From yrohinkumar at gmail.com  Tue Aug  1 09:15:56 2017
From: yrohinkumar at gmail.com (Rohin Kumar)
Date: Tue, 1 Aug 2017 18:45:56 +0530
Subject: [scikit-learn] Nearest neighbor search with 2 distance measures
In-Reply-To: <CAARwKuszb0QxfNce+b5640+3OuG=Mk26Mnt3cEiQ7Lzx80=BJA@mail.gmail.com>
References: <379121501436421@mxfront4j.mail.yandex.net>
 <CAARwKuu+zHGw61=q1t8rO5F-KDUoi+Y7bCFC3SRYBgJAEvyXvw@mail.gmail.com>
 <CACpqBg0uRrMxWENyz1vgHwDXQiSgU6Kdeh+=ATao29e2Lz0EEw@mail.gmail.com>
 <CAARwKuszb0QxfNce+b5640+3OuG=Mk26Mnt3cEiQ7Lzx80=BJA@mail.gmail.com>
Message-ID: <CAARwKut+OH-vHUn3coTySrKA=bk3prhe1btkjBpJRb5P3iwEug@mail.gmail.com>

Since you seem to be from Astrophysics/Cosmology background (I am assuming
you are jakevdp - the creator of astroML - if you are - I am lucky!), I can
explain my application scenario. I am trying to calculate the anisotropic
two-point correlation function something like done in rp_pi_tpcf
<http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.rp_pi_tpcf.html#halotools.mock_observables.rp_pi_tpcf>
 or s_mu_tpcf
<http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.s_mu_tpcf.html#halotools.mock_observables.s_mu_tpcf>
using pairs (DD,DR,RR) calculated from BallTree.two_point_correlation

In halotools (
http://halotools.readthedocs.io/en/latest/function_usage/mock_observables_functions.html)
it is implemented using rectangular grids. I could calculate 2pcf with
custom metrics using one variable with BallTree as done in astroML. I
intend to find the anisotropic counter part.

Thanks & Regards,
Rohin

Y.Rohin Kumar,
+919818092877.

On Tue, Aug 1, 2017 at 5:18 PM, Rohin Kumar <yrohinkumar at gmail.com> wrote:

> Dear Jake,
>
> Thanks for your response. I meant to group/count pairs in boxes (using two
> arrays simultaneously-hence needing 2 metrics) instead of one distance
> array as the binning parameter. I don't know if the algorithm supports such
> a thing. For now, I am proceeding with your suggestion of two ball trees at
> huge computational cost. I hope I am able to frame my question properly.
>
> Thanks & Regards,
> Rohin.
>
>
>
> On Mon, Jul 31, 2017 at 8:16 PM, Jacob Vanderplas <
> jakevdp at cs.washington.edu> wrote:
>
>> On Sun, Jul 30, 2017 at 11:18 AM, Rohin Kumar <yrohinkumar at gmail.com>
>> wrote:
>>
>>> *update*
>>>
>>> May be it doesn't have to be done at the tree creation level. It could
>>> be using loops and creating two different balltrees. Something like
>>>
>>> tree1=BallTree(X,metric='metric1') #for x-z plane
>>> tree2=BallTree(X,metric='metric2') #for y-z plane
>>>
>>> And then calculate correlation functions in a loop to get tpcf(X,r1,r2)
>>>  using tree1.two_point_correlation(X,r1) and
>>> tree2.two_point_correlation(X,r2)
>>>
>>
>> Hi Rohin,
>> It's not exactly clear to me what you wish the tree to do with the two
>> different metrics, but in any case the ball tree only supports one metric
>> at a time. If you can construct your desired result from two ball trees
>> each with its own metric, then that's probably the best way to proceed,
>>    Jake
>>
>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170801/1cd0cd2e/attachment.html>

From yrohinkumar at gmail.com  Tue Aug  1 07:48:23 2017
From: yrohinkumar at gmail.com (Rohin Kumar)
Date: Tue, 1 Aug 2017 17:18:23 +0530
Subject: [scikit-learn] Nearest neighbor search with 2 distance measures
In-Reply-To: <CACpqBg0uRrMxWENyz1vgHwDXQiSgU6Kdeh+=ATao29e2Lz0EEw@mail.gmail.com>
References: <379121501436421@mxfront4j.mail.yandex.net>
 <CAARwKuu+zHGw61=q1t8rO5F-KDUoi+Y7bCFC3SRYBgJAEvyXvw@mail.gmail.com>
 <CACpqBg0uRrMxWENyz1vgHwDXQiSgU6Kdeh+=ATao29e2Lz0EEw@mail.gmail.com>
Message-ID: <CAARwKuszb0QxfNce+b5640+3OuG=Mk26Mnt3cEiQ7Lzx80=BJA@mail.gmail.com>

Dear Jake,

Thanks for your response. I meant to group/count pairs in boxes (using two
arrays simultaneously-hence needing 2 metrics) instead of one distance
array as the binning parameter. I don't know if the algorithm supports such
a thing. For now, I am proceeding with your suggestion of two ball trees at
huge computational cost. I hope I am able to frame my question properly.

Thanks & Regards,
Rohin.



On Mon, Jul 31, 2017 at 8:16 PM, Jacob Vanderplas <jakevdp at cs.washington.edu
> wrote:

> On Sun, Jul 30, 2017 at 11:18 AM, Rohin Kumar <yrohinkumar at gmail.com>
> wrote:
>
>> *update*
>>
>> May be it doesn't have to be done at the tree creation level. It could be
>> using loops and creating two different balltrees. Something like
>>
>> tree1=BallTree(X,metric='metric1') #for x-z plane
>> tree2=BallTree(X,metric='metric2') #for y-z plane
>>
>> And then calculate correlation functions in a loop to get tpcf(X,r1,r2)
>>  using tree1.two_point_correlation(X,r1) and tree2.two_point_correlation(
>> X,r2)
>>
>
> Hi Rohin,
> It's not exactly clear to me what you wish the tree to do with the two
> different metrics, but in any case the ball tree only supports one metric
> at a time. If you can construct your desired result from two ball trees
> each with its own metric, then that's probably the best way to proceed,
>    Jake
>
>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170801/9b61e2a9/attachment-0001.html>

From Jeremiah.Johnson at unh.edu  Tue Aug  1 12:03:01 2017
From: Jeremiah.Johnson at unh.edu (Johnson, Jeremiah)
Date: Tue, 1 Aug 2017 16:03:01 +0000
Subject: [scikit-learn] question about class_weights in LogisticRegression
Message-ID: <D5A61E73.9E11%jeremiah.johnson@unh.edu>

Hello all,

I'm looking for confirmation on an implementation detail that is somewhere in liblinear, but I haven't found documentation for yet. When the class_weights='balanced' parameter is set in LogisticRegression, then the regularisation parameter for an observation from class I is class_weight[I] * C, where C is the usual regularization parameter - is this correct?

Thanks,
Jeremiah

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170801/404a5898/attachment.html>

From stuart at stuartreynolds.net  Tue Aug  1 12:19:54 2017
From: stuart at stuartreynolds.net (Stuart Reynolds)
Date: Tue, 1 Aug 2017 09:19:54 -0700
Subject: [scikit-learn] question about class_weights in
 LogisticRegression
In-Reply-To: <D5A61E73.9E11%jeremiah.johnson@unh.edu>
References: <D5A61E73.9E11%jeremiah.johnson@unh.edu>
Message-ID: <CAAy-kdkgepButRwe32+SA6D-xAS2be5-HAvMM=Yuf_3vMfFehA@mail.gmail.com>

I hope not. And not accoring to the docs...
https://github.com/scikit-learn/scikit-learn/blob/ab93d65/sklearn/linear_model/logistic.py#L947

class_weight : dict or 'balanced', optional
Weights associated with classes in the form ``{class_label: weight}``.
If not given, all classes are supposed to have weight one.
The "balanced" mode uses the values of y to automatically adjust
weights inversely proportional to class frequencies in the input data
as ``n_samples / (n_classes * np.bincount(y))``.
Note that these weights will be multiplied with sample_weight (passed
through the fit method) if sample_weight is specified.

On Tue, Aug 1, 2017 at 9:03 AM, Johnson, Jeremiah
<Jeremiah.Johnson at unh.edu> wrote:
> Hello all,
>
> I?m looking for confirmation on an implementation detail that is somewhere
> in liblinear, but I haven?t found documentation for yet. When the
> class_weights=?balanced? parameter is set in LogisticRegression, then the
> regularisation parameter for an observation from class I is class_weight[I]
> * C, where C is the usual regularization parameter ? is this correct?
>
> Thanks,
> Jeremiah
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>

From Jeremiah.Johnson at unh.edu  Tue Aug  1 12:30:22 2017
From: Jeremiah.Johnson at unh.edu (Johnson, Jeremiah)
Date: Tue, 1 Aug 2017 16:30:22 +0000
Subject: [scikit-learn] question about class_weights in
 LogisticRegression
In-Reply-To: <CAAy-kdkgepButRwe32+SA6D-xAS2be5-HAvMM=Yuf_3vMfFehA@mail.gmail.com>
References: <D5A61E73.9E11%jeremiah.johnson@unh.edu>
 <CAAy-kdkgepButRwe32+SA6D-xAS2be5-HAvMM=Yuf_3vMfFehA@mail.gmail.com>
Message-ID: <D5A62380.9E1A%jeremiah.johnson@unh.edu>

Right, I know how the class_weight calculation is performed. But then
those class weights are utilized during the model fit process in some way
in liblinear, and that?s what I am interested in. libSVM does
class_weight[I] * C (https://www.csie.ntu.edu.tw/~cjlin/libsvm/); is the
implementation in liblinear the same?

Best,
Jeremiah



On 8/1/17, 12:19 PM, "scikit-learn on behalf of Stuart Reynolds"
<scikit-learn-bounces+jeremiah.johnson=unh.edu at python.org on behalf of
stuart at stuartreynolds.net> wrote:

>I hope not. And not accoring to the docs...
>https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_scikit-2Dl
>earn_scikit-2Dlearn_blob_ab93d65_sklearn_linear-5Fmodel_logistic.py-23L947
>&d=DwIGaQ&c=c6MrceVCY5m5A_KAUkrdoA&r=hQNTLb4Jonm4n54VBW80WEzIAaqvTOcTEjhIk
>rRJWXo&m=2XR2z3VWvEaERt4miGabDte3xkz_FwzMKMwnvEOWj8o&s=4uJZS3EaQgysmQlzjt-
>yuLkSlcXTd5G50LkEFMcbZLQ&e=
>
>class_weight : dict or 'balanced', optional
>Weights associated with classes in the form ``{class_label: weight}``.
>If not given, all classes are supposed to have weight one.
>The "balanced" mode uses the values of y to automatically adjust
>weights inversely proportional to class frequencies in the input data
>as ``n_samples / (n_classes * np.bincount(y))``.
>Note that these weights will be multiplied with sample_weight (passed
>through the fit method) if sample_weight is specified.
>
>On Tue, Aug 1, 2017 at 9:03 AM, Johnson, Jeremiah
><Jeremiah.Johnson at unh.edu> wrote:
>> Hello all,
>>
>> I?m looking for confirmation on an implementation detail that is
>>somewhere
>> in liblinear, but I haven?t found documentation for yet. When the
>> class_weights=?balanced? parameter is set in LogisticRegression, then
>>the
>> regularisation parameter for an observation from class I is
>>class_weight[I]
>> * C, where C is the usual regularization parameter ? is this correct?
>>
>> Thanks,
>> Jeremiah
>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> 
>>https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.python.org_mail
>>man_listinfo_scikit-2Dlearn&d=DwIGaQ&c=c6MrceVCY5m5A_KAUkrdoA&r=hQNTLb4Jo
>>nm4n54VBW80WEzIAaqvTOcTEjhIkrRJWXo&m=2XR2z3VWvEaERt4miGabDte3xkz_FwzMKMwn
>>vEOWj8o&s=MgZoI9VOHFh3omGKHTASFx3NAVjj6AY3j_75mnOUg04&e=
>>
>_______________________________________________
>scikit-learn mailing list
>scikit-learn at python.org
>https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.python.org_mailm
>an_listinfo_scikit-2Dlearn&d=DwIGaQ&c=c6MrceVCY5m5A_KAUkrdoA&r=hQNTLb4Jonm
>4n54VBW80WEzIAaqvTOcTEjhIkrRJWXo&m=2XR2z3VWvEaERt4miGabDte3xkz_FwzMKMwnvEO
>Wj8o&s=MgZoI9VOHFh3omGKHTASFx3NAVjj6AY3j_75mnOUg04&e= 


From jakevdp at cs.washington.edu  Tue Aug  1 13:25:52 2017
From: jakevdp at cs.washington.edu (Jacob Vanderplas)
Date: Tue, 1 Aug 2017 10:25:52 -0700
Subject: [scikit-learn] Nearest neighbor search with 2 distance measures
In-Reply-To: <CAARwKut+OH-vHUn3coTySrKA=bk3prhe1btkjBpJRb5P3iwEug@mail.gmail.com>
References: <379121501436421@mxfront4j.mail.yandex.net>
 <CAARwKuu+zHGw61=q1t8rO5F-KDUoi+Y7bCFC3SRYBgJAEvyXvw@mail.gmail.com>
 <CACpqBg0uRrMxWENyz1vgHwDXQiSgU6Kdeh+=ATao29e2Lz0EEw@mail.gmail.com>
 <CAARwKuszb0QxfNce+b5640+3OuG=Mk26Mnt3cEiQ7Lzx80=BJA@mail.gmail.com>
 <CAARwKut+OH-vHUn3coTySrKA=bk3prhe1btkjBpJRb5P3iwEug@mail.gmail.com>
Message-ID: <CACpqBg3dSm_FcQLuhy=5Pes3PSYCwYZBB9cjvQrgoOGakwv0SA@mail.gmail.com>

Hi Rohin,
Ah, I see. I don't think a BallTree is the right data structure for an
anisotropic N-point query, because it fundamentally assumes spherical
symmetry of the metric. You may be able to do something like this with a
specialized KD-tree, but scikit-learn doesn't support this, and I don't
imagine that it ever will given the very specialized nature of the
application.

I'm certain someone has written efficient code for this operation in the
astronomy community, but I don't know of any good Python package to
recommend for this ? I'd suggest googling for keywords and seeing where
that gets you.

Thanks,
  Jake

 Jake VanderPlas
 Senior Data Science Fellow
 Director of Open Software
 University of Washington eScience Institute

On Tue, Aug 1, 2017 at 6:15 AM, Rohin Kumar <yrohinkumar at gmail.com> wrote:

> Since you seem to be from Astrophysics/Cosmology background (I am assuming
> you are jakevdp - the creator of astroML - if you are - I am lucky!), I can
> explain my application scenario. I am trying to calculate the anisotropic
> two-point correlation function something like done in rp_pi_tpcf
> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.rp_pi_tpcf.html#halotools.mock_observables.rp_pi_tpcf>
>  or s_mu_tpcf
> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.s_mu_tpcf.html#halotools.mock_observables.s_mu_tpcf>
> using pairs (DD,DR,RR) calculated from BallTree.two_point_correlation
>
> In halotools (http://halotools.readthedocs.io/en/latest/function_usage/
> mock_observables_functions.html) it is implemented using rectangular
> grids. I could calculate 2pcf with custom metrics using one variable with
> BallTree as done in astroML. I intend to find the anisotropic counter part.
>
> Thanks & Regards,
> Rohin
>
> Y.Rohin Kumar,
> +919818092877 <+91%2098180%2092877>.
>
> On Tue, Aug 1, 2017 at 5:18 PM, Rohin Kumar <yrohinkumar at gmail.com> wrote:
>
>> Dear Jake,
>>
>> Thanks for your response. I meant to group/count pairs in boxes (using
>> two arrays simultaneously-hence needing 2 metrics) instead of one distance
>> array as the binning parameter. I don't know if the algorithm supports such
>> a thing. For now, I am proceeding with your suggestion of two ball trees at
>> huge computational cost. I hope I am able to frame my question properly.
>>
>> Thanks & Regards,
>> Rohin.
>>
>>
>>
>> On Mon, Jul 31, 2017 at 8:16 PM, Jacob Vanderplas <
>> jakevdp at cs.washington.edu> wrote:
>>
>>> On Sun, Jul 30, 2017 at 11:18 AM, Rohin Kumar <yrohinkumar at gmail.com>
>>> wrote:
>>>
>>>> *update*
>>>>
>>>> May be it doesn't have to be done at the tree creation level. It could
>>>> be using loops and creating two different balltrees. Something like
>>>>
>>>> tree1=BallTree(X,metric='metric1') #for x-z plane
>>>> tree2=BallTree(X,metric='metric2') #for y-z plane
>>>>
>>>> And then calculate correlation functions in a loop to get tpcf(X,r1,r2)
>>>>  using tree1.two_point_correlation(X,r1) and
>>>> tree2.two_point_correlation(X,r2)
>>>>
>>>
>>> Hi Rohin,
>>> It's not exactly clear to me what you wish the tree to do with the two
>>> different metrics, but in any case the ball tree only supports one metric
>>> at a time. If you can construct your desired result from two ball trees
>>> each with its own metric, then that's probably the best way to proceed,
>>>    Jake
>>>
>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170801/09b22a0f/attachment-0001.html>

From yrohinkumar at gmail.com  Tue Aug  1 13:50:58 2017
From: yrohinkumar at gmail.com (Rohin Kumar)
Date: Tue, 1 Aug 2017 23:20:58 +0530
Subject: [scikit-learn] Nearest neighbor search with 2 distance measures
In-Reply-To: <CACpqBg3dSm_FcQLuhy=5Pes3PSYCwYZBB9cjvQrgoOGakwv0SA@mail.gmail.com>
References: <379121501436421@mxfront4j.mail.yandex.net>
 <CAARwKuu+zHGw61=q1t8rO5F-KDUoi+Y7bCFC3SRYBgJAEvyXvw@mail.gmail.com>
 <CACpqBg0uRrMxWENyz1vgHwDXQiSgU6Kdeh+=ATao29e2Lz0EEw@mail.gmail.com>
 <CAARwKuszb0QxfNce+b5640+3OuG=Mk26Mnt3cEiQ7Lzx80=BJA@mail.gmail.com>
 <CAARwKut+OH-vHUn3coTySrKA=bk3prhe1btkjBpJRb5P3iwEug@mail.gmail.com>
 <CACpqBg3dSm_FcQLuhy=5Pes3PSYCwYZBB9cjvQrgoOGakwv0SA@mail.gmail.com>
Message-ID: <CAARwKuupVuwYtBCZPbQ58gggHmTKn1ReQ=ZSCqjZQC40dFTeXw@mail.gmail.com>

Dear Jake,

Thank you for your prompt reply. I started with KD-tree but after realising
it doesn't support custom metrics (I don't know the reason for this - would
be nice feature) I shifted to BallTree and was looking for a 2 metric based
categorisation. After looking around, the best I could find at most were
brute-force methods written in python (had my own version too) or better
optimised ones in C or FORTRAN. The closest one was halotools which again
works with euclidean metric. For now, I will try to get my work done with 2
different BallTrees iteratively in bins. If I find a better option will try
to post an update.

Regards,
Rohin.


On Tue, Aug 1, 2017 at 10:55 PM, Jacob Vanderplas <jakevdp at cs.washington.edu
> wrote:

> Hi Rohin,
> Ah, I see. I don't think a BallTree is the right data structure for an
> anisotropic N-point query, because it fundamentally assumes spherical
> symmetry of the metric. You may be able to do something like this with a
> specialized KD-tree, but scikit-learn doesn't support this, and I don't
> imagine that it ever will given the very specialized nature of the
> application.
>
> I'm certain someone has written efficient code for this operation in the
> astronomy community, but I don't know of any good Python package to
> recommend for this ? I'd suggest googling for keywords and seeing where
> that gets you.
>
> Thanks,
>   Jake
>
>  Jake VanderPlas
>  Senior Data Science Fellow
>  Director of Open Software
>  University of Washington eScience Institute
>
> On Tue, Aug 1, 2017 at 6:15 AM, Rohin Kumar <yrohinkumar at gmail.com> wrote:
>
>> Since you seem to be from Astrophysics/Cosmology background (I am
>> assuming you are jakevdp - the creator of astroML - if you are - I am
>> lucky!), I can explain my application scenario. I am trying to calculate
>> the anisotropic two-point correlation function something like done in
>> rp_pi_tpcf
>> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.rp_pi_tpcf.html#halotools.mock_observables.rp_pi_tpcf>
>>  or s_mu_tpcf
>> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.s_mu_tpcf.html#halotools.mock_observables.s_mu_tpcf>
>> using pairs (DD,DR,RR) calculated from BallTree.two_point_correlation
>>
>> In halotools (http://halotools.readthedocs.io/en/latest/function_usage/mo
>> ck_observables_functions.html) it is implemented using rectangular
>> grids. I could calculate 2pcf with custom metrics using one variable with
>> BallTree as done in astroML. I intend to find the anisotropic counter part.
>>
>> Thanks & Regards,
>> Rohin
>>
>>
>> On Tue, Aug 1, 2017 at 5:18 PM, Rohin Kumar <yrohinkumar at gmail.com>
>> wrote:
>>
>>> Dear Jake,
>>>
>>> Thanks for your response. I meant to group/count pairs in boxes (using
>>> two arrays simultaneously-hence needing 2 metrics) instead of one distance
>>> array as the binning parameter. I don't know if the algorithm supports such
>>> a thing. For now, I am proceeding with your suggestion of two ball trees at
>>> huge computational cost. I hope I am able to frame my question properly.
>>>
>>> Thanks & Regards,
>>> Rohin.
>>>
>>>
>>>
>>> On Mon, Jul 31, 2017 at 8:16 PM, Jacob Vanderplas <
>>> jakevdp at cs.washington.edu> wrote:
>>>
>>>> On Sun, Jul 30, 2017 at 11:18 AM, Rohin Kumar <yrohinkumar at gmail.com>
>>>> wrote:
>>>>
>>>>> *update*
>>>>>
>>>>> May be it doesn't have to be done at the tree creation level. It could
>>>>> be using loops and creating two different balltrees. Something like
>>>>>
>>>>> tree1=BallTree(X,metric='metric1') #for x-z plane
>>>>> tree2=BallTree(X,metric='metric2') #for y-z plane
>>>>>
>>>>> And then calculate correlation functions in a loop to get
>>>>> tpcf(X,r1,r2) using tree1.two_point_correlation(X,r1) and
>>>>> tree2.two_point_correlation(X,r2)
>>>>>
>>>>
>>>> Hi Rohin,
>>>> It's not exactly clear to me what you wish the tree to do with the two
>>>> different metrics, but in any case the ball tree only supports one metric
>>>> at a time. If you can construct your desired result from two ball trees
>>>> each with its own metric, then that's probably the best way to proceed,
>>>>    Jake
>>>>
>>>>
>>>>>
>>>>> _______________________________________________
>>>>> scikit-learn mailing list
>>>>> scikit-learn at python.org
>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170801/61fc6f04/attachment-0001.html>

From jakevdp at cs.washington.edu  Tue Aug  1 13:59:21 2017
From: jakevdp at cs.washington.edu (Jacob Vanderplas)
Date: Tue, 1 Aug 2017 10:59:21 -0700
Subject: [scikit-learn] Nearest neighbor search with 2 distance measures
In-Reply-To: <CAARwKuupVuwYtBCZPbQ58gggHmTKn1ReQ=ZSCqjZQC40dFTeXw@mail.gmail.com>
References: <379121501436421@mxfront4j.mail.yandex.net>
 <CAARwKuu+zHGw61=q1t8rO5F-KDUoi+Y7bCFC3SRYBgJAEvyXvw@mail.gmail.com>
 <CACpqBg0uRrMxWENyz1vgHwDXQiSgU6Kdeh+=ATao29e2Lz0EEw@mail.gmail.com>
 <CAARwKuszb0QxfNce+b5640+3OuG=Mk26Mnt3cEiQ7Lzx80=BJA@mail.gmail.com>
 <CAARwKut+OH-vHUn3coTySrKA=bk3prhe1btkjBpJRb5P3iwEug@mail.gmail.com>
 <CACpqBg3dSm_FcQLuhy=5Pes3PSYCwYZBB9cjvQrgoOGakwv0SA@mail.gmail.com>
 <CAARwKuupVuwYtBCZPbQ58gggHmTKn1ReQ=ZSCqjZQC40dFTeXw@mail.gmail.com>
Message-ID: <CACpqBg10=c4Skkw1SZQdPw4k5cEACEmhPAXp_DRPvYQNj-_4ig@mail.gmail.com>

On Tue, Aug 1, 2017 at 10:50 AM, Rohin Kumar <yrohinkumar at gmail.com> wrote:

> I started with KD-tree but after realising it doesn't support custom
> metrics (I don't know the reason for this - would be nice feature)
>

The scikit-learn KD-tree doesn't support custom metrics because it utilizes
relatively strong assumptions about the form of the metric when
constructing the tree. The Ball Tree makes fewer assumptions, which is why
it can support arbitrary metrics. It would in principal be possible to
create a KD Tree that supports custom *axis-aligned* metrics, but again I
think that would be too specialized for inclusion in scikit-learn.

One project you might check out is cykdtree:
https://pypi.python.org/pypi/cykdtree
I'm not certain whether it supports the queries you need, but I would bet
the team behind that would be willing to work toward these sorts of
specialized queries if they don't already exist.

   Jake



> I shifted to BallTree and was looking for a 2 metric based categorisation.
> After looking around, the best I could find at most were brute-force
> methods written in python (had my own version too) or better optimised ones
> in C or FORTRAN. The closest one was halotools which again works with
> euclidean metric. For now, I will try to get my work done with 2 different
> BallTrees iteratively in bins. If I find a better option will try to post
> an update.
>
> Regards,
> Rohin.
>
>
> On Tue, Aug 1, 2017 at 10:55 PM, Jacob Vanderplas <
> jakevdp at cs.washington.edu> wrote:
>
>> Hi Rohin,
>> Ah, I see. I don't think a BallTree is the right data structure for an
>> anisotropic N-point query, because it fundamentally assumes spherical
>> symmetry of the metric. You may be able to do something like this with a
>> specialized KD-tree, but scikit-learn doesn't support this, and I don't
>> imagine that it ever will given the very specialized nature of the
>> application.
>>
>> I'm certain someone has written efficient code for this operation in the
>> astronomy community, but I don't know of any good Python package to
>> recommend for this ? I'd suggest googling for keywords and seeing where
>> that gets you.
>>
>> Thanks,
>>   Jake
>>
>>  Jake VanderPlas
>>  Senior Data Science Fellow
>>  Director of Open Software
>>  University of Washington eScience Institute
>>
>> On Tue, Aug 1, 2017 at 6:15 AM, Rohin Kumar <yrohinkumar at gmail.com>
>> wrote:
>>
>>> Since you seem to be from Astrophysics/Cosmology background (I am
>>> assuming you are jakevdp - the creator of astroML - if you are - I am
>>> lucky!), I can explain my application scenario. I am trying to calculate
>>> the anisotropic two-point correlation function something like done in
>>> rp_pi_tpcf
>>> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.rp_pi_tpcf.html#halotools.mock_observables.rp_pi_tpcf>
>>>  or s_mu_tpcf
>>> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.s_mu_tpcf.html#halotools.mock_observables.s_mu_tpcf>
>>> using pairs (DD,DR,RR) calculated from BallTree.two_point_correlation
>>>
>>> In halotools (http://halotools.readthedocs.
>>> io/en/latest/function_usage/mock_observables_functions.html) it is
>>> implemented using rectangular grids. I could calculate 2pcf with custom
>>> metrics using one variable with BallTree as done in astroML. I intend to
>>> find the anisotropic counter part.
>>>
>>> Thanks & Regards,
>>> Rohin
>>>
>>>
>>> On Tue, Aug 1, 2017 at 5:18 PM, Rohin Kumar <yrohinkumar at gmail.com>
>>> wrote:
>>>
>>>> Dear Jake,
>>>>
>>>> Thanks for your response. I meant to group/count pairs in boxes (using
>>>> two arrays simultaneously-hence needing 2 metrics) instead of one distance
>>>> array as the binning parameter. I don't know if the algorithm supports such
>>>> a thing. For now, I am proceeding with your suggestion of two ball trees at
>>>> huge computational cost. I hope I am able to frame my question properly.
>>>>
>>>> Thanks & Regards,
>>>> Rohin.
>>>>
>>>>
>>>>
>>>> On Mon, Jul 31, 2017 at 8:16 PM, Jacob Vanderplas <
>>>> jakevdp at cs.washington.edu> wrote:
>>>>
>>>>> On Sun, Jul 30, 2017 at 11:18 AM, Rohin Kumar <yrohinkumar at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> *update*
>>>>>>
>>>>>> May be it doesn't have to be done at the tree creation level. It
>>>>>> could be using loops and creating two different balltrees. Something like
>>>>>>
>>>>>> tree1=BallTree(X,metric='metric1') #for x-z plane
>>>>>> tree2=BallTree(X,metric='metric2') #for y-z plane
>>>>>>
>>>>>> And then calculate correlation functions in a loop to get
>>>>>> tpcf(X,r1,r2) using tree1.two_point_correlation(X,r1) and
>>>>>> tree2.two_point_correlation(X,r2)
>>>>>>
>>>>>
>>>>> Hi Rohin,
>>>>> It's not exactly clear to me what you wish the tree to do with the two
>>>>> different metrics, but in any case the ball tree only supports one metric
>>>>> at a time. If you can construct your desired result from two ball trees
>>>>> each with its own metric, then that's probably the best way to proceed,
>>>>>    Jake
>>>>>
>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> scikit-learn mailing list
>>>>>> scikit-learn at python.org
>>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> scikit-learn mailing list
>>>>> scikit-learn at python.org
>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>
>>>>>
>>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170801/969160c2/attachment-0001.html>

From sambarnett95 at gmail.com  Wed Aug  2 08:38:50 2017
From: sambarnett95 at gmail.com (Sam Barnett)
Date: Wed, 2 Aug 2017 13:38:50 +0100
Subject: [scikit-learn] Problems with running GridSearchCV on a pipeline
 with a custom transformer
Message-ID: <CACWGMfo_O3ySYr8vPZ8CpUeZ1XO=Ykq+XkjpOq4OxrdBgEzDbw@mail.gmail.com>

Dear all,

I have created a 2-step pipeline with a custom transformer followed by a
simple SVC classifier, and I wish to run a grid-search over it. I am able
to successfully create the transformer and the pipeline, and each of these
elements work fine. However, when I try to use the fit() method on my
GridSearchCV object, I get the following error:

     57         # during fit.
     58         if X.shape != self.input_shape_:
---> 59             raise ValueError('Shape of input is different from what
was seen '
     60                              'in `fit`')
     61

ValueError: Shape of input is different from what was seen in `fit`

For a full breakdown of the problem, I have written a Jupyter notebook
showing exactly how the error occurs (this also contains all .py files
necessary to run the notebook). Can anybody see how to work through this?

