documentation is less
ambiguous.
If you still see issues room for improvement don't hesitate to share your
insights.
Cheers,
Denis
On Tue, Sep 24, 2013 at 3:56 PM, Shishir Pandey wrote:
> On 24-09-2013 18:03, Denis-Alexander Engemann wrote:
>
> Hi Shishir,
>
>
Hi Shishir,
please note that the ProbabilisiticPCA got recently refactored which lead to
API changes and improved documentation + examples.
https://github.com/scikit-learn/scikit-learn/pull/2404
Did you take these changes into account? If not it would be great to know which
exmample / reduplic
Hi Safi,
On Sun, Sep 8, 2013 at 1:58 PM, Safi Ullah Marwat
mailto:[email protected]>> wrote:
Thank you Gael,
I actually want to know the distance between the entries of existing mixing
matrix and the newly calculated ones (the questioned thing)
As Gael you compute your `A` only once.
Dear scikit-learners,
during the last sprint we've spotted an efficiency issue with the numpy.dot for
numpy versions < 1.8. Apparently, the dot allocates additional copies in order
to deliver appropriate input to the underlying BLAS gemm function which expects
Fortran contiguous memory layout f
Hehe..
Just for logging purposes:
I saw theses old SDKs under /Developer which were conflicting with the recent
10.8 SDK from /Applications
So
sudo /Developer/Library/uninstall-devtools --mode=all
did the trick.
Denis
---
Ok... I think I got it. Xcode path is heavily messed up...
Thanks though (sometimes just posting it helps to get it ;-).
D
Hi Folks, @agramfort & @amueller -- unfortunately getting my 32/64 bit issue
fixed before getting back to #1350 was not a matter of a few minutes. Instead I
found myself somewhere here, related to this issue.
I gave up my macports a few days a ago replaced with 32bit and then 64bit EPD,
since th
