-
-
Notifications
You must be signed in to change notification settings - Fork 561
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
A proper random_element() method for PerfectMatchings #9890
Comments
comment:1
I forgot to click on Nathann |
comment:2
Patch updated ! On the way, I also fixed this "small problem" :
Nathann |
Attachment: trac_9890.patch.gz |
comment:3
hello, looks good to me. I have taken the opportunity to make a complete cosmetic clean-up of the file (using pep8 and pyflakes) if my review patch is ok for you, you can set a positive review |
Author: Nathann Cohen |
comment:4
Attachment: trac_9890_review.patch.gz
Okayyyyyyy ! Good to go, then By the way, how do you use pep8 and pyflakes ? Do you run them externally on files or do you have a way to use them ? Nathann |
Reviewer: Frédéric Chapoton |
comment:5
Yes, I just run them on the *.py files. Pyflakes is more important, as it finds missing imports and unused variables. Pep8 is much more for the cosmetic, but can check that raise statements are in python3 syntax and find other deprecated syntax. |
Merged: sage-5.10.beta5 |
At the moment, the method
random_element()
from PerfectMatchings computes a random matching by picking a random integer between 0 and the number of matchings on n elements, then (I presume) enumerating all the matchings according to some ordering until thek
th has been computed. This is impressively useless.By the way, I was not able to write a method to obtain the list of pairs describing the matching from an PerfectMatching object.. I don't understand how this class is written, and I have no idea why it needs to be so complicated (but it would be nice to add it to this ticket during the review, if someone gets how it works).
The method random_element (and also an_element) both raise an exception when the set of elements is EMPTY. I also fixed the doctests.
(I don't even get why you can build a PerfectMatchings class on an odd number of elements in the first place)
I expect this ticket could be heavily modified during review, but there is a problem with these classes at the moment.
Nathann
CC: @nthiery @hivert
Component: combinatorics
Author: Nathann Cohen
Reviewer: Frédéric Chapoton
Merged: sage-5.10.beta5
Issue created by migration from https://trac.sagemath.org/ticket/9890
The text was updated successfully, but these errors were encountered: