake and ${MYProject}.cmake for common conditional compilation options to be set. These options must be passed to all ExternalProjects (including ${MYProject})

  • SuperBuild.cmake

    Infrastructure to build all the pre-requisite packages where ${MYProject} is the last one listed

  • ${MYProject}.cmake

    Standard cmake build instructions for ${MYProject}

  • SuperBuild/

    A directory full of External_${extProjName}.cmake files defining how to build external dependencies.

  • CMake/

    A directory of support files.

Build Instruction

  1. Install Anaconda Python Environment: https://www.continuum.io/downloads

    (Nipype/BRAINS Auto Workup dependes on python 2.7v)

    $ ${conda_dir}/conda create --name namicAnacondaEnv python=2.7
    export ${conda_dir}:PATH # Add conda build on your PATH. 
    source activate namicAnacondaEnv
  2. Install SimpleITK Against above Anaconda Python Environment: http://www.itk.org/Wiki/SimpleITK/GettingStarted

    $ conda install -c https://conda.binstar.org/simpleitk SimpleITK
  3. Install NiPype against above Anaconda Python Environment: http://nipy.org/nipype/users/install.html

    $ pip install nipype

    Please check out following if you encounter any issues at this stage.

    BLAS/LAPACK for OSX:

    See https://www.continuum.io/blog/developer/mkl-optimizations-anaconda for blas/lapack if not installed.

    $ pip install mkl

    Scipy Issue for Mac/Linux:

    Failed building wheel for scipy

    $ conda install numpy 
    $ conda install scipy
    $ pip install nipype

    If that does not work try:

    Scipy Issue for Linux:

    Failed building wheel for scipy

    See http://stackoverflow.com/questions/24353267/build-wheel-for-a-package-like-scipy-lacking-dependency-declaration

    $ pip wheel numpy
    $ pip install nipype
  4. TBB Build: https://www.threadingbuildingblocks.org/download

    Download source code

     $ cd "tbb44_20151115oss"
     $ make
     $ mv macos_intel64_clang_cc4.2.1_os10.10.4_release/ ../lib
     
     $ cd Namic_build_Dir
     $ cmake -DCMAKE_CXX_STANDARD:STRING=11 -DTBB_ROOT:PATH=/Shared/sinapse/scratch/eunyokim/src/tbb44_20151115oss  ../NAMICExternalProjects  -DCMAKE_OSX_DEPLOYMENT_TARGET:STRING=10.9
  5. Build Namic against all above:

$ git clone https://github.com/BRAINSia/NAMICExternalProjects.git
$ mkdir ${build_dir_name}
$ cd ${build_dir_name}
$ ccmake ../NAMICExternalProjects
$ make
  1. Additional Package Installation for BRAINS Auto Workup:
$pip install docopt
$pip install pydas
$pip install joblib
$pip install dipy
$pip install future
$pip install ipython
$pip install simplejson
  1. Deactivate Anaconda environment when finished:
$ source deactivate