headers
Christopher Burns | 1 Aug 01:08
Picon
Favicon

Re: patches to reduce import costs

Kevin,

Do you mean add a file on the Wiki or in the source tree somewhere?

Chris

On Thu, Jul 31, 2008 at 2:14 PM, Kevin Jacobs <jacobs <at> bioinformed.com>
<bioinformed <at> gmail.com> wrote:
> For the help of freeze packages, it would be great if you could add a file
> that lists all of the deferred imports that you run across.  That way, we
> can add/update recipes more easily for py2app, py2exe, bbfreeze, etc.
>
> Thanks,
> -Kevin
>
Permalink | Reply | 1 comment |
headers
Stéfan van der Walt | 1 Aug 01:14
Picon
Picon
Favicon

Re: patches to reduce import costs

2008/7/31 Andrew Dalke <dalke <at> dalkescientific.com>:
> I don't see a place to submit patches.  Is there a patch manager for
> numpy?
>
> Here's a patch to defer importing 'tempfile' until needed.  I
> previously mentioned one other place that didn't need tempfile.  With
> this there is no 'import tempfile' during 'import numpy'
>
> This improves startup by about 7%

Thanks for the patch.  Applied in r5588.

Cheers
Stéfan
Permalink | Reply | 0 comments |
headers
Kevin Jacobs | 1 Aug 01:14
Picon

Re: patches to reduce import costs

On Thu, Jul 31, 2008 at 7:08 PM, Christopher Burns <cburns <at> berkeley.edu> wrote:
Do you mean add a file on the Wiki or in the source tree somewhere?


Either or both-- so long as there is a convenient place to find them.  I suppose a Wiki page would be most flexible, since it could be expanded to discuss deeper packaging issues.

Thanks,
-Kevin

_______________________________________________
Numpy-discussion mailing list
Numpy-discussion <at> scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
Permalink | Reply | 0 comments |
headers
Gael Varoquaux | 1 Aug 02:38
Favicon
Gravatar

Re: [ANNOUNCE] ETS 3.0.0b1 released!

On Thu, Jul 31, 2008 at 05:48:02PM -0500, Dave Peterson wrote:
> 	easy_install Mayavi

Just a precision (I just got caught by that), to install mayavi2, the
application, and not just the 3d visualization library that can be used eg in
ipython, you need to do:

easy_install "Mayavi[ui]"

ie the Mayavi project, with the optional ui component. In addition, you
need to choose a toolkit. You can do this with the following command:

easy_install "Traits[ui,wx]"

Finally, under Linux, the above commands will install by default the
packages to your /usr/lib/python2.x/site-packages. If like me you believe
this is owned by the package manager and you would like to install to 
/usr/local/lib/python2.x/site-packages, just use the follwing switch:

easy_install --prefix "Mayavi[ui]"

Gaël
Permalink | Reply | 501 comments |
headers
Robert Kern | 1 Aug 04:23
Picon
Gravatar

Re: "import numpy" is slow

On Thu, Jul 31, 2008 at 11:45, Christopher Barker <Chris.Barker <at> noaa.gov> wrote:

> On my OS-X box (10.4.11, python2.5, numpy '1.1.1rc2'), it takes about 7
> seconds to import numpy!

Can you try running a Python process that just imports numpy under Shark.app?

  http://developer.apple.com/tools/shark_optimize.html

This will help us see what's eating up the time at the C level, at least.

--

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless
enigma that is made terrible by our own mad attempt to interpret it as
though it had an underlying truth."
 -- Umberto Eco
Permalink | Reply | 0 comments |
headers
Bruce Southey | 1 Aug 04:38
Picon

Re: numpy 1.1.rc2: win32 binaries

Hi,
The installation worked on my old Athlon XP running Windows XP and
'numpy.test(level=1)' gave no errors.

I did not get an error for the code provided for ticket 844 so I
presume this ticket is fixed:
'numpy.inner(F,F)' results in 'array([[ 0.]])'

Also, the installer gives this information:
    Author: Travis E. Oliphant, et.al.
    Author_email: oliphant <at> ee.byu.edu
    Description: NumPy: array processing for numbers, strings,
records, and objects.
    Maintainer: NumPy Developers
    Maintainer_email: numpy-discussion <at> lists.sourceforge.net
    Name: numpy
    Url: http://numpy.scipy.org
    Version: 1.1.1.dev5559

I think that at least the ' Author_email' and 'Maintainer_email'
should be updated.

