On Tue, Feb 19, 2013 at 10:38 PM, Leonard P <leonard14...@gmail.com> wrote:
> On Tue, Feb 5, 2013 at 6:23 PM, John Benediktsson <mrj...@gmail.com>wrote:
>
>> I think you'd want virtual sequences as well, so getrow wouldn't
>> necessary copy data. Doing the API right would involve walking through
>> various numerical programming examples. We'd love any contributions you'd
>> want to make!
>>
>> Thanks,
>> John.
>>
>
> A neat idea might be to implement two versions of the Laplace expansion...
> one as BFS and one as DFS.
>
> The BFS version would do the regular expansion by copying data.
>
> The DFS version could be a virtual sequence. Each element in the sequence
> is a leaf node in the tree of the Laplace expansion.
>
> The DFS version would save space, at the expense of redundant calculations.
>
Er, cancel that. Realized that with DFS it would be impossible to keep
track of the path to the nth leaf node.
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb
_______________________________________________
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk
