Advanced Computing Platform for Theoretical Physics

README.md 1.24 KB
Newer Older
Justin Bayer's avatar
Justin Bayer committed
1
2
3
climin
------

Justin Bayer's avatar
Justin Bayer committed
4
climin is a Python package for optimization, heavily biased to machine learning
Justin Bayer's avatar
typo    
Justin Bayer committed
5
scenarios distributed under the BSD 3-clause license. It works on top of numpy
Justin Bayer's avatar
Justin Bayer committed
6
and (partially) gnumpy.
Justin Bayer's avatar
Justin Bayer committed
7

Justin Bayer's avatar
Justin Bayer committed
8
The project was started in winter 2011 by Christian Osendorfer and Justin Bayer.
Justin Bayer's avatar
Justin Bayer committed
9
Since then, Sarah Diot-Girard, Thomas Rueckstiess and Sebastian Urban have 
Christian's avatar
Christian committed
10
11
12
contributed. If you use climin in your (academic) work, please cite as 
(tech report is in preparation):

13
> J. Bayer and C. Osendorfer and S. Diot-Girard and T. Rückstiess and Sebastian Urban. climin - A pythonic framework for gradient-based function optimization. TUM Tech Report. 2016. http://github.com/BRML/climin
Justin Bayer's avatar
Justin Bayer committed
14
15
16
17
18
19


Important links
---------------

 - Official repository of source: http://github.com/BRML/climin
Justin Bayer's avatar
Justin Bayer committed
20
21
 - Documentation: http://climin.readthedocs.org
 - Mailing list: climin@librelist.com (archive: http://librelist.com/browser/climin/)
Justin Bayer's avatar
Justin Bayer committed
22
23
24
25
26


Dependencies
------------

27
The software is tested under Python 2.7 with numpy 1.10.4, scipy 0.17. The tests
Justin Bayer's avatar
Justin Bayer committed
28
29
30
31
32
33
34
35
36
37
38
39
40
41
are run with nosetests.


Installation
------------

Use git to clone the official repository; then run `pip install --user -e .`
in the clone to intall in your local user space.


Testing
-------

From the download directory run ``nosetests tests/``.