Difference between revisions of "OSX"
From ACL@NCU
Line 61: | Line 61: | ||
This is because when you import mvpa2 under the pymvpa source code directory (where you make it), python takes the mvpa-subdirectory as the place from where is imports the mvpa-package. This won't work, at least for SVM / SMLR, that use C-code bindings that have to be compiled ([https://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/2011q1/001571.html source]). | This is because when you import mvpa2 under the pymvpa source code directory (where you make it), python takes the mvpa-subdirectory as the place from where is imports the mvpa-package. This won't work, at least for SVM / SMLR, that use C-code bindings that have to be compiled ([https://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/2011q1/001571.html source]). | ||
=[http://blog.manbolo.com/2014/09/27/use-python-effectively-on-os-x Use python effectively in OSX]= |
Revision as of 07:30, 3 June 2016
System
Install Homebrew on OS earlier than 10.9
ruby -e "$(curl -fsSkL raw.github.com/mistydemeo/tigerbrew/go/install)" (in .bash_profile) export PATH=/usr/local/sbin:/usr/local/bin:$PATH
Install pymvpa (adapted from Moffet's Crash Log)
1) Install Homebrew: ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install" 2) mkdir /Users/Shared/PyMVPA; cd /Users/Shared/ 3) sudo easy_install pip 4) pip install virtualenv 5) virtualenv PyMVPA 6) cd PyMVPA 7) . bin/activate 8) pip install numpy 9) pip install scipy 10) pip install nibabel 11) pip install ipython'[all]' 12) pip install scikit-learn 13) pip install matplotlib 14) brew install swig 15) Grab the PyMVPA source (and place into your virtualenv folder): git clone git://github.com/PyMVPA/PyMVPA.git 16)Install libsvm: brew install gnuplot brew install qt4 brew install libsvm 17)Install PyMVPA: make 3rd python setup.py build_ext –with-libsvm python setup.py install –with-libsvm 18) Download tutorial data at: http://data.pymvpa.org/datasets/haxby2001/ 19) Unzip and Place data in /Users/Shared/PyMVPA/Tutorial_Data
And remember that you should cd to another directory when calling the test command:
from mvpa2.tutorial_suite import *
Otherwise you will see the following complaint:
WARNING: Failed to load fast implementation of SMLR. May be you forgotten to build it. We will use much slower pure-Python version. Original exception was dlopen(mvpa2/clfs/libsmlrc/smlrc.so, 6): image not found * Please note: warnings are printed only once, but underlying problem might occur many times * WARNING: SMLR: C implementation is not available. Using pure Python one
This is because when you import mvpa2 under the pymvpa source code directory (where you make it), python takes the mvpa-subdirectory as the place from where is imports the mvpa-package. This won't work, at least for SVM / SMLR, that use C-code bindings that have to be compiled (source).