Home    About    Documentation    Install    Newsline    Links    Feedback    Appeal    Guestbook

since 2007/10/24

OpenOpt Installation

---

1. If you have either problems with installation or bug reports you can use scipy-user mail list (this one is connected to scipy-user google group)

---

Note! If you have numpy.__version__ > 1.1, using OpenOpt from latest tarball or subversion repository is recommended.

---

2. Choose one of the following:

• get OpenOpt from svn (Subversion - a version control system, MS VSS or cvs equivalent):
  

svn co http://svn.scipy.org/svn/scikits/trunk/openopt openopt

for Windows using subversion client like TortoiseSVN is recommended

• openopt.tar.bz2 (~ 200 Kb) - tarball with latest changes (mentioned in newsline), for last modification date see bottom of the page
• scikits.zip (~ 10 Mb) - zip-file with all scikits automatically generated by Trac system from subversion repository. The link has been taken from the page.
• One of stable releases: openopt0.19.tar.bz2 (2008-September-15), openopt0.18.tar.gz (2008-June-15), openopt0.17.tar.gz (2008-March-15), openopt0.15.tar.gz (2007-Dec-15)
• Try installation from the Linux deb file made by Yaroslav Halchenko (you can try using aptitude, apt-get, adept manager etc).

---

3. Ensure you have Python v. 2.4 or 2.5 (there is no OO for Python 2.6 and there will be for 3.0 only) and NumPy (ver. >= 1.1.0 is preferred, some users inform of errors due to v 1.0.5, 1.0.4 or even more obsolete) installed (as a rule, Linux systems already have Python installed, and "python-numpy" package is available in software channels, for example use "apt-get install python-numpy" or "aptitude install python-numpy" for Debian or *UBUNTU Linux).

• Some Python distributions have no setuptools included, then you need either install it by yourself (for Linux users using software channels is recommended, for example "[sudo] aptitude install python-setuptools"; or, if any problems will be encountered, just add the directory containing "scikits" folder to your PYTHONPATH, hence further steps are not required).
• Linux users, using software update channels to get numpy, as well as OpenOpt-related software (like cvxopt, setuptools, matplotlib, scipy, with all required and optional dependencies) can yield a little bit obsolete version, but could greatly economy your time and efforts.
  
• Windows users may be interested in EPD ("Enthought Python Distribution") - it already includes numpy, scipy, matplotlib and much more.
• Linux, Windows, Mac OS X users may be interested in SAGE (license: GPL) - "viable free open source alternative to Magma, Maple, Mathematica, and Matlab" - it contains Python + numpy + scipy + cvxopt + much more (BTW, 1/2 there is 0.5, not CPython- or C-like zero).
• Python(x,y) is new Eclipse-based scientific Python distribution, currently available for Windows, but developers promise Linux and MacOS versions in future (check the site, mb it's already available).

---

4. Run installation:

• Linux:
  [sudo] python setup.py install
  

• Windows: you should run setup.py and choose "install" option in menu.
• You could also run setup.py from a filemanager (or python setup.py from terminal) and then chose option 2 (install) and (optional) change installation path to another (however, that one should be present in PYTHONPATH)

---

5. To ensure OpenOpt has been installed successfully you could run .../examples/nlp_1.py
Found OpenOpt useful?
Please mention it in OpenOpt guestbook and / or users page

---

Update OpenOpt

# type in scikits.openopt directory:
svn up .
[sudo] python setup.py install
# (or see the remark above about adding "scikits" folder to Python path)

---

Using docstrings

# type in Python interpreter:
from scikits.openopt import NLP, QP, NSP # etc
# and see help(NLP), help(QP), etc
# or see documentation online

---

Downloads

openopt.tar.bz2 contains latest changes (mentioned in newsline)
openopt0.15.tar.gz, openopt0.17.tar.gz, openopt0.18.tar.gz are stable releases