[SciPy-user] random numbers in scipy

Giovanni Samaey giovanni.samaey@cs.kuleuven...
Wed Apr 25 12:30:26 CDT 2007

Andrew Straw wrote:
> Giovanni Samaey wrote:
>> I now understand the situation that scipy.stats builds upon 
>> numpy.random, and adds stuff without claiming to expose all that is in 
>> numpy.random (such as the seeding).  Thanks for helping !
>> I repeat that I am willing to document this, if this effort will come in 
>> an "authorative" documentation (ie at the right place) and will be 
>> properly reviewed. 
>> That part of my previous message has not been answered.
> Gael already answered if you want to make a wiki page, which is one of 
> the "authoritative" means of documentation we have. Another, probably 
> slightly more so, is the docstrings. Your efforts in this regard will 
> certainly be reviewed and incorporated if acceptable to a developer if 
> you make a patch against the svn version of scipy and create a Trac 
> ticket containing the patch.
Thanks:  I will go through the docstrings that I find for other numpy 
modules and add one for numpy.random in the same style.
Since numpy.random is, strictly speaking, a numpy package and not a 
scipy package, I don't know where I should put it on the wiki.
Can I just create an account to make a trac ticket or should I ask for one?

This being said: don't be surprised if it takes a few weeks...  I put it 
on my to-do list and it will show up when I am able to do this.
But I am making a hard commitment for myself and towards you guys.  It 
is my pleasure to help you create documentation if it only involves 
translating loose emails and experience into a proper docstring :-)



More information about the SciPy-user mailing list