On Fri, Jan 6, 2012 at 7:00 AM, Olivier Grisel wrote:
> 2012/1/5 Fabian Pedregosa :
> > On Thu, Jan 5, 2012 at 11:30 PM, Gael Varoquaux
> > wrote:
> >> On Thu, Jan 05, 2012 at 11:28:45PM +0100, Andreas wrote:
> >>> As I said, I don't have multiple versions and the only
> >>> thing that fails is s
On Jan 5, 2012, at 23:45 , Fabian Pedregosa wrote:
> and that was
> quite convenient for testing on systems on which nosetest fails
> (windows).
Hi Fabian
Could you please be more specific regarding this point, since as a former
Windows user, I find that I don't know what you mean.
On topic, I
2012/1/5 Fabian Pedregosa :
> On Thu, Jan 5, 2012 at 11:30 PM, Gael Varoquaux
> wrote:
>> On Thu, Jan 05, 2012 at 11:28:45PM +0100, Andreas wrote:
>>> As I said, I don't have multiple versions and the only
>>> thing that fails is sklearn.test().
>>
>> OK, so let's move the warning there.
>
> +1. R
On Thu, Jan 5, 2012 at 11:30 PM, Gael Varoquaux
wrote:
> On Thu, Jan 05, 2012 at 11:28:45PM +0100, Andreas wrote:
>> As I said, I don't have multiple versions and the only
>> thing that fails is sklearn.test().
>
> OK, so let's move the warning there.
+1. Raising an exception is also OK and might
2012/1/5 Olivier Grisel :
> We could rename it "as_float_data" instead. I have no strong opinion on this.
Hmm... well, as it's internal, we might as well leave it as is.
--
Lars Buitinck
Scientific programmer, ILPS
University of Amsterdam
On Thu, Jan 05, 2012 at 11:28:45PM +0100, Andreas wrote:
> As I said, I don't have multiple versions and the only
> thing that fails is sklearn.test().
OK, so let's move the warning there.
G
--
Ridiculously easy VDI. Wit
On 01/05/2012 10:45 PM, Fabian Pedregosa wrote:
> On Thu, Jan 5, 2012 at 11:47 AM, Andreas
> wrote:
>> On 01/05/2012 07:37 AM, Gael Varoquaux wrote:
>>> On Wed, Jan 04, 2012 at 03:38:41PM -0800, Jacob VanderPlas wrote:
>>>
Importing sklearn from within the scikit-learn source directory
2012/1/5 Lars Buitinck :
> 2012/1/5 Gael Varoquaux :
>> On Thu, Jan 05, 2012 at 10:40:20PM +0100, Gael Varoquaux wrote:
>>> Right now, when a sparse matrix is given at the validation utility
>>> 'as_float_array', it crashes with the following incomprehensible error
>>> message:
>>
>> Alright, it's
On Thu, Jan 05, 2012 at 11:17:34PM +0100, Lars Buitinck wrote:
> > Alright, it's not what I thought. Forget this message, apparently this
> > incomprehensible error message is raised for another reason.
> But I'm still surprised at the behavior:
> In [1]: from sklearn.utils import as_float_array
On Thu, Jan 5, 2012 at 11:14 PM, Lars Buitinck wrote:
> 2012/1/5 Fabian Pedregosa :
>> Good look with the building issues. A safe solution would be to do
>> conditional compilation. Scipy does it and we use the same build
>> system, unfortunately it's not as easy as it sounds :-).
>
> I hope you m
2012/1/5 Gael Varoquaux :
> On Thu, Jan 05, 2012 at 10:40:20PM +0100, Gael Varoquaux wrote:
>> Right now, when a sparse matrix is given at the validation utility
>> 'as_float_array', it crashes with the following incomprehensible error
>> message:
>
> Alright, it's not what I thought. Forget this m
2012/1/5 Fabian Pedregosa :
> Good look with the building issues. A safe solution would be to do
> conditional compilation. Scipy does it and we use the same build
> system, unfortunately it's not as easy as it sounds :-).
