Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

numpy.linalg #155

Open
TheFausap opened this issue Oct 8, 2013 · 10 comments
Open

numpy.linalg #155

TheFausap opened this issue Oct 8, 2013 · 10 comments

Comments

@TheFausap
Copy link

Hello all,

i'm trying to use numpy.linalg functions with pythran but I receive an error : Attribute 'linalg' unknown.

How can I fix this ?

thanks,
Fausto

@pbrunet
Copy link
Collaborator

pbrunet commented Oct 8, 2013

Hi TheFausap,

I am sorry, for now, numpy.linalg is not available with pythran.

The only way to fix this is not to use this module or waiting for us to implement it.

As we use nt2 for mathematical aspect of pythran (we are computer scientist, not mathematician), It will not be implemented soon as we are waiting for their C like array.
If you need specifics functions, we may try to add them but it may not be as fast as you expect.

Cheers,
Pierrick

@TheFausap
Copy link
Author

Hello Pierrick,

thanks for the reply and clarification.

I'm using basically norm and eigval functions. So, the norm is easy to
re-implement, for the eigen values (of a complex matrix) it's not so
immediate but it should be feasible anyway :-)

I'll try to do in this way.

thanks a lot,
Fausto

2013/10/8 pbrunet [email protected]

Hi TheFausap,

I am sorry, for now, numpy.linalg is not available with pythran.

The only way to fix this is not to use this module or waiting for us to
implement it.

As we use nt2 for mathematical aspect of pythran (we are computer
scientist, not mathematician), It will not be implemented soon as we are
waiting for their C like array.
If you need specifics functions, we may try to add them but it may not be
as fast as you expect.

Cheers,
Pierrick


Reply to this email directly or view it on GitHubhttps://github.com//issues/155#issuecomment-25871848
.

@serge-sans-paille
Copy link
Owner

Hi Fausto,

can you send us the code you're trying to convert?
It should be possible to add suport to norm and eigval depending on how you're using them before we make the full switch to nt?.

Thxs
I'd be interested in adding the support for linalg and

@TheFausap
Copy link
Author

Hi!

sure.
This code is written by Paul Pham (skc-python) and I'm trying to do some
optimizations in order to speed-up the entire process that is very very
time consuming.

Attached there's the file I'm trying to convert.

thanks,
Fausto

2013/10/8 serge-sans-paille [email protected]

Hi Fausto,

can you send us the code you're trying to convert?
It should be possible to add suport to norm and eigval depending on how
you're using them before we make the full switch to nt?.

Thxs
I'd be interested in adding the support for linalg and


Reply to this email directly or view it on GitHubhttps://github.com//issues/155#issuecomment-25881239
.

@serge-sans-paille
Copy link
Owner

mmmh, no attachement there...

@TheFausap
Copy link
Author

hello, maybe there's a filter, deleting attachment to the list... i'll use
the web interface, maybe I have more luck :-D

2013/10/8 serge-sans-paille [email protected]

mmmh, no attachement there...


Reply to this email directly or view it on GitHubhttps://github.com//issues/155#issuecomment-25884343
.

@serge-sans-paille
Copy link
Owner

Or just put it into a gist
https://gist.github.com/

@TheFausap
Copy link
Author

ok... done this is the link... never used GIST :)

https://gist.github.com/TheFausap/6883837

it's public gist, so I hope you can access it...

@serge-sans-paille
Copy link
Owner

Great! Lot of work for us, I'll give you more info later one.

@serge-sans-paille
Copy link
Owner

Hi again. I've been digging through your code and it appears most of it relies on the matrix class, which is still not supported in pythran, and will likely not be supported in a near future.

A good test bed though :-)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants