On Tue, Mar 12, 2013 at 7:37 AM, ThanhVu Nguyen
<[email protected]> wrote:
> That sounds very interesting.  Can you share some details on what you did
> that makes it much faster,  even when using pure Python coefficients ?  why
> is that the current solve in sympy is so slow comparing to the one in Sage ?
>
> How do I reproduce what you did after obtaining the patch, assuming the
> input is my original list of 165 equations as given in
> http://pastebin.com/cE87W9m3 ?  What is eqs_165x165() in the code eqs =
> eqs_165x165()   ?   Also, how long usually does it take for such a patch to
> get into the main sympy branch ?

Normal patch just a few hours or days. This patch is quite big and it
might need some polishing.
If you want to help us out, please try the code out:

https://github.com/sympy/sympy/pull/1840

and provide us with feedback. That would be very helpful.

Just yesterday Mateusz explained to me a bit how it works over a beer.
I am very excited about this new code.
It's using a better polynomial representation and then Mateusz did
some profiling and was able to optimize some slow parts. It's quite
amazing that this is still pure Python (optionally with gmpy).

Ondrej

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" 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/sympy?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to