I hope you mean just passing compiler options optionally; you don't
need co
On Thu, Jan 5, 2012 at 11:47 AM, Andreas wrote:
> On 01/05/2012 07:37 AM, Gael Varoquaux wrote:
>> On Wed, Jan 04, 2012 at 03:38:41PM -0800, Jacob VanderPlas wrote:
>>
>>> Importing sklearn from within the scikit-learn source directory produces
>>> no such error. Perhaps this would be a good fix
On Thu, Jan 05, 2012 at 10:40:20PM +0100, Gael Varoquaux wrote:
> Right now, when a sparse matrix is given at the validation utility
> 'as_float_array', it crashes with the following incomprehensible error
> message:
Alright, it's not what I thought. Forget this message, apparently this
incomprehe
Hi list,
Right now, when a sparse matrix is given at the validation utility
'as_float_array', it crashes with the following incomprehensible error
message:
/usr/local/lib/python2.7/dist-packages/scikit_learn-0.10_git-py2.7-linux-x86_64.
egg/sklearn/utils/validation.pyc in as_float_array(X, copy)
On Thu, Jan 5, 2012 at 9:35 AM, Benjamin Hepp
wrote:
> My implementation is assuming all data fits in memory. I'll do some
> benchmarks and look into the openmp/building/cython issues.
Hey,
Good look with the building issues. A safe solution would be to do
conditional compilation. Scipy does it
OK I'll push to master
Jake
Andreas wrote:
> It does :)
> Thanks a lot!
>
>
> On 01/05/2012 09:19 PM, Jacob VanderPlas wrote:
>
>> See if this fixes things:
>> https://github.com/jakevdp/scikit-learn/tree/doc-math-fix
>>Jake
>>
>> Andreas wrote:
>>
>>
>>> On 01/05/2012 08:58 PM,
It does :)
Thanks a lot!
On 01/05/2012 09:19 PM, Jacob VanderPlas wrote:
> See if this fixes things:
> https://github.com/jakevdp/scikit-learn/tree/doc-math-fix
>Jake
>
> Andreas wrote:
>
>> On 01/05/2012 08:58 PM, Andreas wrote:
>>
>>
>>> Thanks Jake for looking into this.
>>> I ha
See if this fixes things:
https://github.com/jakevdp/scikit-learn/tree/doc-math-fix
Jake
Andreas wrote:
> On 01/05/2012 08:58 PM, Andreas wrote:
>
>> Thanks Jake for looking into this.
>> I have 8 pngs in this directory
>>
>>
>>
> If I do a "make clean" and a "make" again, I get
I get 328, followed by 244. I think a more careful removal of the
images will fix this.
Jake
Andreas wrote:
> On 01/05/2012 08:58 PM, Andreas wrote:
>
>> Thanks Jake for looking into this.
>> I have 8 pngs in this directory
>>
>>
>>
> If I do a "make clean" and a "make" again,
On 01/05/2012 08:58 PM, Andreas wrote:
> Thanks Jake for looking into this.
> I have 8 pngs in this directory
>
>
If I do a "make clean" and a "make" again, I get 328 ;)
> On 01/05/2012 08:47 PM, Jacob VanderPlas wrote:
>
>> I'm having trouble replicating the problem.
>> When you ``mak
Thanks Jake for looking into this.
I have 8 pngs in this directory
On 01/05/2012 08:47 PM, Jacob VanderPlas wrote:
> I'm having trouble replicating the problem.
> When you ``make html`` twice in a row, do you see anything in the
> _build/html/_images/math directory?
>Jake
>
> Andreas wrote
I'm having trouble replicating the problem.
When you ``make html`` twice in a row, do you see anything in the
_build/html/_images/math directory?
Jake
Andreas wrote:
> On 01/05/2012 08:03 PM, Jacob VanderPlas wrote:
>
>> I wonder if this is a problem with that doc/image fix I put up during
>
Definitely related.
I guess the code should be modified to not use "rmtree" but to just
remove the figure images alone. I'll take a look
Jake
Andreas wrote:
> On 01/05/2012 08:03 PM, Jacob VanderPlas wrote:
>
>> I wonder if this is a problem with that doc/image fix I put up during
>> the s
Emanuele,
I should also note that a distinct advantage of cover trees is that,
unlike ball tree, there is no need to compute the mean/median point of
each node. This means that their storage can be much more compact, and
they'd be very suitable to computing distances within sparse data. For
t
On 01/05/2012 08:03 PM, Jacob VanderPlas wrote:
> I wonder if this is a problem with that doc/image fix I put up during
> the sprint? When the docs are re-made, all the math images are
> removed. I recall checking and seeing that they were re-generated, but
> I may be wrong. Can you check this,
I wonder if this is a problem with that doc/image fix I put up during
the sprint? When the docs are re-made, all the math images are
removed. I recall checking and seeing that they were re-generated, but
I may be wrong. Can you check this, Andy?
