On Mar 2, 12:09 pm, Robert Miller <rlmills...@gmail.com> wrote:
> My timings are showing different results than yours.
> ...
> What platform were you running your tests on?

Hey,

Just got back from a day of fun in the snow.

I can confirm your timing comparisons, when I run your code
on my machine.  Makes me wonder if I'm doing anything wrong
in my code.  The exact code I used to produce my originally
posted timing results is below.  My machine is a MacBook Pro
running Sage 3.3 on OSX 10.4.11.

I haven't tried your patch, but I'm looking forward to it.

Mark


--------------------- NetworkX ----------------------
-- The interpreter box at the top is set to Python --

import networkx as nx
import time as time

import urllib2
f = urllib2.urlopen('http://cs.anu.edu.au/~bdm/data/graph7.g6')
graph_strings = f.readlines()
f.close()
graphs = [nx.parse_graph6(s.strip()) for s in graph_strings]

t = time.clock()
result = False
for i in xrange(0, len(graphs)-1):
    for j in xrange(i+1, len(graphs)-1):
        if nx.isomorphvf2.GraphMatcher(graphs[i],graphs
[j]).is_isomorphic():
            result=True
            break
        if result:
            break
result
time.clock()-t

CPU time: 27.39 s


----------------- Sage code --------------
import urllib2
f = urllib2.urlopen('http://cs.anu.edu.au/~bdm/data/graph7.g6')  # Or
graph7.g6
graph_strings = f.readlines()
f.close()
some_graphs = [Graph(graph_string) for graph_string in graph_strings]
# Switch to the new c_graph implementation.
better_graphs = [g.copy(implementation='c_graph', sparse=False) for g
in some_graphs]
n = len(some_graphs)

%time
result = false
for i in xrange(0,n):
    for j in xrange(i+1, n):
        if better_graphs[i].is_isomorphic(better_graphs[j]):
            result = true
            break
    if result:
        break
result

CPU time: 38.12 s

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---

Reply via email to