Many thanks,
Sam Barnett
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/1fc59f4b/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Sequential Kernel Test.zip
Type: application/zip
Size: 6759 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/1fc59f4b/attachment.zip>

From viewsonic234 at aim.com  Wed Aug  2 11:36:24 2017
From: viewsonic234 at aim.com (Chris Carrion)
Date: Wed, 2 Aug 2017 11:36:24 -0400
Subject: [scikit-learn] minibatchkmeans deprecation warning?
Message-ID: <3xMy9f2YqXzFqm1@mail.python.org>

Hi,

I?m working in an environment provided by Quantopian, an algorithmic-traders hub for research. I imported the minibatch kmeans from sklearn.clusters in the environment they provided, but I?m getting a deprecation warning. After reaching out to Quantopian support, they claim it?s something with the way sklearn is coded, and nothing can be done on their end. I was wondering whether this was true or not.

Curious,
Chris
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/aba670ec/attachment.html>

From t3kcit at gmail.com  Wed Aug  2 12:05:17 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Wed, 2 Aug 2017 12:05:17 -0400
Subject: [scikit-learn] Problems with running GridSearchCV on a pipeline
 with a custom transformer
In-Reply-To: <CACWGMfo_O3ySYr8vPZ8CpUeZ1XO=Ykq+XkjpOq4OxrdBgEzDbw@mail.gmail.com>
References: <CACWGMfo_O3ySYr8vPZ8CpUeZ1XO=Ykq+XkjpOq4OxrdBgEzDbw@mail.gmail.com>
Message-ID: <e36c375d-97d8-56b0-ddb2-442bfc1f1128@gmail.com>

Hi Sam.
GridSearchCV will do cross-validation, which requires to "transform" the 
test data.
The shape of the test-data will be different from the shape of the 
training data.
You need to have the ability to compute the kernel between the training 
data and new test data.

A more hacky solution would be to compute the full kernel matrix in 
advance and pass that to GridSearchCV.

You probably don't need it here, but you should also checkout what the 
_pairwise attribute does in cross-validation,
because that it likely to come up when playing with kernels.

Hth,
Andy

On 08/02/2017 08:38 AM, Sam Barnett wrote:
> Dear all,
>
> I have created a 2-step pipeline with a custom transformer followed by 
> a simple SVC classifier, and I wish to run a grid-search over it. I am 
> able to successfully create the transformer and the pipeline, and each 
> of these elements work fine. However, when I try to use the fit() 
> method on my GridSearchCV object, I get the following error:
>
> 57 # during fit.
>    58 if X.shape != self.input_shape_:
> ---> 59             raise ValueError('Shape of input is different from 
> what was seen '
>    60                              'in `fit`')
>    61
>
> ValueError: Shape of input is different from what was seen in `fit`
>
> For a full breakdown of the problem, I have written a Jupyter notebook 
> showing exactly how the error occurs (this also contains all .py files 
> necessary to run the notebook). Can anybody see how to work through this?
>
> Many thanks,
> Sam Barnett
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/4b175486/attachment.html>

From t3kcit at gmail.com  Wed Aug  2 12:05:44 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Wed, 2 Aug 2017 12:05:44 -0400
Subject: [scikit-learn] minibatchkmeans deprecation warning?
In-Reply-To: <3xMy9f2YqXzFqm1@mail.python.org>
References: <3xMy9f2YqXzFqm1@mail.python.org>
Message-ID: <ed581a43-0511-1dec-c14d-a6d56117f6ed@gmail.com>

Hi Chris.

What is the warning?

Andy

On 08/02/2017 11:36 AM, Chris Carrion via scikit-learn wrote:
>
> Hi,
>
> I?m working in an environment provided by Quantopian, an 
> algorithmic-traders hub for research. I imported the minibatch kmeans 
> from sklearn.clusters in the environment they provided, but I?m 
> getting a deprecation warning. After reaching out to Quantopian 
> support, they claim it?s something with the way sklearn is coded, and 
> nothing can be done on their end. I was wondering whether this was 
> true or not.
>
> Curious,
>
> Chris
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/a4f48e09/attachment.html>

From viewsonic234 at aim.com  Wed Aug  2 12:10:30 2017
From: viewsonic234 at aim.com (Chris Carrion)
Date: Wed, 2 Aug 2017 12:10:30 -0400
Subject: [scikit-learn] minibatchkmeans deprecation warning?
In-Reply-To: <ed581a43-0511-1dec-c14d-a6d56117f6ed@gmail.com>
References: <3xMy9f2YqXzFqm1@mail.python.org>
 <ed581a43-0511-1dec-c14d-a6d56117f6ed@gmail.com>
Message-ID: <3xMypN0FjFzFqw2@mail.python.org>

Hi Andy,
WARN sklearn/cluster/k_means_.py:1301: DeprecationWarning: This function is deprecated. Please call randint(0, 179 + 1) instead

That?s all I?m given
From: Andreas Mueller
Sent: Wednesday, August 2, 2017 12:09 PM
To: Chris Carrion via scikit-learn
Subject: Re: [scikit-learn] minibatchkmeans deprecation warning?

Hi Chris.

What is the warning?

Andy
On 08/02/2017 11:36 AM, Chris Carrion via scikit-learn wrote:
Hi,
?
I?m working in an environment provided by Quantopian, an algorithmic-traders hub for research. I imported the minibatch kmeans from sklearn.clusters in the environment they provided, but I?m getting a deprecation warning. After reaching out to Quantopian support, they claim it?s something with the way sklearn is coded, and nothing can be done on their end. I was wondering whether this was true or not.
?
Curious,
Chris



_______________________________________________
scikit-learn mailing list
scikit-learn at python.org
https://mail.python.org/mailman/listinfo/scikit-learn


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/68eb6b4c/attachment.html>

From t3kcit at gmail.com  Wed Aug  2 12:32:03 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Wed, 2 Aug 2017 12:32:03 -0400
Subject: [scikit-learn] minibatchkmeans deprecation warning?
In-Reply-To: <3xMypN0FjFzFqw2@mail.python.org>
References: <3xMy9f2YqXzFqm1@mail.python.org>
 <ed581a43-0511-1dec-c14d-a6d56117f6ed@gmail.com>
 <3xMypN0FjFzFqw2@mail.python.org>
Message-ID: <66043d0e-dce5-ebac-a100-31bc02760aa3@gmail.com>

Ah.
That's actually a deprecation warning coming from numpy, and it think 
it'll be removed in 0.19 (if not already in 0.18.1).
It's really nothing to worry about, though.

Andy


On 08/02/2017 12:10 PM, Chris Carrion via scikit-learn wrote:
>
> Hi Andy,
>
> WARNsklearn/cluster/k_means_.py:1301: DeprecationWarning: This 
> function is deprecated. Please call randint(0, 179 + 1) instead
>
> That?s all I?m given
>
> *From: *Andreas Mueller <mailto:t3kcit at gmail.com>
> *Sent: *Wednesday, August 2, 2017 12:09 PM
> *To: *Chris Carrion via scikit-learn <mailto:scikit-learn at python.org>
> *Subject: *Re: [scikit-learn] minibatchkmeans deprecation warning?
>
> Hi Chris.
>
> What is the warning?
>
> Andy
>
> On 08/02/2017 11:36 AM, Chris Carrion via scikit-learn wrote:
>
>     Hi,
>
>     I?m working in an environment provided by Quantopian, an
>     algorithmic-traders hub for research. I imported the minibatch
>     kmeans from sklearn.clusters in the environment they provided, but
>     I?m getting a deprecation warning. After reaching out to
>     Quantopian support, they claim it?s something with the way sklearn
>     is coded, and nothing can be done on their end. I was wondering
>     whether this was true or not.
>
>     Curious,
>
>     Chris
>
>
>
>
>     _______________________________________________
>
>     scikit-learn mailing list
>
>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>
>     https://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/4a62c311/attachment-0001.html>

From viewsonic234 at aim.com  Wed Aug  2 12:38:47 2017
From: viewsonic234 at aim.com (Chris Carrion)
Date: Wed, 2 Aug 2017 12:38:47 -0400
Subject: [scikit-learn] minibatchkmeans deprecation warning?
In-Reply-To: <66043d0e-dce5-ebac-a100-31bc02760aa3@gmail.com>
References: <3xMy9f2YqXzFqm1@mail.python.org>
 <ed581a43-0511-1dec-c14d-a6d56117f6ed@gmail.com>
 <3xMypN0FjFzFqw2@mail.python.org>
 <66043d0e-dce5-ebac-a100-31bc02760aa3@gmail.com>
Message-ID: <3xMzR06sl2zFqwt@mail.python.org>

That?s great to hear, thanks!

Chris

From: Andreas Mueller
Sent: Wednesday, August 2, 2017 12:34 PM
To: Chris Carrion via scikit-learn
Subject: Re: [scikit-learn] minibatchkmeans deprecation warning?

Ah.
That's actually a deprecation warning coming from numpy, and it think it'll be removed in 0.19 (if not already in 0.18.1).
It's really nothing to worry about, though.

Andy

On 08/02/2017 12:10 PM, Chris Carrion via scikit-learn wrote:
Hi Andy,
WARN sklearn/cluster/k_means_.py:1301: DeprecationWarning: This function is deprecated. Please call randint(0, 179 + 1) instead
?
That?s all I?m given
From: Andreas Mueller
Sent: Wednesday, August 2, 2017 12:09 PM
To: Chris Carrion via scikit-learn
Subject: Re: [scikit-learn] minibatchkmeans deprecation warning?
?
Hi Chris.

What is the warning?

Andy
On 08/02/2017 11:36 AM, Chris Carrion via scikit-learn wrote:
Hi,
?
I?m working in an environment provided by Quantopian, an algorithmic-traders hub for research. I imported the minibatch kmeans from sklearn.clusters in the environment they provided, but I?m getting a deprecation warning. After reaching out to Quantopian support, they claim it?s something with the way sklearn is coded, and nothing can be done on their end. I was wondering whether this was true or not.
?
Curious,
Chris




_______________________________________________
scikit-learn mailing list
scikit-learn at python.org
https://mail.python.org/mailman/listinfo/scikit-learn
?
?



_______________________________________________
scikit-learn mailing list
scikit-learn at python.org
https://mail.python.org/mailman/listinfo/scikit-learn


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/fc74e787/attachment.html>

From viewsonic234 at aim.com  Wed Aug  2 12:48:06 2017
From: viewsonic234 at aim.com (Chris Carrion)
Date: Wed, 2 Aug 2017 12:48:06 -0400
Subject: [scikit-learn] minibatchkmeans deprecation warning?
In-Reply-To: <66043d0e-dce5-ebac-a100-31bc02760aa3@gmail.com>
References: <3xMy9f2YqXzFqm1@mail.python.org>
 <ed581a43-0511-1dec-c14d-a6d56117f6ed@gmail.com>
 <3xMypN0FjFzFqw2@mail.python.org>
 <66043d0e-dce5-ebac-a100-31bc02760aa3@gmail.com>
Message-ID: <3xMzdn0TcWzFqVr@mail.python.org>

Before I forget, is there an ETA for .19, or an average time between upgrades?


From: Andreas Mueller
Sent: Wednesday, August 2, 2017 12:34 PM
To: Chris Carrion via scikit-learn
Subject: Re: [scikit-learn] minibatchkmeans deprecation warning?

Ah.
That's actually a deprecation warning coming from numpy, and it think it'll be removed in 0.19 (if not already in 0.18.1).
It's really nothing to worry about, though.

Andy

On 08/02/2017 12:10 PM, Chris Carrion via scikit-learn wrote:
Hi Andy,
WARN sklearn/cluster/k_means_.py:1301: DeprecationWarning: This function is deprecated. Please call randint(0, 179 + 1) instead
?
That?s all I?m given
From: Andreas Mueller
Sent: Wednesday, August 2, 2017 12:09 PM
To: Chris Carrion via scikit-learn
Subject: Re: [scikit-learn] minibatchkmeans deprecation warning?
?
Hi Chris.

What is the warning?

Andy
On 08/02/2017 11:36 AM, Chris Carrion via scikit-learn wrote:
Hi,
?
I?m working in an environment provided by Quantopian, an algorithmic-traders hub for research. I imported the minibatch kmeans from sklearn.clusters in the environment they provided, but I?m getting a deprecation warning. After reaching out to Quantopian support, they claim it?s something with the way sklearn is coded, and nothing can be done on their end. I was wondering whether this was true or not.
?
Curious,
Chris




_______________________________________________
scikit-learn mailing list
scikit-learn at python.org
https://mail.python.org/mailman/listinfo/scikit-learn
?
?



_______________________________________________
scikit-learn mailing list
scikit-learn at python.org
https://mail.python.org/mailman/listinfo/scikit-learn


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/a30d3ade/attachment.html>

From t3kcit at gmail.com  Wed Aug  2 14:36:02 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Wed, 2 Aug 2017 14:36:02 -0400
Subject: [scikit-learn] minibatchkmeans deprecation warning?
In-Reply-To: <3xMzdn0TcWzFqVr@mail.python.org>
References: <3xMy9f2YqXzFqm1@mail.python.org>
 <ed581a43-0511-1dec-c14d-a6d56117f6ed@gmail.com>
 <3xMypN0FjFzFqw2@mail.python.org>
 <66043d0e-dce5-ebac-a100-31bc02760aa3@gmail.com>
 <3xMzdn0TcWzFqVr@mail.python.org>
Message-ID: <31bc0362-f3b4-94af-b240-0a1d4bb9e7e0@gmail.com>

The docs say 3 month, I think. Though it's been more like 8.
0.19 will come out in August.


On 08/02/2017 12:48 PM, Chris Carrion via scikit-learn wrote:
>
> Before I forget, is there an ETA for .19, or an average time between 
> upgrades?
>
> *From: *Andreas Mueller <mailto:t3kcit at gmail.com>
> *Sent: *Wednesday, August 2, 2017 12:34 PM
> *To: *Chris Carrion via scikit-learn <mailto:scikit-learn at python.org>
> *Subject: *Re: [scikit-learn] minibatchkmeans deprecation warning?
>
> Ah.
> That's actually a deprecation warning coming from numpy, and it think 
> it'll be removed in 0.19 (if not already in 0.18.1).
> It's really nothing to worry about, though.
>
> Andy
>
> On 08/02/2017 12:10 PM, Chris Carrion via scikit-learn wrote:
>
>     Hi Andy,
>
>     WARNsklearn/cluster/k_means_.py:1301: DeprecationWarning: This
>     function is deprecated. Please call randint(0, 179 + 1) instead
>
>     That?s all I?m given
>
>     *From: *Andreas Mueller <mailto:t3kcit at gmail.com>
>     *Sent: *Wednesday, August 2, 2017 12:09 PM
>     *To: *Chris Carrion via scikit-learn <mailto:scikit-learn at python.org>
>     *Subject: *Re: [scikit-learn] minibatchkmeans deprecation warning?
>
>     Hi Chris.
>
>     What is the warning?
>
>     Andy
>
>     On 08/02/2017 11:36 AM, Chris Carrion via scikit-learn wrote:
>
>         Hi,
>
>         I?m working in an environment provided by Quantopian, an
>         algorithmic-traders hub for research. I imported the minibatch
>         kmeans from sklearn.clusters in the environment they provided,
>         but I?m getting a deprecation warning. After reaching out to
>         Quantopian support, they claim it?s something with the way
>         sklearn is coded, and nothing can be done on their end. I was
>         wondering whether this was true or not.
>
>         Curious,
>
>         Chris
>
>
>
>
>
>         _______________________________________________
>
>         scikit-learn mailing list
>
>         scikit-learn at python.org <mailto:scikit-learn at python.org>
>
>         https://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
>
>     _______________________________________________
>
>     scikit-learn mailing list
>
>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>
>     https://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/76e5f967/attachment-0001.html>

From sambarnett95 at gmail.com  Wed Aug  2 15:08:07 2017
From: sambarnett95 at gmail.com (Sam Barnett)
Date: Wed, 2 Aug 2017 20:08:07 +0100
Subject: [scikit-learn] Problems with running GridSearchCV on a pipeline
 with a custom transformer
In-Reply-To: <e36c375d-97d8-56b0-ddb2-442bfc1f1128@gmail.com>
References: <CACWGMfo_O3ySYr8vPZ8CpUeZ1XO=Ykq+XkjpOq4OxrdBgEzDbw@mail.gmail.com>
 <e36c375d-97d8-56b0-ddb2-442bfc1f1128@gmail.com>
Message-ID: <CACWGMfqv=HZxnxYao+tC5L9gc_a6OZe_hZOLNuEO37ac-7xfyA@mail.gmail.com>

Hi Andy,

The purpose of the transformer is to take an ordinary kernel (in this case
I have taken 'rbf' as a default) and return a 'sequentialised' kernel using
a few extra parameters. Hence, the transformer takes an ordinary
data-target pair X, y as its input, and the fit_transform(X, y) method will
output the Gram matrix for X that is associated with this sequentialised
kernel. In the pipeline, this Gram matrix is passed into an SVC classifier
with the kernel parameter set to 'precomputed'.

Therefore, I do not think your hacky solution would be possible. However, I
am still unsure how to implement your first solution: won't the Gram matrix
from the transformer contain all the necessary kernel values? Could you
elaborate further?


Best,
Sam

On Wed, Aug 2, 2017 at 5:05 PM, Andreas Mueller <t3kcit at gmail.com> wrote:

> Hi Sam.
> GridSearchCV will do cross-validation, which requires to "transform" the
> test data.
> The shape of the test-data will be different from the shape of the
> training data.
> You need to have the ability to compute the kernel between the training
> data and new test data.
>
> A more hacky solution would be to compute the full kernel matrix in
> advance and pass that to GridSearchCV.
>
> You probably don't need it here, but you should also checkout what the
> _pairwise attribute does in cross-validation,
> because that it likely to come up when playing with kernels.
>
> Hth,
> Andy
>
>
> On 08/02/2017 08:38 AM, Sam Barnett wrote:
>
> Dear all,
>
> I have created a 2-step pipeline with a custom transformer followed by a
> simple SVC classifier, and I wish to run a grid-search over it. I am able
> to successfully create the transformer and the pipeline, and each of these
> elements work fine. However, when I try to use the fit() method on my
> GridSearchCV object, I get the following error:
>
>      57         # during fit.
>      58         if X.shape != self.input_shape_:
> ---> 59             raise ValueError('Shape of input is different from
> what was seen '
>      60                              'in `fit`')
>      61
>
> ValueError: Shape of input is different from what was seen in `fit`
>
> For a full breakdown of the problem, I have written a Jupyter notebook
> showing exactly how the error occurs (this also contains all .py files
> necessary to run the notebook). Can anybody see how to work through this?
>
> Many thanks,
> Sam Barnett
>
>
>
> _______________________________________________
> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/d0cbe35e/attachment.html>

From pybokeh at gmail.com  Wed Aug  2 22:01:36 2017
From: pybokeh at gmail.com (pybokeh)
Date: Wed, 2 Aug 2017 22:01:36 -0400
Subject: [scikit-learn] Help With Text Classification
Message-ID: <CAOoMBViVEfV61SJaUq4Z+z6sQegQc8w=Hf_4vb-X08Es1Jr1cA@mail.gmail.com>

Hello,
I am studying this example from scikit-learn's site:
http://scikit-learn.org/stable/tutorial/text_analytics/working_with_text_
data.html

The problem that I need to solve is very similar to this example, except I
have one
additional feature column (part #) that is categorical of type string.  My
label or target
values consist of just 2 values: 0 or 1.

With that additional feature column, I am transforming it with a
LabelEncoder and
then I am encoding it with the OneHotEncoder.

Then I am concatenating that one-hot encoded column (part #) to the
text/document
feature column (complaint), which I had applied the CountVectorizer and
TfidfTransformer transformations.

Then I chose the MultinomialNB model to fit my concatenated training data
with.

The problem I run into is when I invoke the prediction, I get a dimension
mis-match error.

Here's my jupyter notebook gist:
http://nbviewer.jupyter.org/gist/anonymous/59ba930a783571c85ef86ba41424b311

I would gladly appreciate it if someone can guide me where I went wrong.
Thanks!

- Daniel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/172acff4/attachment.html>

From joel.nothman at gmail.com  Wed Aug  2 22:38:34 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Thu, 3 Aug 2017 12:38:34 +1000
Subject: [scikit-learn] Help With Text Classification
In-Reply-To: <CAOoMBViVEfV61SJaUq4Z+z6sQegQc8w=Hf_4vb-X08Es1Jr1cA@mail.gmail.com>
References: <CAOoMBViVEfV61SJaUq4Z+z6sQegQc8w=Hf_4vb-X08Es1Jr1cA@mail.gmail.com>
Message-ID: <CAAkaFLUuKF7ek2+wb2+EpBbxCmaUnENC=tJx87fU2vQDivqZ9A@mail.gmail.com>

Use a Pipeline to help avoid this kind of issue (and others). You might
also want to do something like
http://scikit-learn.org/stable/auto_examples/hetero_feature_union.html

On 3 August 2017 at 12:01, pybokeh <pybokeh at gmail.com> wrote:

> Hello,
> I am studying this example from scikit-learn's site:
> http://scikit-learn.org/stable/tutorial/text_analytics/
> working_with_text_data.html
>
> The problem that I need to solve is very similar to this example, except I
> have one
> additional feature column (part #) that is categorical of type string.  My
> label or target
> values consist of just 2 values: 0 or 1.
>
> With that additional feature column, I am transforming it with a
> LabelEncoder and
> then I am encoding it with the OneHotEncoder.
>
> Then I am concatenating that one-hot encoded column (part #) to the
> text/document
> feature column (complaint), which I had applied the CountVectorizer and
> TfidfTransformer transformations.
>
> Then I chose the MultinomialNB model to fit my concatenated training data
> with.
>
> The problem I run into is when I invoke the prediction, I get a dimension
> mis-match error.
>
> Here's my jupyter notebook gist:
> http://nbviewer.jupyter.org/gist/anonymous/59ba930a783571c85
> ef86ba41424b311
>
> I would gladly appreciate it if someone can guide me where I went wrong.
> Thanks!
>
> - Daniel
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/1746a230/attachment-0001.html>

From pybokeh at gmail.com  Wed Aug  2 23:12:36 2017
From: pybokeh at gmail.com (pybokeh)
Date: Wed, 2 Aug 2017 23:12:36 -0400
Subject: [scikit-learn] Help With Text Classification
In-Reply-To: <CAAkaFLUuKF7ek2+wb2+EpBbxCmaUnENC=tJx87fU2vQDivqZ9A@mail.gmail.com>
References: <CAOoMBViVEfV61SJaUq4Z+z6sQegQc8w=Hf_4vb-X08Es1Jr1cA@mail.gmail.com>
 <CAAkaFLUuKF7ek2+wb2+EpBbxCmaUnENC=tJx87fU2vQDivqZ9A@mail.gmail.com>
Message-ID: <CAOoMBViLs0=nStMaU+KvGnXY+pexoXMScAynBC1g8E4U-1iZyw@mail.gmail.com>

Thanks Joel for recommending FeatureUnion.  I did run across that.  But for
just 2 features, I thought that might be overkill.  I am aware of Pipeline
which the scikit-learn example explains very well, which I was going to
utilize once I finalize my script.  I did not want to abstract away too
much early on since I am in the beginning stages of learning machine
learning and scikit-learn.

- Daniel

On Wed, Aug 2, 2017 at 10:38 PM, Joel Nothman <joel.nothman at gmail.com>
wrote:

> Use a Pipeline to help avoid this kind of issue (and others). You might
> also want to do something like http://scikit-learn.org/
> stable/auto_examples/hetero_feature_union.html
>
> On 3 August 2017 at 12:01, pybokeh <pybokeh at gmail.com> wrote:
>
>> Hello,
>> I am studying this example from scikit-learn's site:
>> http://scikit-learn.org/stable/tutorial/text_analytics/worki
>> ng_with_text_data.html
>>
>> The problem that I need to solve is very similar to this example, except
>> I have one
>> additional feature column (part #) that is categorical of type string.
>> My label or target
>> values consist of just 2 values: 0 or 1.
>>
>> With that additional feature column, I am transforming it with a
>> LabelEncoder and
>> then I am encoding it with the OneHotEncoder.
>>
>> Then I am concatenating that one-hot encoded column (part #) to the
>> text/document
>> feature column (complaint), which I had applied the CountVectorizer and
>> TfidfTransformer transformations.
>>
>> Then I chose the MultinomialNB model to fit my concatenated training data
>> with.
>>
>> The problem I run into is when I invoke the prediction, I get a dimension
>> mis-match error.
>>
>> Here's my jupyter notebook gist:
>> http://nbviewer.jupyter.org/gist/anonymous/59ba930a783571c85
>> ef86ba41424b311
>>
>> I would gladly appreciate it if someone can guide me where I went wrong.
>> Thanks!
>>
>> - Daniel
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170802/10685ecf/attachment.html>

From yrohinkumar at gmail.com  Wed Aug  2 23:42:58 2017
From: yrohinkumar at gmail.com (Rohin Kumar)
Date: Thu, 3 Aug 2017 09:12:58 +0530
Subject: [scikit-learn] Nearest neighbor search with 2 distance measures
In-Reply-To: <CACpqBg10=c4Skkw1SZQdPw4k5cEACEmhPAXp_DRPvYQNj-_4ig@mail.gmail.com>
References: <379121501436421@mxfront4j.mail.yandex.net>
 <CAARwKuu+zHGw61=q1t8rO5F-KDUoi+Y7bCFC3SRYBgJAEvyXvw@mail.gmail.com>
 <CACpqBg0uRrMxWENyz1vgHwDXQiSgU6Kdeh+=ATao29e2Lz0EEw@mail.gmail.com>
 <CAARwKuszb0QxfNce+b5640+3OuG=Mk26Mnt3cEiQ7Lzx80=BJA@mail.gmail.com>
 <CAARwKut+OH-vHUn3coTySrKA=bk3prhe1btkjBpJRb5P3iwEug@mail.gmail.com>
 <CACpqBg3dSm_FcQLuhy=5Pes3PSYCwYZBB9cjvQrgoOGakwv0SA@mail.gmail.com>
 <CAARwKuupVuwYtBCZPbQ58gggHmTKn1ReQ=ZSCqjZQC40dFTeXw@mail.gmail.com>
 <CACpqBg10=c4Skkw1SZQdPw4k5cEACEmhPAXp_DRPvYQNj-_4ig@mail.gmail.com>
Message-ID: <CAARwKuu6n-hFdTXAx5Psue7XqVLH_xPa7eSznZJmS1Wk-XavGw@mail.gmail.com>

Dear Jake,

Thank you for your inputs. Had a look at cykdtree. Core implementation of
the algorithm is in C/C++ modifying which is currently beyond my skill.
Will try to contact their team if they entertain special requests. I should
be able fork and modify the sklearn algorithm in cython once my current
project is complete. Currently going ahead with brute-force method. For
now, this thread may be considered closed. Thanks once again!

Regards,
Rohin.



On Tue, Aug 1, 2017 at 11:29 PM, Jacob Vanderplas <jakevdp at cs.washington.edu
> wrote:

> On Tue, Aug 1, 2017 at 10:50 AM, Rohin Kumar <yrohinkumar at gmail.com>
> wrote:
>
>> I started with KD-tree but after realising it doesn't support custom
>> metrics (I don't know the reason for this - would be nice feature)
>>
>
> The scikit-learn KD-tree doesn't support custom metrics because it
> utilizes relatively strong assumptions about the form of the metric when
> constructing the tree. The Ball Tree makes fewer assumptions, which is why
> it can support arbitrary metrics. It would in principal be possible to
> create a KD Tree that supports custom *axis-aligned* metrics, but again I
> think that would be too specialized for inclusion in scikit-learn.
>
> One project you might check out is cykdtree: https://pypi.python.
> org/pypi/cykdtree
> I'm not certain whether it supports the queries you need, but I would bet
> the team behind that would be willing to work toward these sorts of
> specialized queries if they don't already exist.
>
>    Jake
>
>
>
>
>> I shifted to BallTree and was looking for a 2 metric based
>> categorisation. After looking around, the best I could find at most were
>> brute-force methods written in python (had my own version too) or better
>> optimised ones in C or FORTRAN. The closest one was halotools which again
>> works with euclidean metric. For now, I will try to get my work done with 2
>> different BallTrees iteratively in bins. If I find a better option will try
>> to post an update.
>>
>> Regards,
>> Rohin.
>>
>>
>> On Tue, Aug 1, 2017 at 10:55 PM, Jacob Vanderplas <
>> jakevdp at cs.washington.edu> wrote:
>>
>>> Hi Rohin,
>>> Ah, I see. I don't think a BallTree is the right data structure for an
>>> anisotropic N-point query, because it fundamentally assumes spherical
>>> symmetry of the metric. You may be able to do something like this with a
>>> specialized KD-tree, but scikit-learn doesn't support this, and I don't
>>> imagine that it ever will given the very specialized nature of the
>>> application.
>>>
>>> I'm certain someone has written efficient code for this operation in the
>>> astronomy community, but I don't know of any good Python package to
>>> recommend for this ? I'd suggest googling for keywords and seeing where
>>> that gets you.
>>>
>>> Thanks,
>>>   Jake
>>>
>>>  Jake VanderPlas
>>>  Senior Data Science Fellow
>>>  Director of Open Software
>>>  University of Washington eScience Institute
>>>
>>> On Tue, Aug 1, 2017 at 6:15 AM, Rohin Kumar <yrohinkumar at gmail.com>
>>> wrote:
>>>
>>>> Since you seem to be from Astrophysics/Cosmology background (I am
>>>> assuming you are jakevdp - the creator of astroML - if you are - I am
>>>> lucky!), I can explain my application scenario. I am trying to calculate
>>>> the anisotropic two-point correlation function something like done in
>>>> rp_pi_tpcf
>>>> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.rp_pi_tpcf.html#halotools.mock_observables.rp_pi_tpcf>
>>>>  or s_mu_tpcf
>>>> <http://halotools.readthedocs.io/en/latest/api/halotools.mock_observables.s_mu_tpcf.html#halotools.mock_observables.s_mu_tpcf>
>>>> using pairs (DD,DR,RR) calculated from BallTree.two_point_correlation
>>>>
>>>> In halotools (http://halotools.readthedocs.
>>>> io/en/latest/function_usage/mock_observables_functions.html) it is
>>>> implemented using rectangular grids. I could calculate 2pcf with custom
>>>> metrics using one variable with BallTree as done in astroML. I intend to
>>>> find the anisotropic counter part.
>>>>
>>>> Thanks & Regards,
>>>> Rohin
>>>>
>>>>
>>>> On Tue, Aug 1, 2017 at 5:18 PM, Rohin Kumar <yrohinkumar at gmail.com>
>>>> wrote:
>>>>
>>>>> Dear Jake,
>>>>>
>>>>> Thanks for your response. I meant to group/count pairs in boxes (using
>>>>> two arrays simultaneously-hence needing 2 metrics) instead of one distance
>>>>> array as the binning parameter. I don't know if the algorithm supports such
>>>>> a thing. For now, I am proceeding with your suggestion of two ball trees at
>>>>> huge computational cost. I hope I am able to frame my question properly.
>>>>>
>>>>> Thanks & Regards,
>>>>> Rohin.
>>>>>
>>>>>
>>>>>
>>>>> On Mon, Jul 31, 2017 at 8:16 PM, Jacob Vanderplas <
>>>>> jakevdp at cs.washington.edu> wrote:
>>>>>
>>>>>> On Sun, Jul 30, 2017 at 11:18 AM, Rohin Kumar <yrohinkumar at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> *update*
>>>>>>>
>>>>>>> May be it doesn't have to be done at the tree creation level. It
>>>>>>> could be using loops and creating two different balltrees. Something like
>>>>>>>
>>>>>>> tree1=BallTree(X,metric='metric1') #for x-z plane
>>>>>>> tree2=BallTree(X,metric='metric2') #for y-z plane
>>>>>>>
>>>>>>> And then calculate correlation functions in a loop to get
>>>>>>> tpcf(X,r1,r2) using tree1.two_point_correlation(X,r1) and
>>>>>>> tree2.two_point_correlation(X,r2)
>>>>>>>
>>>>>>
>>>>>> Hi Rohin,
>>>>>> It's not exactly clear to me what you wish the tree to do with the
>>>>>> two different metrics, but in any case the ball tree only supports one
>>>>>> metric at a time. If you can construct your desired result from two ball
>>>>>> trees each with its own metric, then that's probably the best way to
>>>>>> proceed,
>>>>>>    Jake
>>>>>>
>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> scikit-learn mailing list
>>>>>>> scikit-learn at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> scikit-learn mailing list
>>>>>> scikit-learn at python.org
>>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>>
>>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/cb0f5db7/attachment-0001.html>

From joel.nothman at gmail.com  Thu Aug  3 00:54:18 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Thu, 3 Aug 2017 14:54:18 +1000
Subject: [scikit-learn] Help With Text Classification
In-Reply-To: <CAOoMBViLs0=nStMaU+KvGnXY+pexoXMScAynBC1g8E4U-1iZyw@mail.gmail.com>
References: <CAOoMBViVEfV61SJaUq4Z+z6sQegQc8w=Hf_4vb-X08Es1Jr1cA@mail.gmail.com>
 <CAAkaFLUuKF7ek2+wb2+EpBbxCmaUnENC=tJx87fU2vQDivqZ9A@mail.gmail.com>
 <CAOoMBViLs0=nStMaU+KvGnXY+pexoXMScAynBC1g8E4U-1iZyw@mail.gmail.com>
Message-ID: <CAAkaFLX5+FHyX=x1JGiRLD39WPNM5ngcQDuGJ7jHsT40rnibPA@mail.gmail.com>

One of the key advantages of Pipeline is that it makes sure that equivalent
processing happens at training and prediction time (assuming you do not
write your own transformers that break their contract). This is what
appears to have broken in your current attempts.

On 3 August 2017 at 13:12, pybokeh <pybokeh at gmail.com> wrote:

> Thanks Joel for recommending FeatureUnion.  I did run across that.  But
> for just 2 features, I thought that might be overkill.  I am aware of
> Pipeline which the scikit-learn example explains very well, which I was
> going to utilize once I finalize my script.  I did not want to abstract
> away too much early on since I am in the beginning stages of learning
> machine learning and scikit-learn.
>
> - Daniel
>
> On Wed, Aug 2, 2017 at 10:38 PM, Joel Nothman <joel.nothman at gmail.com>
> wrote:
>
>> Use a Pipeline to help avoid this kind of issue (and others). You might
>> also want to do something like http://scikit-learn.org/stable
>> /auto_examples/hetero_feature_union.html
>>
>> On 3 August 2017 at 12:01, pybokeh <pybokeh at gmail.com> wrote:
>>
>>> Hello,
>>> I am studying this example from scikit-learn's site:
>>> http://scikit-learn.org/stable/tutorial/text_analytics/worki
>>> ng_with_text_data.html
>>>
>>> The problem that I need to solve is very similar to this example, except
>>> I have one
>>> additional feature column (part #) that is categorical of type string.
>>> My label or target
>>> values consist of just 2 values: 0 or 1.
>>>
>>> With that additional feature column, I am transforming it with a
>>> LabelEncoder and
>>> then I am encoding it with the OneHotEncoder.
>>>
>>> Then I am concatenating that one-hot encoded column (part #) to the
>>> text/document
>>> feature column (complaint), which I had applied the CountVectorizer and
>>> TfidfTransformer transformations.
>>>
>>> Then I chose the MultinomialNB model to fit my concatenated training
>>> data with.
>>>
>>> The problem I run into is when I invoke the prediction, I get a
>>> dimension mis-match error.
>>>
>>> Here's my jupyter notebook gist:
>>> http://nbviewer.jupyter.org/gist/anonymous/59ba930a783571c85
>>> ef86ba41424b311
>>>
>>> I would gladly appreciate it if someone can guide me where I went
>>> wrong.  Thanks!
>>>
>>> - Daniel
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/9506c147/attachment.html>

From abhishekraj10 at yahoo.com  Thu Aug  3 06:15:50 2017
From: abhishekraj10 at yahoo.com (Abhishek Raj)
Date: Thu, 3 Aug 2017 15:45:50 +0530
Subject: [scikit-learn] OneClassSvm | Different results on different runs
Message-ID: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>

Hi,

I am using one class svm for developing an anomaly detection model. I
observed that different runs of training on the same data set outputs
different accuracy. One run takes the accuracy as high as 98% and another
run on the same data brings it down to 93%. Googling a little bit I found
out that this is happening because of the random_state
<http://scikit-learn.org/stable/modules/generated/sklearn.utils.check_random_state.html>
parameter
but I am not clear of the details.

Can anyone expand on how is the parameter exactly affecting my training and
how I can figure out the best value to get the model with best accuracy?

Thanks,
Abhishek
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/cd094e3f/attachment.html>

From jaquesgrobler at gmail.com  Thu Aug  3 06:39:44 2017
From: jaquesgrobler at gmail.com (Jaques Grobler)
Date: Thu, 3 Aug 2017 12:39:44 +0200
Subject: [scikit-learn] OneClassSvm | Different results on different runs
In-Reply-To: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>
References: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>
Message-ID: <CAHcSOR=st2v7fj=GMX+Jg7yBNaoEOBasNLPDE8Av-7ziYWJEYQ@mail.gmail.com>

Hi,

The random_state parameter is used to generate a pseudo random number that
is used when shuffling your data for probability estimation

The seed of the pseudo random number generator to use when shuffling the
data for probability estimation.
A seed can be provided to control the shuffling for reproducible behavior.

Also, from the SVM docs
<http://scikit-learn.org/stable/modules/svm.html#svm-outlier-detection>

The underlying LinearSVC
> <http://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVC.html#sklearn.svm.LinearSVC> implementation
> uses a random number generator to select features when fitting the model.
> It is thus not uncommon, to have slightly different results for the same
> input data. If that happens, try with a smaller *tol *parameter.


Hope that helps

2017-08-03 12:15 GMT+02:00 Abhishek Raj via scikit-learn <
scikit-learn at python.org>:

> Hi,
>
> I am using one class svm for developing an anomaly detection model. I
> observed that different runs of training on the same data set outputs
> different accuracy. One run takes the accuracy as high as 98% and another
> run on the same data brings it down to 93%. Googling a little bit I found
> out that this is happening because of the random_state
> <http://scikit-learn.org/stable/modules/generated/sklearn.utils.check_random_state.html> parameter
> but I am not clear of the details.
>
> Can anyone expand on how is the parameter exactly affecting my training
> and how I can figure out the best value to get the model with best accuracy?
>
> Thanks,
> Abhishek
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/137fb6d1/attachment-0001.html>

From albertthomas88 at gmail.com  Thu Aug  3 07:26:17 2017
From: albertthomas88 at gmail.com (Albert Thomas)
Date: Thu, 03 Aug 2017 11:26:17 +0000
Subject: [scikit-learn] OneClassSvm | Different results on different runs
In-Reply-To: <CAHcSOR=st2v7fj=GMX+Jg7yBNaoEOBasNLPDE8Av-7ziYWJEYQ@mail.gmail.com>
References: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>
 <CAHcSOR=st2v7fj=GMX+Jg7yBNaoEOBasNLPDE8Av-7ziYWJEYQ@mail.gmail.com>
Message-ID: <CAK6amUPF_ZPPXzjZQJtZ5+_KuaWNETpbNVAGiv1+YsDThzRi1Q@mail.gmail.com>

Hi Abhishek,

Could you provide a small code snippet? I don't think the random_state
parameter should influence the result of the OneClassSVM as there is no
probability estimation for this estimator.

Albert

On Thu, Aug 3, 2017 at 12:41 PM Jaques Grobler <jaquesgrobler at gmail.com>
wrote:

> Hi,
>
> The random_state parameter is used to generate a pseudo random number that
> is used when shuffling your data for probability estimation
>
> The seed of the pseudo random number generator to use when shuffling the
> data for probability estimation.
> A seed can be provided to control the shuffling for reproducible behavior.
>
> Also, from the SVM docs
> <http://scikit-learn.org/stable/modules/svm.html#svm-outlier-detection>
>
> The underlying LinearSVC
>> <http://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVC.html#sklearn.svm.LinearSVC> implementation
>> uses a random number generator to select features when fitting the model.
>> It is thus not uncommon, to have slightly different results for the same
>> input data. If that happens, try with a smaller *tol *parameter.
>
>
> Hope that helps
>
> 2017-08-03 12:15 GMT+02:00 Abhishek Raj via scikit-learn <
> scikit-learn at python.org>:
>
>> Hi,
>>
>> I am using one class svm for developing an anomaly detection model. I
>> observed that different runs of training on the same data set outputs
>> different accuracy. One run takes the accuracy as high as 98% and another
>> run on the same data brings it down to 93%. Googling a little bit I found
>> out that this is happening because of the random_state
>> <http://scikit-learn.org/stable/modules/generated/sklearn.utils.check_random_state.html> parameter
>> but I am not clear of the details.
>>
>> Can anyone expand on how is the parameter exactly affecting my training
>> and how I can figure out the best value to get the model with best accuracy?
>>
>> Thanks,
>> Abhishek
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/7975490f/attachment.html>

From goix.nicolas at gmail.com  Thu Aug  3 07:54:37 2017
From: goix.nicolas at gmail.com (Nicolas Goix)
Date: Thu, 3 Aug 2017 13:54:37 +0200
Subject: [scikit-learn] OneClassSvm | Different results on different runs
In-Reply-To: <CAK6amUPF_ZPPXzjZQJtZ5+_KuaWNETpbNVAGiv1+YsDThzRi1Q@mail.gmail.com>
References: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>
 <CAHcSOR=st2v7fj=GMX+Jg7yBNaoEOBasNLPDE8Av-7ziYWJEYQ@mail.gmail.com>
 <CAK6amUPF_ZPPXzjZQJtZ5+_KuaWNETpbNVAGiv1+YsDThzRi1Q@mail.gmail.com>
Message-ID: <CAPV6P2yKhucq=zHfvTk_8kc0K1OzAORjcnQa3b2-Ky-h=p3xzA@mail.gmail.com>

@albertcthomas isn't there some randomness in SMO which could influence the
result if the tolerance parameter is too large?

On Aug 3, 2017 1:28 PM, "Albert Thomas" <albertthomas88 at gmail.com> wrote:

> Hi Abhishek,
>
> Could you provide a small code snippet? I don't think the random_state
> parameter should influence the result of the OneClassSVM as there is no
> probability estimation for this estimator.
>
> Albert
>
> On Thu, Aug 3, 2017 at 12:41 PM Jaques Grobler <jaquesgrobler at gmail.com>
> wrote:
>
>> Hi,
>>
>> The random_state parameter is used to generate a pseudo random number
>> that is used when shuffling your data for probability estimation
>>
>> The seed of the pseudo random number generator to use when shuffling the
>> data for probability estimation.
>> A seed can be provided to control the shuffling for reproducible behavior.
>>
>> Also, from the SVM docs
>> <http://scikit-learn.org/stable/modules/svm.html#svm-outlier-detection>
>>
>> The underlying LinearSVC
>>> <http://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVC.html#sklearn.svm.LinearSVC>
>>>  implementation uses a random number generator to select features when
>>> fitting the model. It is thus not uncommon, to have slightly different
>>> results for the same input data. If that happens, try with a smaller *tol
>>> *parameter.
>>
>>
>> Hope that helps
>>
>> 2017-08-03 12:15 GMT+02:00 Abhishek Raj via scikit-learn <
>> scikit-learn at python.org>:
>>
>>> Hi,
>>>
>>> I am using one class svm for developing an anomaly detection model. I
>>> observed that different runs of training on the same data set outputs
>>> different accuracy. One run takes the accuracy as high as 98% and another
>>> run on the same data brings it down to 93%. Googling a little bit I found
>>> out that this is happening because of the random_state
>>> <http://scikit-learn.org/stable/modules/generated/sklearn.utils.check_random_state.html> parameter
>>> but I am not clear of the details.
>>>
>>> Can anyone expand on how is the parameter exactly affecting my training
>>> and how I can figure out the best value to get the model with best accuracy?
>>>
>>> Thanks,
>>> Abhishek
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/58b9beca/attachment.html>

From albertthomas88 at gmail.com  Thu Aug  3 09:17:38 2017
From: albertthomas88 at gmail.com (Albert Thomas)
Date: Thu, 03 Aug 2017 13:17:38 +0000
Subject: [scikit-learn] OneClassSvm | Different results on different runs
In-Reply-To: <CAPV6P2yKhucq=zHfvTk_8kc0K1OzAORjcnQa3b2-Ky-h=p3xzA@mail.gmail.com>
References: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>
 <CAHcSOR=st2v7fj=GMX+Jg7yBNaoEOBasNLPDE8Av-7ziYWJEYQ@mail.gmail.com>
 <CAK6amUPF_ZPPXzjZQJtZ5+_KuaWNETpbNVAGiv1+YsDThzRi1Q@mail.gmail.com>
 <CAPV6P2yKhucq=zHfvTk_8kc0K1OzAORjcnQa3b2-Ky-h=p3xzA@mail.gmail.com>
Message-ID: <CAK6amUO_aC3GTR=eEhy-HnYxRoAVnLXeEWb=vJzvOn1kh_yFcg@mail.gmail.com>

Yes, in fact, changing the random_state might have an influence on the
result. The docstring of the random_state parameter for the OneClassSVM
seems incorrect though...

Albert

On Thu, Aug 3, 2017 at 1:55 PM Nicolas Goix <goix.nicolas at gmail.com> wrote:

> @albertcthomas isn't there some randomness in SMO which could influence
> the result if the tolerance parameter is too large?
>
> On Aug 3, 2017 1:28 PM, "Albert Thomas" <albertthomas88 at gmail.com> wrote:
>
>> Hi Abhishek,
>>
>> Could you provide a small code snippet? I don't think the random_state
>> parameter should influence the result of the OneClassSVM as there is no
>> probability estimation for this estimator.
>>
>> Albert
>>
>> On Thu, Aug 3, 2017 at 12:41 PM Jaques Grobler <jaquesgrobler at gmail.com>
>> wrote:
>>
>>> Hi,
>>>
>>> The random_state parameter is used to generate a pseudo random number
>>> that is used when shuffling your data for probability estimation
>>>
>>> The seed of the pseudo random number generator to use when shuffling the
>>> data for probability estimation.
>>> A seed can be provided to control the shuffling for reproducible
>>> behavior.
>>>
>>> Also, from the SVM docs
>>> <http://scikit-learn.org/stable/modules/svm.html#svm-outlier-detection>
>>>
>>> The underlying LinearSVC
>>>> <http://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVC.html#sklearn.svm.LinearSVC> implementation
>>>> uses a random number generator to select features when fitting the model.
>>>> It is thus not uncommon, to have slightly different results for the same
>>>> input data. If that happens, try with a smaller *tol *parameter.
>>>
>>>
>>> Hope that helps
>>>
>>> 2017-08-03 12:15 GMT+02:00 Abhishek Raj via scikit-learn <
>>> scikit-learn at python.org>:
>>>
>>>> Hi,
>>>>
>>>> I am using one class svm for developing an anomaly detection model. I
>>>> observed that different runs of training on the same data set outputs
>>>> different accuracy. One run takes the accuracy as high as 98% and another
>>>> run on the same data brings it down to 93%. Googling a little bit I found
>>>> out that this is happening because of the random_state
>>>> <http://scikit-learn.org/stable/modules/generated/sklearn.utils.check_random_state.html> parameter
>>>> but I am not clear of the details.
>>>>
>>>> Can anyone expand on how is the parameter exactly affecting my training
>>>> and how I can figure out the best value to get the model with best accuracy?
>>>>
>>>> Thanks,
>>>> Abhishek
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/4f2e6522/attachment-0001.html>

From m.waseem.ahmad at gmail.com  Thu Aug  3 10:37:02 2017
From: m.waseem.ahmad at gmail.com (muhammad waseem)
Date: Thu, 3 Aug 2017 15:37:02 +0100
Subject: [scikit-learn] Extra trees tuning parameters
Message-ID: <CAGWVJJW+uu2G=SBfSROGci1M4L+4Vei7E+a7ztYkadDyLjy21A@mail.gmail.com>

Hi All,
I was wondering if you could please tell me what is the "nmin , the minimum
sample size for splitting a node" (referred by Geurts et al., 2006) in
scikit-learn API for Extra trees? Is it min_samples_split in skearn?

Regards
Waseem
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/f49ae12d/attachment.html>

From tom.duprelatour at orange.fr  Thu Aug  3 11:18:17 2017
From: tom.duprelatour at orange.fr (Tom DLT)
Date: Thu, 3 Aug 2017 17:18:17 +0200
Subject: [scikit-learn] question about class_weights in
 LogisticRegression
In-Reply-To: <D5A62380.9E1A%jeremiah.johnson@unh.edu>
References: <D5A61E73.9E11%jeremiah.johnson@unh.edu>
 <CAAy-kdkgepButRwe32+SA6D-xAS2be5-HAvMM=Yuf_3vMfFehA@mail.gmail.com>
 <D5A62380.9E1A%jeremiah.johnson@unh.edu>
Message-ID: <CAGKmC=twRkxQWQ50B2Z1nB9bqZckH6TiYUX7VeZJjkdcAMM5aA@mail.gmail.com>

The class weights and sample weights are used in the same way, as a factor
specific to each sample, in the loss function.
In LogisticRegression, it is equivalent to incorporate this factor into a
regularization parameter C specific to each sample.

Tom

2017-08-01 18:30 GMT+02:00 Johnson, Jeremiah <Jeremiah.Johnson at unh.edu>:

> Right, I know how the class_weight calculation is performed. But then
> those class weights are utilized during the model fit process in some way
> in liblinear, and that?s what I am interested in. libSVM does
> class_weight[I] * C (https://www.csie.ntu.edu.tw/~cjlin/libsvm/); is the
> implementation in liblinear the same?
>
> Best,
> Jeremiah
>
>
>
> On 8/1/17, 12:19 PM, "scikit-learn on behalf of Stuart Reynolds"
> <scikit-learn-bounces+jeremiah.johnson=unh.edu at python.org on behalf of
> stuart at stuartreynolds.net> wrote:
>
> >I hope not. And not accoring to the docs...
> >https://urldefense.proofpoint.com/v2/url?u=https-
> 3A__github.com_scikit-2Dl
> >earn_scikit-2Dlearn_blob_ab93d65_sklearn_linear-
> 5Fmodel_logistic.py-23L947
> >&d=DwIGaQ&c=c6MrceVCY5m5A_KAUkrdoA&r=hQNTLb4Jonm4n54VBW80WEzIAaqvTO
> cTEjhIk
> >rRJWXo&m=2XR2z3VWvEaERt4miGabDte3xkz_FwzMKMwnvEOWj8o&s=
> 4uJZS3EaQgysmQlzjt-
> >yuLkSlcXTd5G50LkEFMcbZLQ&e=
> >
> >class_weight : dict or 'balanced', optional
> >Weights associated with classes in the form ``{class_label: weight}``.
> >If not given, all classes are supposed to have weight one.
> >The "balanced" mode uses the values of y to automatically adjust
> >weights inversely proportional to class frequencies in the input data
> >as ``n_samples / (n_classes * np.bincount(y))``.
> >Note that these weights will be multiplied with sample_weight (passed
> >through the fit method) if sample_weight is specified.
> >
> >On Tue, Aug 1, 2017 at 9:03 AM, Johnson, Jeremiah
> ><Jeremiah.Johnson at unh.edu> wrote:
> >> Hello all,
> >>
> >> I?m looking for confirmation on an implementation detail that is
> >>somewhere
> >> in liblinear, but I haven?t found documentation for yet. When the
> >> class_weights=?balanced? parameter is set in LogisticRegression, then
> >>the
> >> regularisation parameter for an observation from class I is
> >>class_weight[I]
> >> * C, where C is the usual regularization parameter ? is this correct?
> >>
> >> Thanks,
> >> Jeremiah
> >>
> >>
> >> _______________________________________________
> >> scikit-learn mailing list
> >> scikit-learn at python.org
> >>
> >>https://urldefense.proofpoint.com/v2/url?u=https-
> 3A__mail.python.org_mail
> >>man_listinfo_scikit-2Dlearn&d=DwIGaQ&c=c6MrceVCY5m5A_
> KAUkrdoA&r=hQNTLb4Jo
> >>nm4n54VBW80WEzIAaqvTOcTEjhIkrRJWXo&m=2XR2z3VWvEaERt4miGabDte3xkz_
> FwzMKMwn
> >>vEOWj8o&s=MgZoI9VOHFh3omGKHTASFx3NAVjj6AY3j_75mnOUg04&e=
> >>
> >_______________________________________________
> >scikit-learn mailing list
> >scikit-learn at python.org
> >https://urldefense.proofpoint.com/v2/url?u=https-
> 3A__mail.python.org_mailm
> >an_listinfo_scikit-2Dlearn&d=DwIGaQ&c=c6MrceVCY5m5A_
> KAUkrdoA&r=hQNTLb4Jonm
> >4n54VBW80WEzIAaqvTOcTEjhIkrRJWXo&m=2XR2z3VWvEaERt4miGabDte3xkz_
> FwzMKMwnvEO
> >Wj8o&s=MgZoI9VOHFh3omGKHTASFx3NAVjj6AY3j_75mnOUg04&e=
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/d0e0f6b3/attachment.html>

From t3kcit at gmail.com  Thu Aug  3 12:12:12 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Thu, 3 Aug 2017 12:12:12 -0400
Subject: [scikit-learn] OneClassSvm | Different results on different runs
In-Reply-To: <CAK6amUO_aC3GTR=eEhy-HnYxRoAVnLXeEWb=vJzvOn1kh_yFcg@mail.gmail.com>
References: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>
 <CAHcSOR=st2v7fj=GMX+Jg7yBNaoEOBasNLPDE8Av-7ziYWJEYQ@mail.gmail.com>
 <CAK6amUPF_ZPPXzjZQJtZ5+_KuaWNETpbNVAGiv1+YsDThzRi1Q@mail.gmail.com>
 <CAPV6P2yKhucq=zHfvTk_8kc0K1OzAORjcnQa3b2-Ky-h=p3xzA@mail.gmail.com>
 <CAK6amUO_aC3GTR=eEhy-HnYxRoAVnLXeEWb=vJzvOn1kh_yFcg@mail.gmail.com>
Message-ID: <ca580a2d-6cf5-a29a-c0f3-77ae1a832a54@gmail.com>



On 08/03/2017 09:17 AM, Albert Thomas wrote:
> Yes, in fact, changing the random_state might have an influence on the 
> result. The docstring of the random_state parameter for the 
> OneClassSVM seems incorrect though...
PR or issue welcome.

From t3kcit at gmail.com  Thu Aug  3 13:35:46 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Thu, 3 Aug 2017 13:35:46 -0400
Subject: [scikit-learn] Problems with running GridSearchCV on a pipeline
 with a custom transformer
In-Reply-To: <CACWGMfouHpma7Yb-kP=CmQhiH7oPkT+sPnY4_Wo=wpbveRB5eg@mail.gmail.com>
References: <CACWGMfo_O3ySYr8vPZ8CpUeZ1XO=Ykq+XkjpOq4OxrdBgEzDbw@mail.gmail.com>
 <e36c375d-97d8-56b0-ddb2-442bfc1f1128@gmail.com>
 <CACWGMfqv=HZxnxYao+tC5L9gc_a6OZe_hZOLNuEO37ac-7xfyA@mail.gmail.com>
 <a940dc7a-05c2-7b02-14a0-1c22fc4df14f@gmail.com>
 <CACWGMfqNt7uKoGApCwP+gm+Q9LdcCwggH56p5=EypGAO5KJgQQ@mail.gmail.com>
 <CACWGMfouHpma7Yb-kP=CmQhiH7oPkT+sPnY4_Wo=wpbveRB5eg@mail.gmail.com>
Message-ID: <bda5a454-b7f1-c71b-7055-96901a1e391f@gmail.com>

Hi Sam.
You need to put these into a reachable namespace (possibly as private 
functions) so that they can be pickled.
Please stay on the sklearn mailing list, I might not have time to reply.

Andy

On 08/03/2017 01:24 PM, Sam Barnett wrote:
> Hi Andy,
>
> I've since tried a different solution: instead of a pipeline, I've 
> simply created a classifier that is for the most part like svm.SVC, 
> though it takes a few extra inputs for the sequentialisation step. 
> I've used a Python function that can compute the Gram matrix between 
> two datasets of any shape to pass into SVC(), though I'm now having 
> trouble with pickling on the check_estimator test. It appears that 
> SeqSVC.fit() doesn't like to have methods defined within it. Can you 
> see how to pass this test? (the .ipynb file shows the error).
>
> Best,
> Sam
>
> On Wed, Aug 2, 2017 at 9:44 PM, Sam Barnett <sambarnett95 at gmail.com 
> <mailto:sambarnett95 at gmail.com>> wrote:
>
>     You're right: it does fail without GridSearchCV when I change the
>     size of seq_test. I will look at the transform tomorrow to see if
>     I can work this out. Thank you for your help so far!
>
>     On Wed, Aug 2, 2017 at 9:20 PM, Andreas Mueller <t3kcit at gmail.com
>     <mailto:t3kcit at gmail.com>> wrote:
>
>         Change the size of seq_test in your notebook and you'll see
>         the failure without GridSearchCV.
>         I haven't looked at your code in detail, but transform is
>         supposed to work on arbitrary new data with the same number of
>         features.
>         Your code requires the test data to have the same shape as the
>         training data.
>         Cross-validation will lead to training data and test data
>         having different sizes. But I feel like something is already
>         wrong if your
>         test data size depends on your training data size.
>
>
>
>         On 08/02/2017 03:08 PM, Sam Barnett wrote:
>>         Hi Andy,
>>
>>         The purpose of the transformer is to take an ordinary kernel
>>         (in this case I have taken 'rbf' as a default) and return a
>>         'sequentialised' kernel using a few extra parameters. Hence,
>>         the transformer takes an ordinary data-target pair X, y as
>>         its input, and the fit_transform(X, y) method will output the
>>         Gram matrix for X that is associated with this sequentialised
>>         kernel. In the pipeline, this Gram matrix is passed into an
>>         SVC classifier with the kernel parameter set to 'precomputed'.
>>
>>         Therefore, I do not think your hacky solution would be
>>         possible. However, I am still unsure how to implement your
>>         first solution: won't the Gram matrix from the transformer
>>         contain all the necessary kernel values? Could you elaborate
>>         further?
>>
>>
>>         Best,
>>         Sam
>>
>>         On Wed, Aug 2, 2017 at 5:05 PM, Andreas Mueller
>>         <t3kcit at gmail.com <mailto:t3kcit at gmail.com>> wrote:
>>
>>             Hi Sam.
>>             GridSearchCV will do cross-validation, which requires to
>>             "transform" the test data.
>>             The shape of the test-data will be different from the
>>             shape of the training data.
>>             You need to have the ability to compute the kernel
>>             between the training data and new test data.
>>
>>             A more hacky solution would be to compute the full kernel
>>             matrix in advance and pass that to GridSearchCV.
>>
>>             You probably don't need it here, but you should also
>>             checkout what the _pairwise attribute does in
>>             cross-validation,
>>             because that it likely to come up when playing with kernels.
>>
>>             Hth,
>>             Andy
>>
>>
>>             On 08/02/2017 08:38 AM, Sam Barnett wrote:
>>>             Dear all,
>>>
>>>             I have created a 2-step pipeline with a custom
>>>             transformer followed by a simple SVC classifier, and I
>>>             wish to run a grid-search over it. I am able to
>>>             successfully create the transformer and the pipeline,
>>>             and each of these elements work fine. However, when I
>>>             try to use the fit() method on my GridSearchCV object, I
>>>             get the following error:
>>>
>>>             57 # during fit.
>>>                  58 if X.shape != self.input_shape_:
>>>             ---> 59             raise ValueError('Shape of input is
>>>             different from what was seen '
>>>                  60                              'in `fit`')
>>>                  61
>>>
>>>             ValueError: Shape of input is different from what was
>>>             seen in `fit`
>>>
>>>             For a full breakdown of the problem, I have written a
>>>             Jupyter notebook showing exactly how the error occurs
>>>             (this also contains all .py files necessary to run the
>>>             notebook). Can anybody see how to work through this?
>>>
>>>             Many thanks,
>>>             Sam Barnett
>>>
>>>
>>>
>>>             _______________________________________________
>>>             scikit-learn mailing list
>>>             scikit-learn at python.org <mailto:scikit-learn at python.org>
>>>             https://mail.python.org/mailman/listinfo/scikit-learn
>>>             <https://mail.python.org/mailman/listinfo/scikit-learn>
>>
>>
>>             _______________________________________________
>>             scikit-learn mailing list
>>             scikit-learn at python.org <mailto:scikit-learn at python.org>
>>             https://mail.python.org/mailman/listinfo/scikit-learn
>>             <https://mail.python.org/mailman/listinfo/scikit-learn>
>>
>>
>>
>>
>>         _______________________________________________
>>         scikit-learn mailing list
>>         scikit-learn at python.org <mailto:scikit-learn at python.org>
>>         https://mail.python.org/mailman/listinfo/scikit-learn
>>         <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/a38e5b90/attachment-0001.html>

From pybokeh at gmail.com  Thu Aug  3 17:48:26 2017
From: pybokeh at gmail.com (pybokeh)
Date: Thu, 3 Aug 2017 17:48:26 -0400
Subject: [scikit-learn] Help With Text Classification
In-Reply-To: <CAAkaFLUuKF7ek2+wb2+EpBbxCmaUnENC=tJx87fU2vQDivqZ9A@mail.gmail.com>
References: <CAOoMBViVEfV61SJaUq4Z+z6sQegQc8w=Hf_4vb-X08Es1Jr1cA@mail.gmail.com>
 <CAAkaFLUuKF7ek2+wb2+EpBbxCmaUnENC=tJx87fU2vQDivqZ9A@mail.gmail.com>
Message-ID: <CAOoMBVjtk025aafoMf03BiFYqZcnzPgwWbDSeR+pJ-ZeNmiE1w@mail.gmail.com>

I found my problem.  When I one-hot encoded my test part #, it resulted in
being a 1x1 matrix, when I need it to be a 1x153.  This happened because I
used the default setting ('auto') for n_values, when I needed it set it to
153.  Now when I horizontally stacked it to my other feature matrix, the
resulting total # of columns now correctly comes to 1294, instead of
1142.  Looking back now, not sure if using Pipeline or using FeatureUnion
would have helped in this case or prevented this since this error occurred
on the prediction side, not on training or modeling side.

On Wed, Aug 2, 2017 at 10:38 PM, Joel Nothman <joel.nothman at gmail.com>
wrote:

> Use a Pipeline to help avoid this kind of issue (and others). You might
> also want to do something like http://scikit-learn.org/
> stable/auto_examples/hetero_feature_union.html
>
> On 3 August 2017 at 12:01, pybokeh <pybokeh at gmail.com> wrote:
>
>> Hello,
>> I am studying this example from scikit-learn's site:
>> http://scikit-learn.org/stable/tutorial/text_analytics/worki
>> ng_with_text_data.html
>>
>> The problem that I need to solve is very similar to this example, except
>> I have one
>> additional feature column (part #) that is categorical of type string.
>> My label or target
>> values consist of just 2 values: 0 or 1.
>>
>> With that additional feature column, I am transforming it with a
>> LabelEncoder and
>> then I am encoding it with the OneHotEncoder.
>>
>> Then I am concatenating that one-hot encoded column (part #) to the
>> text/document
>> feature column (complaint), which I had applied the CountVectorizer and
>> TfidfTransformer transformations.
>>
>> Then I chose the MultinomialNB model to fit my concatenated training data
>> with.
>>
>> The problem I run into is when I invoke the prediction, I get a dimension
>> mis-match error.
>>
>> Here's my jupyter notebook gist:
>> http://nbviewer.jupyter.org/gist/anonymous/59ba930a783571c85
>> ef86ba41424b311
>>
>> I would gladly appreciate it if someone can guide me where I went wrong.
>> Thanks!
>>
>> - Daniel
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170803/ebb72335/attachment.html>

From joel.nothman at gmail.com  Thu Aug  3 18:29:10 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Fri, 4 Aug 2017 08:29:10 +1000
Subject: [scikit-learn] Help With Text Classification
In-Reply-To: <CAOoMBVjtk025aafoMf03BiFYqZcnzPgwWbDSeR+pJ-ZeNmiE1w@mail.gmail.com>
References: <CAOoMBViVEfV61SJaUq4Z+z6sQegQc8w=Hf_4vb-X08Es1Jr1cA@mail.gmail.com>
 <CAAkaFLUuKF7ek2+wb2+EpBbxCmaUnENC=tJx87fU2vQDivqZ9A@mail.gmail.com>
 <CAOoMBVjtk025aafoMf03BiFYqZcnzPgwWbDSeR+pJ-ZeNmiE1w@mail.gmail.com>
Message-ID: <CAAkaFLV+Ydchw8hCgwN_LtRF0krgEHY0DNHTOdkj7aFOQMV3rA@mail.gmail.com>

pipeline helps in prediction time too.

On 4 Aug 2017 7:49 am, "pybokeh" <pybokeh at gmail.com> wrote:

> I found my problem.  When I one-hot encoded my test part #, it resulted in
> being a 1x1 matrix, when I need it to be a 1x153.  This happened because I
> used the default setting ('auto') for n_values, when I needed it set it to
> 153.  Now when I horizontally stacked it to my other feature matrix, the
> resulting total # of columns now correctly comes to 1294, instead of
> 1142.  Looking back now, not sure if using Pipeline or using FeatureUnion
> would have helped in this case or prevented this since this error occurred
> on the prediction side, not on training or modeling side.
>
> On Wed, Aug 2, 2017 at 10:38 PM, Joel Nothman <joel.nothman at gmail.com>
> wrote:
>
>> Use a Pipeline to help avoid this kind of issue (and others). You might
>> also want to do something like http://scikit-learn.org/stable
>> /auto_examples/hetero_feature_union.html
>>
>> On 3 August 2017 at 12:01, pybokeh <pybokeh at gmail.com> wrote:
>>
>>> Hello,
>>> I am studying this example from scikit-learn's site:
>>> http://scikit-learn.org/stable/tutorial/text_analytics/worki
>>> ng_with_text_data.html
>>>
>>> The problem that I need to solve is very similar to this example, except
>>> I have one
>>> additional feature column (part #) that is categorical of type string.
>>> My label or target
>>> values consist of just 2 values: 0 or 1.
>>>
>>> With that additional feature column, I am transforming it with a
>>> LabelEncoder and
>>> then I am encoding it with the OneHotEncoder.
>>>
>>> Then I am concatenating that one-hot encoded column (part #) to the
>>> text/document
>>> feature column (complaint), which I had applied the CountVectorizer and
>>> TfidfTransformer transformations.
>>>
>>> Then I chose the MultinomialNB model to fit my concatenated training
>>> data with.
>>>
>>> The problem I run into is when I invoke the prediction, I get a
>>> dimension mis-match error.
>>>
>>> Here's my jupyter notebook gist:
>>> http://nbviewer.jupyter.org/gist/anonymous/59ba930a783571c85
>>> ef86ba41424b311
>>>
>>> I would gladly appreciate it if someone can guide me where I went
>>> wrong.  Thanks!
>>>
>>> - Daniel
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/c3565b77/attachment.html>

From sambarnett95 at gmail.com  Fri Aug  4 06:29:50 2017
From: sambarnett95 at gmail.com (Sam Barnett)
Date: Fri, 4 Aug 2017 11:29:50 +0100
Subject: [scikit-learn] Problems with running GridSearchCV on a pipeline
 with a custom transformer
Message-ID: <CACWGMfr3mb+7YoZe+oE7_dijrNBrBE555epRsH1=E=n-iSvZVQ@mail.gmail.com>

Hi Andy,
I have since been able to resolve the pickling issue, though I am now
getting an error message saying that an error message does not include the
expected string 'fit'. In general, I am trying to use the fit() method of
my classifier to instantiate a separate SVC() classifier with a custom
kernel, fit THAT to the data, then return this instance as the fitted
version of the new classifier. Is this possible in theory? If so, what is
the best way to implement it?

As before, the requisite code and a .ipynb file is attached.

Best,
Sam

On Thu, Aug 3, 2017 at 6:35 PM, Andreas Mueller <t3kcit at gmail.com> wrote:

> Hi Sam.
> You need to put these into a reachable namespace (possibly as private
> functions) so that they can be pickled.
> Please stay on the sklearn mailing list, I might not have time to reply.
>
> Andy
>
>
> On 08/03/2017 01:24 PM, Sam Barnett wrote:
>
> Hi Andy,
>
> I've since tried a different solution: instead of a pipeline, I've simply
> created a classifier that is for the most part like svm.SVC, though it
> takes a few extra inputs for the sequentialisation step. I've used a Python
> function that can compute the Gram matrix between two datasets of any shape
> to pass into SVC(), though I'm now having trouble with pickling on the
> check_estimator test. It appears that SeqSVC.fit() doesn't like to have
> methods defined within it. Can you see how to pass this test? (the .ipynb
> file shows the error).
>
> Best,
> Sam
>
> On Wed, Aug 2, 2017 at 9:44 PM, Sam Barnett <sambarnett95 at gmail.com>
> wrote:
>
>> You're right: it does fail without GridSearchCV when I change the size of
>> seq_test. I will look at the transform tomorrow to see if I can work this
>> out. Thank you for your help so far!
>>
>> On Wed, Aug 2, 2017 at 9:20 PM, Andreas Mueller <t3kcit at gmail.com> wrote:
>>
>>> Change the size of seq_test in your notebook and you'll see the failure
>>> without GridSearchCV.
>>> I haven't looked at your code in detail, but transform is supposed to
>>> work on arbitrary new data with the same number of features.
>>> Your code requires the test data to have the same shape as the training
>>> data.
>>> Cross-validation will lead to training data and test data having
>>> different sizes. But I feel like something is already wrong if your
>>> test data size depends on your training data size.
>>>
>>>
>>>
>>> On 08/02/2017 03:08 PM, Sam Barnett wrote:
>>>
>>> Hi Andy,
>>>
>>> The purpose of the transformer is to take an ordinary kernel (in this
>>> case I have taken 'rbf' as a default) and return a 'sequentialised' kernel
>>> using a few extra parameters. Hence, the transformer takes an ordinary
>>> data-target pair X, y as its input, and the fit_transform(X, y) method will
>>> output the Gram matrix for X that is associated with this sequentialised
>>> kernel. In the pipeline, this Gram matrix is passed into an SVC classifier
>>> with the kernel parameter set to 'precomputed'.
>>>
>>> Therefore, I do not think your hacky solution would be possible.
>>> However, I am still unsure how to implement your first solution: won't the
>>> Gram matrix from the transformer contain all the necessary kernel values?
>>> Could you elaborate further?
>>>
>>>
>>> Best,
>>> Sam
>>>
>>> On Wed, Aug 2, 2017 at 5:05 PM, Andreas Mueller <t3kcit at gmail.com>
>>> wrote:
>>>
>>>> Hi Sam.
>>>> GridSearchCV will do cross-validation, which requires to "transform"
>>>> the test data.
>>>> The shape of the test-data will be different from the shape of the
>>>> training data.
>>>> You need to have the ability to compute the kernel between the training
>>>> data and new test data.
>>>>
>>>> A more hacky solution would be to compute the full kernel matrix in
>>>> advance and pass that to GridSearchCV.
>>>>
>>>> You probably don't need it here, but you should also checkout what the
>>>> _pairwise attribute does in cross-validation,
>>>> because that it likely to come up when playing with kernels.
>>>>
>>>> Hth,
>>>> Andy
>>>>
>>>>
>>>> On 08/02/2017 08:38 AM, Sam Barnett wrote:
>>>>
>>>> Dear all,
>>>>
>>>> I have created a 2-step pipeline with a custom transformer followed by
>>>> a simple SVC classifier, and I wish to run a grid-search over it. I am able
>>>> to successfully create the transformer and the pipeline, and each of these
>>>> elements work fine. However, when I try to use the fit() method on my
>>>> GridSearchCV object, I get the following error:
>>>>
>>>>      57         # during fit.
>>>>      58         if X.shape != self.input_shape_:
>>>> ---> 59             raise ValueError('Shape of input is different from
>>>> what was seen '
>>>>      60                              'in `fit`')
>>>>      61
>>>>
>>>> ValueError: Shape of input is different from what was seen in `fit`
>>>>
>>>> For a full breakdown of the problem, I have written a Jupyter notebook
>>>> showing exactly how the error occurs (this also contains all .py files
>>>> necessary to run the notebook). Can anybody see how to work through this?
>>>>
>>>> Many thanks,
>>>> Sam Barnett
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/0d150d4c/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: seqsvc.py
Type: text/x-python-script
Size: 3051 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/0d150d4c/attachment-0002.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Sequential Kernel SVC GridSearchCV Test.ipynb
Type: application/octet-stream
Size: 7678 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/0d150d4c/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SeqKernelLucy.py
Type: text/x-python-script
Size: 2628 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/0d150d4c/attachment-0003.bin>

From albertthomas88 at gmail.com  Fri Aug  4 08:49:16 2017
From: albertthomas88 at gmail.com (Albert Thomas)
Date: Fri, 04 Aug 2017 12:49:16 +0000
Subject: [scikit-learn] OneClassSvm | Different results on different runs
In-Reply-To: <ca580a2d-6cf5-a29a-c0f3-77ae1a832a54@gmail.com>
References: <CANvDnHp95FdifURMGdCvJWns64rLUJEJ-TcbE5Uaa3LH7Uru_Q@mail.gmail.com>
 <CAHcSOR=st2v7fj=GMX+Jg7yBNaoEOBasNLPDE8Av-7ziYWJEYQ@mail.gmail.com>
 <CAK6amUPF_ZPPXzjZQJtZ5+_KuaWNETpbNVAGiv1+YsDThzRi1Q@mail.gmail.com>
 <CAPV6P2yKhucq=zHfvTk_8kc0K1OzAORjcnQa3b2-Ky-h=p3xzA@mail.gmail.com>
 <CAK6amUO_aC3GTR=eEhy-HnYxRoAVnLXeEWb=vJzvOn1kh_yFcg@mail.gmail.com>
 <ca580a2d-6cf5-a29a-c0f3-77ae1a832a54@gmail.com>
Message-ID: <CAK6amUM=rjcraUYCrmhq04f072OF2+bKajQOsvy3CST2Q7HbzA@mail.gmail.com>

I opened an issue https://github.com/scikit-learn/scikit-learn/issues/9497

Albert

On Thu, Aug 3, 2017 at 6:16 PM Andreas Mueller <t3kcit at gmail.com> wrote:

>
>
> On 08/03/2017 09:17 AM, Albert Thomas wrote:
> > Yes, in fact, changing the random_state might have an influence on the
> > result. The docstring of the random_state parameter for the
> > OneClassSVM seems incorrect though...
> PR or issue welcome.
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/68a84c13/attachment.html>

From olivier.grisel at ensta.org  Fri Aug  4 09:54:00 2017
From: olivier.grisel at ensta.org (Olivier Grisel)
Date: Fri, 4 Aug 2017 15:54:00 +0200
Subject: [scikit-learn] Extra trees tuning parameters
In-Reply-To: <CAGWVJJW+uu2G=SBfSROGci1M4L+4Vei7E+a7ztYkadDyLjy21A@mail.gmail.com>
References: <CAGWVJJW+uu2G=SBfSROGci1M4L+4Vei7E+a7ztYkadDyLjy21A@mail.gmail.com>
Message-ID: <CAFvE7K7L0fv4_iUm5xeVyvem_=UPSjtp6SKHdXU29MydiZFYTw@mail.gmail.com>

I believe so even though it's always better to check in the code to see how
this parameter is actually used.

