[sage-devel] Testing new graph algorithms

I've been working on some new algorithms for computing some properties (eccentricity, diameter, radius, ...) of (undirected) graphs and I'm looking for families to test these on. I've written a reasonable efficient C extension which implements these algorithms and so I've started comparing

[sage-devel] Re: Polyhedron.integral_points_count() gives inconsistent answers on slices

Great suggestion. Could you ask polymake to compute what the number of lattice points in each x_0 = k (for k = 1, ..., 100) slice? Does it match: [84, 84, 84, 81, 81, 81, 75, 75, 75, 66, 66, 66, 54, 54, 54, 39, 39, 39, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,

[sage-devel] Polyhedron.integral_points_count() gives inconsistent answers on slices

I want to count the number of integral points inside of a bounded polyhedron. The polyhedron is defined by the following system of equations: > eqns = [ [-100, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [0, -1, 0, 0, 0, 0, -1, 1, 1, 0, 0, 0, 0], [0, 0, 0, 0, 0, -1, 0, 0, 1, 1, -1, 0, 0], [0, 0, 0, 0,

[sage-devel] Re: Polyhedron.integral_points_count() gives inconsistent answers on slices

Here is the full CoCalc output when I run "sage error.py" in a terminal, either via the web interface or via sshing into the machine: Traceback (most recent call last): File "error.py", line 42, in print(P.integral_points_count(preprocess=False)) File

Re: [sage-devel] Re: Polyhedron.integral_points_count() gives inconsistent answers on slices

Thanks for the suggestion. That might be the case but if I run "/usr/bin/time -v sage error.py" then I get Command exited with non-zero status 1 Command being timed: "sage error.py" User time (seconds): 2.22 System time (seconds): 0.67 Percent of CPU this job got: 86% Elapsed (wall clock) time

[sage-devel] Re: Polyhedron.integral_points_count() gives inconsistent answers on slices

Excellent! Adding the option prevents the crash. So on CoCalc (and locally) I can do: print(P.integral_points_count(preprocess=False, triangulation='cddlib')) And it prints out 19958. Without this extra option this example crashes on CoCalc but runs fine locally. What is this this option

[sage-devel] Re: Polyhedron.integral_points_count() gives inconsistent answers on slices

Ah, I see. Is it possible that by doing "sage -i latte_int" I have installed a different version of 4ti2 than the one that CoCalc uses? I'm still not sure why (without this option) I see different behaviours on my machine and on CoCalc. If it matters I have Sage 8.0 running on Mac OS Sierra

[sage-devel] Re: Polyhedron.integral_points_count() gives inconsistent answers on slices

So when k = 19, sage.interfaces.latte.count and P.integral_points_count() give different answers. This seems to be the cause of the difference. Here is some more code showing the difference. -- You received

[sage-devel] Re: Polyhedron.integral_points_count() gives inconsistent answers on slices

Thanks! I'm not sure if this is the right place to add this but the attached example also causes LattE to crash, but I only see this behaviour when I run this code on CoCalc. When I run the script on my local machine with latte_int installed it completes and prints out 19958. Surely this

[sage-devel] Missing Sage 8.2 binaries

The main sagemath.org page lists Sage 8.2 as the latest version to download and the source and git links appear to be tagged as sage-8.2 as of three days ago. However within the main download page, there don't appear to be any precompiled versions of 8.2 available. For example here:

[sage-devel] Re: Missing Sage 8.2 binaries

Excellent, things have started appearing now. Is the 8.2 changelog (http://www.sagemath.org/changelogs/index.html) also missing though? On Tuesday, May 8, 2018 at 11:05:54 PM UTC+1, Volker Braun wrote: > > Binaries take a couple of days to build. On the way to the mirrors now... > > > > On

[sage-devel] Typo in trac instructions

I think there is a minor typo in the *To get an account* instructions at https://trac.sagemath.org/. It says that: the default username is the first letter of the surname followed by the > name (i.e. mjohnson for Mike Johnson or tlei for Tan Lei). Should this not say: the default username is

[sage-devel] _integer_ for QQbar and QQ raise different errors

The coersion sage: ZZ(1/2) fails and (correctly) raises a TypeError while sage: ZZ(QQbar(sqrt(17))) fails and (incorrectly) raises a ValueError. This has several knock on effects, in particular vector(ZZ, [list of algebraic numbers]) all of whoms try / except blocks are set up to catch

[sage-devel] Making Integer and Rational compatible with Python Fraction

During Sage Days 100, I experienced an issue with Sage's Integers and Rational not being compatible with Python's built in Fraction class. This made me unable to run my Python code within Sage and so I raised ticket 28234 . For example, in Sage 8.8: