These ongoing discussions about Numpy and matrices versus arrays might
be of interest:
http://groups.google.com/group/Numpy-discussion/browse_thread/thread/fe349bb7f1a1809a
http://groups.google.com/group/Numpy-discussion/browse_thread/thread/63441776acd05034
Joachim
On May 11, 8:27 am, mabshoff
On May 11, 6:39 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Sat, May 10, 2008 at 9:27 PM, Jason Grout
>
>
>
> <[EMAIL PROTECTED]> wrote:
>
> > [EMAIL PROTECTED] wrote:
>
> >> On Sat, 10 May 2008, Jason Grout wrote:
>
> >>> Currently CDF and RDF matrices wrap GSL matrices and use GSL algo
On Sat, May 10, 2008 at 9:27 PM, Jason Grout
<[EMAIL PROTECTED]> wrote:
>
> [EMAIL PROTECTED] wrote:
>>
>>
>>
>> On Sat, 10 May 2008, Jason Grout wrote:
>>
>>> Currently CDF and RDF matrices wrap GSL matrices and use GSL algorithms
>>> for part of the computations. After talking with a few lead d
[EMAIL PROTECTED] wrote:
>
>
>
> On Sat, 10 May 2008, Jason Grout wrote:
>
>> Currently CDF and RDF matrices wrap GSL matrices and use GSL algorithms
>> for part of the computations. After talking with a few lead developers
>> on IRC, it seems that the consensus is that numpy is generally bet
On Sat, 10 May 2008, Jason Grout wrote:
>
> Currently CDF and RDF matrices wrap GSL matrices and use GSL algorithms
> for part of the computations. After talking with a few lead developers
> on IRC, it seems that the consensus is that numpy is generally better
> and has a much, much stronger
Yi Qiang wrote:
> Are there going to be any API incompatibilities going to be introduced
> by this move? That is, will existing code written against RDF and CDF
> matrices still "just work"?
My hope is that we will just replace the backend and the underlying code
and the user-interface function
Are there going to be any API incompatibilities going to be introduced
by this move? That is, will existing code written against RDF and CDF
matrices still "just work"?
On Sat, May 10, 2008 at 6:41 PM, Jason Grout
<[EMAIL PROTECTED]> wrote:
>
> Currently CDF and RDF matrices wrap GSL matrices and