Comment #14 on issue 2130 by smi...@gmail.com: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
How now?
h[1] >>> limit(Sum(1/x, (x, 1, n))-1/n, n, oo)
Sum(1/x, (x, 1, oo))
h[2] >>> limit(Sum(1/x, (x, 1, n))-log(n), n, oo)
Traceback (most recent call last):
Comment #12 on issue 2130 by Vinzent.Steinberg: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
Well, it does not make sense. (When you return oo you loose information
about the speed of divergence.) But I think summations are not well
supported anyway.
Doesn't gruntz() ha
Comment #13 on issue 2130 by asmeurer: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
I don't like that. It is basically oo - oo (if you call .doit() for
example, it will be converted to nan). I think we should either special
case the code to return EulerGamma, or else ju
Comment #11 on issue 2130 by smi...@gmail.com: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
What about this?
>>> limit(summation(1/k, (k, 1, n))-log(n), n, oo)
-oo + Sum(1/k, (k, 1, oo))
--
You received this message because you are subscribed to the Google Groups
Comment #10 on issue 2130 by asmeurer: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
Can someone who understands sizes of infinity comment on whether that
summation should be considered a smaller infinity.
See http://en.wikipedia.org/wiki/Euler–Mascheroni_constant.
--
You
Updates:
Labels: -NeedsBetterPatch NeedsReview
Comment #9 on issue 2130 by Vinzent.Steinberg: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
(No comment was entered for this change.)
--
You received this message because you are subscribed to the Google Groups
"sympy-p
Comment #8 on issue 2130 by smi...@gmail.com: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
vks: sum is now summation
>>> limit(summation(1/k, (k, 1, n))-log(n), n, oo)
-oo
re #6: the NotImplementedError is raised for a bare Sum, but a separate
issue is how to prop
Comment #7 on issue 2130 by smi...@gmail.com: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
Perhaps the better fix is to make subs return NaN when it make a subs that
leads to a an undefined state:
(1/x).subs(x, 0) -> NaN, not 1
tan(x).subs(x, pi/2) -> NaN
What do you thi
Comment #6 on issue 2130 by asmeurer: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
I think the "NotImplementedError: Don't know how to calculate the mrv
of 'Sum(1/k, (k, 1, n))'" was the correct error, i.e., the Gruntz algorithm
doesn't know how to calculate the mrv of Su
Updates:
Status: Started
Labels: -PassedReview NeedsBetterPatch
Comment #5 on issue 2130 by Vinzent.Steinberg: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
I get this with your branch:
In [1]: limit(sum(1/k, (k, 1, n))-log(n), n, oo)
-
Comment #4 on issue 2130 by smi...@gmail.com: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
Excellent idea! Since I am working on the limit/series issues I found that
this has caused a problem with an old issue 1092. The limit can't handle
the sum so the whole thing fails
Updates:
Labels: -NeedsReview PassedReview
Comment #3 on issue 2130 by Vinzent.Steinberg: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
Thanks, this looks good and works. Let's push this in after waiting 24
hours.
--
You received this message because you are subscr
Updates:
Labels: smichr NeedsReview
Comment #2 on issue 2130 by smi...@gmail.com: wrong limit
http://code.google.com/p/sympy/issues/detail?id=2130
See [ https://github.com/sympy/sympy/pull/91 ]
--
You received this message because you are subscribed to the Google Groups
"sympy-patches"
13 matches
Mail list logo
