Setup

General

$ mkvirtualenv ctree
$ pip install -r requirements.txt
$ pip install nose
$ workon ctree
$ nosetests
......................................................................
.......................S.....S.S.S....................................
......................................................................
................................................SS....................
.......................S........................
----------------------------------------------------------------------
Ran 328 tests in 3.315s

OK (SKIP=7)

Millennium Machines

NOTE: Generate an ssh key and add it to your github account (https://help.github.com/articles/generating-ssh-keys/) or use the https repo url (https://github.com/ucb-sejits/ctree.git) for the git clone step.

$ cd /scratch
$ mkdir personal_folder && cd personal_folder
$ wget http://repo.continuum.io/miniconda/Miniconda-3.8.3-Linux-x86_64.sh
$ sh Miniconda-3.8.3-Linux-x86_64.sh  
# Be sure to set the install directory to your scratch folder (i.e. /scratch/personal_folder/miniconda)
# Also be sure to append the install directory to your path
$ exec bash
$ conda install pip nose numpy
$ git clone [email protected]:ucb-sejits/ctree.git && cd ctree
$ pip install -r requirements.txt
$ nosetests

If you see this error

: CL/cl.h: No such file or directory
#include <CL/cl.h>
                  ^
compilation terminated.

You need to add the location of the OpenCL header to your include path, i.e (for a19)

export C_INCLUDE_PATH=/usr/local/cuda/include:$C_INCLUDE_PATH

You can put this in your .bashrc and source it when you log in.