Jake
Andreas wrote:
> On 01/05/2012 07:53 PM
On 01/05/2012 07:53 PM, Gael Varoquaux wrote:
> On Thu, Jan 05, 2012 at 07:44:12PM +0100, Andreas wrote:
>
>> Can anyone explain to me how to build the html docs
>> so that the math is rendered with latex?
>>
> It should be. You need to use the ..math directive.
>
>
I compared
http:/
On Thu, Jan 05, 2012 at 07:44:12PM +0100, Andreas wrote:
> Can anyone explain to me how to build the html docs
> so that the math is rendered with latex?
It should be. You need to use the ..math directive.
> This is the pngmath_latex sphinx plugin, right?
It's actually fully done through matplot
Hi everybody.
Can anyone explain to me how to build the html docs
so that the math is rendered with latex?
This is the pngmath_latex sphinx plugin, right?
I have latex and dvipng installed but the images
don't show up.
Anything else I need or any other build command?
There seems to be nothing in th
Also we should ensure that all of those naming conventions for
distances are consistent with what we already have in the
sklearn.metrics.pairwise module.
--
Olivier
--
Ridiculously easy VDI. With Citrix VDI-in-a-Box, you
I would rename "tmp" as "work_buffer".
Same for "VI" I don't understand the meaning either.
I would also like to avoid "p" in the public API of kNN methods. Maybe
use a "distance" parameter that could accept string values such as
distance="euclidean" or distance="manhattan" or a float value that
On Thu, Jan 05, 2012 at 09:33:28AM -0800, Jacob VanderPlas wrote:
> In computing the mahalanobis distance, a temporary storage array is
> needed. To avoid repeated allocation within the distance C-function
> (and to avoid the need for malloc/free), I pre-allocate this temporary
> array via nump
Gael Varoquaux wrote:
> You are cimporting malloc and free. I have a personnal difficult
> relationship with those two old friends. However, it seems not to be used
> in the code. I just wanted to check.
>
I initially used malloc and free, but settled on the `tmp` pointer to
avoid this (see be
Mathias,
I'm glad you're excited to work on this! I think starting with just the
minkowski p-distance in this case is a good idea, and it would be a
great way for you to gain experience with the code. I'd do the following:
- in sklearn/neighbors/base.py, add a parameter `p` to the NeighborsBas
On Thu, Jan 05, 2012 at 08:33:01AM -0800, Jacob VanderPlas wrote:
> Here's a small example I coded up that shows how I envision including
> multiple distance metrics in BallTree
> https://gist.github.com/1565998
> The idea is that you create functions to compute distance which expose C
> functi
Hi,
First, thanks for all the answers! Waauw, really interesting discussion. I
have only basic Python skills, and never programmed in Cython (together
with a lot of time constraints, as most of you probably), but I would like
to give it a try to add new distance metrics to the brute force method.
Here's a small example I coded up that shows how I envision including
multiple distance metrics in BallTree
https://gist.github.com/1565998
The idea is that you create functions to compute distance which expose C
function pointers, so that the ball tree cython code can call these
without pytho
On 01/05/2012 07:37 AM, Gael Varoquaux wrote:
> On Wed, Jan 04, 2012 at 03:38:41PM -0800, Jacob VanderPlas wrote:
>
>> Importing sklearn from within the scikit-learn source directory produces
>> no such error. Perhaps this would be a good fix
>>
> +1
>
>
+10
---
My implementation is assuming all data fits in memory. I'll do some
benchmarks and look into the openmp/building/cython issues.
Benni
On 12/23/11 10:46 PM, Gael Varoquaux wrote:
> On Fri, Dec 23, 2011 at 10:42:45PM +0100, Gael wrote:
>> The reason that we have integrated openmp code in the scikit
40 matches
Mail list logo
