>>> [..]
- Structured SVM / CRF learning
This is a big one. Not sure what other people think of it.
I think having a structured SVM would be great.
>>
>>> +100 on this one...
>>>
>> For this, do we need to have our own SVM solver? This is a naive
>> question, I have ne
On Thu, Jan 19, 2012 at 12:08:41AM +0100, Andreas wrote:
> I have no experience with GSoC and I will totally bow
> to you wisdom there. My thinking was that single
> algorithms are more "project-like" than doing polishing here and
> there.
Yes. My point was that I'd like to see project that help u
On Thu, Jan 19, 2012 at 12:13:38PM +0900, Mathieu Blondel wrote:
> Since your data is sparse, you need to use svm.sparse.SVC, not svm.SVC.
Those error messages are really not enlightning. Mathieu, you were saying
in the thread about GSOC that sparse functionality in the scikit could
use some love.
On Wed, Jan 18, 2012 at 11:53:01PM +0100, Andreas wrote:
> > Factor analysis is a decomposition with a particular
> > assumption about the noise.
> > See Bishop page 583
> For a bit more detail:
> The idea is that the data is a linear transform of some
> underlying, lower dimensional data plus so
On Thu, Jan 19, 2012 at 9:57 AM, Manish Katyal wrote:
> Just getting started with scikits and was running into a problem:
>
> My large dataset is in SVMLight format. I load it like (X_train, y_train =
> load_svmlight_file(f))
> When I try using the SVM
> Classifier: svm.SVC(gamma=0.001).fit(X_trai
On 19 January 2012 11:57, Manish Katyal wrote:
> Just getting started with scikits and was running into a problem:
>
> My large dataset is in SVMLight format. I load it like (X_train, y_train =
> load_svmlight_file(f))
> When I try using the SVM
> Classifier: svm.SVC(gamma=0.001).fit(X_train,y_tr
Just getting started with scikits and was running into a problem:
My large dataset is in SVMLight format. I load it like (X_train, y_train =
load_svmlight_file(f))
When I try using the SVM
Classifier: svm.SVC(gamma=0.001).fit(X_train,y_train), I get the following
error:
"File
"/var/tmp/scilearn/l
FA gets used a lot in finance for getting tractable factorizations of large
covariance matrices. Definitely would be very useful to have in sklearn.
On Wed, Jan 18, 2012 at 5:56 PM, David Warde-Farley <
warde...@iro.umontreal.ca> wrote:
> On Wed, Jan 18, 2012 at 11:49:07PM +0100, Andreas wrote:
On Thu, Jan 19, 2012 at 7:44 AM, Gael Varoquaux
wrote:
> On Wed, Jan 18, 2012 at 11:37:15PM +0100, Andreas wrote:
>> Having this feature might get us a LOT of attention.
>> But this is really not a simple project.
>
> Before trying to jump to the super fancy features, I'd rather have a
> polished
On 01/18/2012 11:44 PM, Gael Varoquaux wrote:
> On Wed, Jan 18, 2012 at 11:37:15PM +0100, Andreas wrote:
>
>> Having this feature might get us a LOT of attention.
>> But this is really not a simple project.
>>
> Before trying to jump to the super fancy features, I'd rather have a
> polish
On Jan 19, 2012, at 00:23 , Gael Varoquaux wrote:
> On Wed, Jan 18, 2012 at 07:37:12PM +0900, Mathieu Blondel wrote:
>> It would be nice if you could make a few contributions to scikit-learn
>> before the application process starts. This will allow you to
>> familiarize with the code base, us to
On Wed, Jan 18, 2012 at 11:49:07PM +0100, Andreas wrote:
> > Actually, I am not sure what FA means. For me ICA, PCA, or any
> > decomposition model is an FA. Joris, what do you have in mind in
> > particular?
> >
> >
> Factor analysis is a decomposition with a particular
> assumption about the
On 01/18/2012 11:49 PM, Andreas wrote:
> On 01/18/2012 11:45 PM, Gael Varoquaux wrote:
>
>> On Wed, Jan 18, 2012 at 11:41:24PM +0100, Andreas wrote:
>>
>>
>>> As far as I can tell, FA is not implemented yet.
>>>
>>>
>> Actually, I am not sure what FA means. For me ICA, PCA, o
On 01/18/2012 11:45 PM, Gael Varoquaux wrote:
> On Wed, Jan 18, 2012 at 11:41:24PM +0100, Andreas wrote:
>
>> As far as I can tell, FA is not implemented yet.
>>
> Actually, I am not sure what FA means. For me ICA, PCA, or any
> decomposition model is an FA. Joris, what do you have in
On Jan 19, 2012, at 00:45 , Gael Varoquaux wrote:
> On Wed, Jan 18, 2012 at 11:41:24PM +0100, Andreas wrote:
>> As far as I can tell, FA is not implemented yet.
>
> Actually, I am not sure what FA means. For me ICA, PCA, or any
> decomposition model is an FA. Joris, what do you have in mind in
On Wed, Jan 18, 2012 at 11:41:24PM +0100, Andreas wrote:
>As far as I can tell, FA is not implemented yet.
Actually, I am not sure what FA means. For me ICA, PCA, or any
decomposition model is an FA. Joris, what do you have in mind in
particular?
Gael
On Wed, Jan 18, 2012 at 11:37:15PM +0100, Andreas wrote:
> Having this feature might get us a LOT of attention.
> But this is really not a simple project.
Before trying to jump to the super fancy features, I'd rather have a
polished and versatile version of the scikit. They are many things that I
Hi Joris.
As far as I can tell, FA is not implemented yet.
