I was playing with more complex functions and found a memory leak. I have
been trying to figure out where is it, and here is result so far:
def foo(L, s):
return sum( [L.mobius_function(y, L.top())/prod(s[x] for x in
L.interval(L.bottom(), y)) for y in s.keys()] )
S.<x,y>=QQ[]
def bar(L, x):
x2=copy(x)
x2[9]=var('x')
x2[10]=var('y')
x2[7]=1
P=S(foo(L, x2)*var('x')*var('y').full_simplify())
P=P*lcm([x.denominator() for x in P.coefficients()])
a=P.coefficients()[0]
b=P.coefficients()[1]
c=P.coefficients()[2]
d=-P.coefficients()[3]
for t in divisors(a*d+b*c):
if (t-c)/a in ZZ and (t-c)/a > 0 and ((a*d+b*c)/t-b)/a in ZZ and
((a*d+b*c)/t-b)/a > 0:
print "Foo"
L=Poset(([0,1,2,3,4,5,6,7],
[[0,1],[0,2],[0,3],[1,4],[3,6],[1,5],[3,5],[2,4],[2,6],[4,7],[5,7],[6,7],[1,8],[8,7],[1,9],[9,7],[1,10],[10,7]
]))
print sage.misc.getusage.get_memory_usage()
for n in range(1,100):
bar(L, {0: 1, 1: n, 2: n, 3: 1, 4: 24, 5: 2, 6: n, 8: 25})
print sage.misc.getusage.get_memory_usage()
for n in range(1,100):
bar(L, {0: 1, 1: 3, 2: 5, 3: 67, 4: n, 5: n, 6: n, 8: 54})
print sage.misc.getusage.get_memory_usage()
This outputs
940.765625
989.78125
991.875
If I change, say, divisors(a*d+b*c) to divisors(123), it will output
989.78125 twice. So what is happening? Does Python somehow save variables
defined inside function (x2) if they are used somewhere (x2 is used for
calculating value of P, P is used for getting a,...,d and they are used as
argument for divisors().)
On the other hand, when I change range(1,100) to range(1,1000) this
outputs
940.765625
998.4375
1012.27734375
--
Jori Mäntysalo
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
