#10792: Upgrade numpy to 1.5.1
-------------------------------------------------------------------------+--
Reporter: jason |
Owner: tbd
Type: task |
Status: needs_work
Priority: major |
Milestone: sage-4.7
Component: packages |
Keywords:
Author: Jason Grout |
Upstream: N/A
Reviewer: David Kirkby |
Merged:
Work_issues: doctest changes for patch, verify a couple analytic cases |
-------------------------------------------------------------------------+--
Comment(by kcrisman):
Jason's patch is correct. Positive review on that part.
Here's another thought.
{{{
self.cps = np.zeros([self.B, N], dtype=np.complex128)
# Find the points on the boundaries and their derivatives.
for k in xrange(self.B):
for i in xrange(N):
self.cps[k, i] = np.complex(fs[k](self.tk[i]))
}}}
So when the original code does
{{{
temp = np.zeros([self.B, 1])
for k in xrange(self.B):
temp[k, 0] = self.cps[k, 0]
}}}
then I would say the problem is in taking np.zeros and assigning complex
numbers to that temp array, which is defined to be real a priori. Maybe
that is the change in how Numpy did things - it now complains?
My point being that if we change
{{{
- temp = np.zeros([self.B, 1])
+ temp = np.zeros([self.B, 1],dtype=np.complex128)
}}}
and that also removes the errors, and we get the same doctest changes,
then we can be sure that the problem is simply that this is a really hard
numerical thing to get right, and Ethan should be commended for even
getting within epsilon.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/10792#comment:29>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