Thanks for pointing that out. It should definitely be included.
Cheers,
Andy
On 01/18/2012 02:51 PM, Joris A. wrote:
Hello All,
Sorry if it's stupid but it's not so obvious to me.
Is it possible to perform a factorial analysis with sk
On 01/18/2012 11:26 PM, Gael Varoquaux wrote:
> On Wed, Jan 18, 2012 at 11:28:52AM +0100, Lars Buitinck wrote:
>
>> 2012/1/18 Andreas:
>>
>>> - Structured SVM / CRF learning
>>> This is a big one. Not sure what other people think of it.
>>> I think having a structured SVM would
On Wed, Jan 18, 2012 at 11:28:52AM +0100, Lars Buitinck wrote:
> 2012/1/18 Andreas :
> > - Structured SVM / CRF learning
> > This is a big one. Not sure what other people think of it.
> > I think having a structured SVM would be great.
> +100 on this one...
For this, do we need to have ou
On Wed, Jan 18, 2012 at 07:37:12PM +0900, Mathieu Blondel wrote:
> It would be nice if you could make a few contributions to scikit-learn
> before the application process starts. This will allow you to
> familiarize with the code base, us to evaluate your potential and, if
> I remember correctly, t
2012/1/18 Vlad Niculae :
> The bdist_wininst is only if you want to make a neat executable installer
> like the ones we ship. Maybe we should add a `make wininst` target for
> windows users who have Make. But I don't think this is used often.
+1; this may be useful for distributing hacked versio
On Jan 18, 2012, at 20:23 , Andreas wrote:
> On 01/18/2012 07:19 PM, Vlad Niculae wrote:
>> I am quoting from http://docs.python.org/distutils/builtdist.html
>>
>>
>>> By default the installer will display the cool “Python Powered”
>>> logo when it is run, but you can also supply your own> 152
On 01/18/2012 07:19 PM, Vlad Niculae wrote:
> I am quoting from http://docs.python.org/distutils/builtdist.html
>
>
>> By default the installer will display the cool “Python Powered”
>> logo when it is run, but you can also supply your own> 152x261
>> bitmap which must be a Windows .bmpfile wi
I am quoting from http://docs.python.org/distutils/builtdist.html
> By default the installer will display the cool “Python Powered”
> logo when it is run, but you can also supply your own > 152x261
> bitmap which must be a Windows .bmpfile with the --bitmap option.
I'm assuming -b is short for -
Hello,
This doesn't seem right to me. Is the path to the logo bitmap supposed
to appear in this line?
In doc/install.rst:
the command to execute is::
python setup.py bdist_wininst -b doc/logos/scikit-learn-logo.bmp
This will create an insta
Hi Jaidev,
Well, the two of us do have a busy summer coming up, but a word of
> caution - Google hasn't decided yet whether they will hold GSoC this
> year. Please join the GSoC mailing list too.
>
Hm... Let us hope for the best.
>
> We'll talk more tonight if you are free...
>
Sure.
See you
Hi Bala,
Well, the two of us do have a busy summer coming up, but a word of
caution - Google hasn't decided yet whether they will hold GSoC this
year. Please join the GSoC mailing list too.
We'll talk more tonight if you are free...
Cheers
---
Hello All,
Sorry if it's stupid but it's not so obvious to me.
Is it possible to perform a factorial analysis with sklearn or do I have to
use other libraries?
Thanks and regards,
Joris
--
Keep Your Developer Skills Curre
Hi :)
You might start on this one:
> https://github.com/scikit-learn/scikit-learn/issues/559
> It should be fairly easy to do.
>
Okay... Sure ! I'll try to do this.
>
>
> --
> Keep Your Developer Skills Current with Lea
You might start on this one:
https://github.com/scikit-learn/scikit-learn/issues/559
It should be fairly easy to do.
--
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library f
On Wed, Jan 18, 2012 at 7:47 PM, Bala Subrahmanyam Varanasi
wrote:
> Upto now, I pulled two commits regarding the documentation. I hope I could
> do more in the coming days. Here are my commits.
>
> https://github.com/Balu-Varanasi/scikit-learn/commit/36d0adb8c14b8105b9ba690073d0501955bce328
>
>
Hi list,
Is there any reason the Windows binary for the 0.10 release does not appear
on the sourceforge repos?
http://sourceforge.net/projects/scikit-learn/files/
Thanks,
Vincent
2012/1/13 Yaroslav Halchenko
> Thanks! btw -- in general you might prefer annotated or signed tags
> instead of "l
Dear Mathieu,
It would be nice if you could make a few contributions to scikit-learn
> before the application process starts. This will allow you to
> familiarize with the code base, us to evaluate your potential and, if
> I remember correctly, this is actually a requirement from the PSF.
>
I wou
On Wed, Jan 18, 2012 at 3:12 PM, Bala Subrahmanyam Varanasi
wrote:
> Also... I'm attending to Stanford's Online courses - ML class and NLP class.
> I believe this is the right time to discuss. Because, I can learn new things
> before the start of GSoC and can work on challenging implementations i
2012/1/18 Andreas :
> - Structured SVM / CRF learning
> This is a big one. Not sure what other people think of it.
> I think having a structured SVM would be great.
+100 on this one...
> Designing the interface is also non-trivial.
Indeed. I suspect different APIs would be needed for
Hi Bela.
I'm not sure how this usually goes but here is my current wish list.
We'd have to discuss whether any of that actually fits into the scikits,
thou ;)
- Multilayer Perceptron and Multinomial Logistic regression
I have been working on that so maybe there is not enough
left to do
37 matches
Mail list logo