Thanks
Bruce

On Thu, Jul 31, 2008 at 4:08 PM, Jarrod Millman <millman <at> berkeley.edu> wrote:
> On Sun, Jul 27, 2008 at 10:30 PM, David Cournapeau
> <david <at> ar.media.kyoto-u.ac.jp> wrote:
>> http://www.ar.media.kyoto-u.ac.jp/members/david/archives/numpy-1.1.1.dev5559-win32-superpack-python2.5.exe
>
> I want to get the final 1.1.1 release out ASAP, but I need some
> feedback on the windows binaries.  Could someone please try them out
> and let us know if you run into any problems.
>
> In particular, could someone verify that this has been fixed:
> http://projects.scipy.org/scipy/numpy/ticket/844
>
> Thanks,
>
> --
> Jarrod Millman
> Computational Infrastructure for Research Labs
> 10 Giannini Hall, UC Berkeley
> phone: 510.643.4014
> http://cirl.berkeley.edu/
> _______________________________________________
> Numpy-discussion mailing list
> Numpy-discussion <at> scipy.org
> http://projects.scipy.org/mailman/listinfo/numpy-discussion
>
Permalink | Reply | 2 comments |
headers
Francesc Alted | 1 Aug 07:26

User-defined data-types

Hi,

We need to make a dtype that needs to be defined in the user's 
application space.  In the NumPy book there is a section devoted to 
this subject, but also warns:

"""
Adding data-types is one of the less well-tested areas for NumPy 1.0, so 
there may be bugs remaining in the approach.
"""

Anybody out there has tried this successfully?

Thanks,

--

-- 
Francesc Alted
Permalink | Reply | 2 comments |
headers
David Cournapeau | 1 Aug 07:30
Picon
Picon

Re: numpy 1.1.rc2: win32 binaries

Bruce Southey wrote:
> Hi,
> The installation worked on my old Athlon XP running Windows XP and
> 'numpy.test(level=1)' gave no errors.
>   

By old, do you mean it does not have at least SSE2 ? If so, the problem
would not have happened anyway because the numpy  installer does not use
ATLAS there,

Thanks for the report,

David
Permalink | Reply | 1 comment |
headers
Charles R Harris | 1 Aug 08:22
Picon

Re: User-defined data-types



On Thu, Jul 31, 2008 at 11:26 PM, Francesc Alted <faltet <at> pytables.org> wrote:
Hi,

We need to make a dtype that needs to be defined in the user's
application space.  In the NumPy book there is a section devoted to
this subject, but also warns:

"""
Adding data-types is one of the less well-tested areas for NumPy 1.0, so
there may be bugs remaining in the approach.
"""

Anybody out there has tried this successfully?

Not me, but it is on my list. You will probably want to use at least a few ufuncs with it, no? The new type will probably also need a type number. Oh, Heck, give it a shot.  Someone has to go first O_o

Chuck 


_______________________________________________
Numpy-discussion mailing list
Numpy-discussion <at> scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
Permalink | Reply | 1 comment |
headers
Neal Becker | 1 Aug 13:05
Picon

Re: User-defined data-types

Charles R Harris wrote:

> On Thu, Jul 31, 2008 at 11:26 PM, Francesc Alted
> <faltet <at> pytables.org>wrote:
> 
>> Hi,
>>
>> We need to make a dtype that needs to be defined in the user's
>> application space.  In the NumPy book there is a section devoted to
>> this subject, but also warns:
>>
>> """
>> Adding data-types is one of the less well-tested areas for NumPy 1.0, so
>> there may be bugs remaining in the approach.
>> """
>>
>> Anybody out there has tried this successfully?
>>
> 
> Not me, but it is on my list. You will probably want to use at least a few
> ufuncs with it, no? The new type will probably also need a type number.
> Oh,
> Heck, give it a shot.  Someone has to go first O_o
> 
> Chuck

Yes, I have, a few months back.  Check list archives.
Permalink | Reply | 0 comments |

Gmane