-- 
Olivier
?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/2a472f6a/attachment.html>

From t3kcit at gmail.com  Fri Aug  4 10:50:40 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Fri, 4 Aug 2017 10:50:40 -0400
Subject: [scikit-learn] Problems with running GridSearchCV on a pipeline
 with a custom transformer
In-Reply-To: <CACWGMfr3mb+7YoZe+oE7_dijrNBrBE555epRsH1=E=n-iSvZVQ@mail.gmail.com>
References: <CACWGMfr3mb+7YoZe+oE7_dijrNBrBE555epRsH1=E=n-iSvZVQ@mail.gmail.com>
Message-ID: <90aed61d-2d9b-8d02-3c9b-fbe0ccdd98ee@gmail.com>

Yes, that's totally fine. The error is unrelated and just means you need 
to call ``check_is_fitted`` in your predict method
to give a nicer error message.


On 08/04/2017 06:29 AM, Sam Barnett wrote:
> Hi Andy,
> I have since been able to resolve the pickling issue, though I am now 
> getting an error message saying that an error message does not include 
> the expected string 'fit'. In general, I am trying to use the fit() 
> method of my classifier to instantiate a separate SVC() classifier 
> with a custom kernel, fit THAT to the data, then return this instance 
> as the fitted version of the new classifier. Is this possible in 
> theory? If so, what is the best way to implement it?
>
> As before, the requisite code and a .ipynb file is attached.
>
> Best,
> Sam
>
> On Thu, Aug 3, 2017 at 6:35 PM, Andreas Mueller <t3kcit at gmail.com 
> <mailto:t3kcit at gmail.com>> wrote:
>
>     Hi Sam.
>     You need to put these into a reachable namespace (possibly as
>     private functions) so that they can be pickled.
>     Please stay on the sklearn mailing list, I might not have time to
>     reply.
>
>     Andy
>
>
>     On 08/03/2017 01:24 PM, Sam Barnett wrote:
>>     Hi Andy,
>>
>>     I've since tried a different solution: instead of a pipeline,
>>     I've simply created a classifier that is for the most part like
>>     svm.SVC, though it takes a few extra inputs for the
>>     sequentialisation step. I've used a Python function that can
>>     compute the Gram matrix between two datasets of any shape to pass
>>     into SVC(), though I'm now having trouble with pickling on the
>>     check_estimator test. It appears that SeqSVC.fit() doesn't like
>>     to have methods defined within it. Can you see how to pass this
>>     test? (the .ipynb file shows the error).
>>
>>     Best,
>>     Sam
>>
>>     On Wed, Aug 2, 2017 at 9:44 PM, Sam Barnett
>>     <sambarnett95 at gmail.com <mailto:sambarnett95 at gmail.com>> wrote:
>>
>>         You're right: it does fail without GridSearchCV when I change
>>         the size of seq_test. I will look at the transform tomorrow
>>         to see if I can work this out. Thank you for your help so far!
>>
>>         On Wed, Aug 2, 2017 at 9:20 PM, Andreas Mueller
>>         <t3kcit at gmail.com <mailto:t3kcit at gmail.com>> wrote:
>>
>>             Change the size of seq_test in your notebook and you'll
>>             see the failure without GridSearchCV.
>>             I haven't looked at your code in detail, but transform is
>>             supposed to work on arbitrary new data with the same
>>             number of features.
>>             Your code requires the test data to have the same shape
>>             as the training data.
>>             Cross-validation will lead to training data and test data
>>             having different sizes. But I feel like something is
>>             already wrong if your
>>             test data size depends on your training data size.
>>
>>
>>
>>             On 08/02/2017 03:08 PM, Sam Barnett wrote:
>>>             Hi Andy,
>>>
>>>             The purpose of the transformer is to take an ordinary
>>>             kernel (in this case I have taken 'rbf' as a default)
>>>             and return a 'sequentialised' kernel using a few extra
>>>             parameters. Hence, the transformer takes an ordinary
>>>             data-target pair X, y as its input, and the
>>>             fit_transform(X, y) method will output the Gram matrix
>>>             for X that is associated with this sequentialised
>>>             kernel. In the pipeline, this Gram matrix is passed into
>>>             an SVC classifier with the kernel parameter set to
>>>             'precomputed'.
>>>
>>>             Therefore, I do not think your hacky solution would be
>>>             possible. However, I am still unsure how to implement
>>>             your first solution: won't the Gram matrix from the
>>>             transformer contain all the necessary kernel values?
>>>             Could you elaborate further?
>>>
>>>
>>>             Best,
>>>             Sam
>>>
>>>             On Wed, Aug 2, 2017 at 5:05 PM, Andreas Mueller
>>>             <t3kcit at gmail.com <mailto:t3kcit at gmail.com>> wrote:
>>>
>>>                 Hi Sam.
>>>                 GridSearchCV will do cross-validation, which
>>>                 requires to "transform" the test data.
>>>                 The shape of the test-data will be different from
>>>                 the shape of the training data.
>>>                 You need to have the ability to compute the kernel
>>>                 between the training data and new test data.
>>>
>>>                 A more hacky solution would be to compute the full
>>>                 kernel matrix in advance and pass that to GridSearchCV.
>>>
>>>                 You probably don't need it here, but you should also
>>>                 checkout what the _pairwise attribute does in
>>>                 cross-validation,
>>>                 because that it likely to come up when playing with
>>>                 kernels.
>>>
>>>                 Hth,
>>>                 Andy
>>>
>>>
>>>                 On 08/02/2017 08:38 AM, Sam Barnett wrote:
>>>>                 Dear all,
>>>>
>>>>                 I have created a 2-step pipeline with a custom
>>>>                 transformer followed by a simple SVC classifier,
>>>>                 and I wish to run a grid-search over it. I am able
>>>>                 to successfully create the transformer and the
>>>>                 pipeline, and each of these elements work fine.
>>>>                 However, when I try to use the fit() method on my
>>>>                 GridSearchCV object, I get the following error:
>>>>
>>>>                 57 # during fit.
>>>>                      58 if X.shape != self.input_shape_:
>>>>                 ---> 59             raise ValueError('Shape of
>>>>                 input is different from what was seen '
>>>>                      60                              'in `fit`')
>>>>                      61
>>>>
>>>>                 ValueError: Shape of input is different from what
>>>>                 was seen in `fit`
>>>>
>>>>                 For a full breakdown of the problem, I have written
>>>>                 a Jupyter notebook showing exactly how the error
>>>>                 occurs (this also contains all .py files necessary
>>>>                 to run the notebook). Can anybody see how to work
>>>>                 through this?
>>>>
>>>>                 Many thanks,
>>>>                 Sam Barnett
>>>>
>>>>
>>>>
>>>>                 _______________________________________________
>>>>                 scikit-learn mailing list
>>>>                 scikit-learn at python.org
>>>>                 <mailto:scikit-learn at python.org>
>>>>                 https://mail.python.org/mailman/listinfo/scikit-learn
>>>>                 <https://mail.python.org/mailman/listinfo/scikit-learn>
>>>
>>>
>>>                 _______________________________________________
>>>                 scikit-learn mailing list
>>>                 scikit-learn at python.org <mailto:scikit-learn at python.org>
>>>                 https://mail.python.org/mailman/listinfo/scikit-learn
>>>                 <https://mail.python.org/mailman/listinfo/scikit-learn>
>>>
>>>
>>>
>>>
>>>             _______________________________________________
>>>             scikit-learn mailing list
>>>             scikit-learn at python.org <mailto:scikit-learn at python.org>
>>>             https://mail.python.org/mailman/listinfo/scikit-learn
>>>             <https://mail.python.org/mailman/listinfo/scikit-learn>
>>
>>
>>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170804/c5ffbc05/attachment-0001.html>

From georg.kf.heiler at gmail.com  Sat Aug  5 05:10:57 2017
From: georg.kf.heiler at gmail.com (Georg Heiler)
Date: Sat, 05 Aug 2017 09:10:57 +0000
Subject: [scikit-learn] transform categorical data to numerical
 representation
Message-ID: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>

Hi,

the LabelEncooder is only meant for a single column i.e. target variable.
Is the DictVectorizeer or a manual chaining of multiple LabelEncoders (one
per categorical column) the desired way to get values which can be fed into
a subsequent classifier?

Is there some way I have overlooked which works better and possibly also
can handle unseen values by applying most frequent imputation?

regards,
Georg
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170805/308522c4/attachment.html>

From se.raschka at gmail.com  Sat Aug  5 12:13:10 2017
From: se.raschka at gmail.com (Sebastian Raschka)
Date: Sat, 5 Aug 2017 12:13:10 -0400
Subject: [scikit-learn] transform categorical data to numerical
 representation
In-Reply-To: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>
References: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>
Message-ID: <6D0BF22C-9ABA-4C2A-B35B-210673439286@gmail.com>

Hi, Georg,

I bring this up every time here on the mailing list :), and you probably aware of this issue, but it makes a difference whether your categorical data is nominal or ordinal. For instance if you have an ordinal variable like with values like {small, medium, large} you probably want to encode it as {1, 2, 3} or {1, 20, 100} or whatever is appropriate based on your domain knowledge regarding the variable. If you have sth like {blue, red, green} it may make more sense to do a one-hot encoding so that the classifier doesn't assume  a relationship between the variables like blue > red > green or sth like that.

Now, the DictVectorizer and OneHotEncoder are both doing one hot encoding. The LabelEncoder does convert a variable to integer values, but if you have sth like {small, medium, large}, it wouldn't know the order (if that's an ordinal variable) and it would just assign arbitrary integers in increasing order. Thus, if you are dealing ordinal variables, there's no way around doing this manually; for example you could create mapping dictionaries for that (most conveniently done in pandas).

Best,
Sebastian

> On Aug 5, 2017, at 5:10 AM, Georg Heiler <georg.kf.heiler at gmail.com> wrote:
> 
> Hi,
> 
> the LabelEncooder is only meant for a single column i.e. target variable. Is the DictVectorizeer or a manual chaining of multiple LabelEncoders (one per categorical column) the desired way to get values which can be fed into a subsequent classifier?
> 
> Is there some way I have overlooked which works better and possibly also can handle unseen values by applying most frequent imputation?
> 
> regards,
> Georg
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn


From joel.nothman at gmail.com  Sat Aug  5 18:47:23 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Sun, 6 Aug 2017 08:47:23 +1000
Subject: [scikit-learn] transform categorical data to numerical
 representation
In-Reply-To: <6D0BF22C-9ABA-4C2A-B35B-210673439286@gmail.com>
References: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>
 <6D0BF22C-9ABA-4C2A-B35B-210673439286@gmail.com>
Message-ID: <CAAkaFLVr47Viuet6-YWe8qguA0c655cojK3t2i=z+DKevjp+eA@mail.gmail.com>

We are working on CategoricalEncoder in
https://github.com/scikit-learn/scikit-learn/pull/9151 to help users more
with this kind of thing. Feedback and testing is welcome.

On 6 August 2017 at 02:13, Sebastian Raschka <se.raschka at gmail.com> wrote:

> Hi, Georg,
>
> I bring this up every time here on the mailing list :), and you probably
> aware of this issue, but it makes a difference whether your categorical
> data is nominal or ordinal. For instance if you have an ordinal variable
> like with values like {small, medium, large} you probably want to encode it
> as {1, 2, 3} or {1, 20, 100} or whatever is appropriate based on your
> domain knowledge regarding the variable. If you have sth like {blue, red,
> green} it may make more sense to do a one-hot encoding so that the
> classifier doesn't assume  a relationship between the variables like blue >
> red > green or sth like that.
>
> Now, the DictVectorizer and OneHotEncoder are both doing one hot encoding.
> The LabelEncoder does convert a variable to integer values, but if you have
> sth like {small, medium, large}, it wouldn't know the order (if that's an
> ordinal variable) and it would just assign arbitrary integers in increasing
> order. Thus, if you are dealing ordinal variables, there's no way around
> doing this manually; for example you could create mapping dictionaries for
> that (most conveniently done in pandas).
>
> Best,
> Sebastian
>
> > On Aug 5, 2017, at 5:10 AM, Georg Heiler <georg.kf.heiler at gmail.com>
> wrote:
> >
> > Hi,
> >
> > the LabelEncooder is only meant for a single column i.e. target
> variable. Is the DictVectorizeer or a manual chaining of multiple
> LabelEncoders (one per categorical column) the desired way to get values
> which can be fed into a subsequent classifier?
> >
> > Is there some way I have overlooked which works better and possibly also
> can handle unseen values by applying most frequent imputation?
> >
> > regards,
> > Georg
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170806/f7a0a293/attachment.html>

From georg.kf.heiler at gmail.com  Sun Aug  6 06:30:28 2017
From: georg.kf.heiler at gmail.com (Georg Heiler)
Date: Sun, 06 Aug 2017 10:30:28 +0000
Subject: [scikit-learn] transform categorical data to numerical
 representation
In-Reply-To: <CAAkaFLVr47Viuet6-YWe8qguA0c655cojK3t2i=z+DKevjp+eA@mail.gmail.com>
References: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>
 <6D0BF22C-9ABA-4C2A-B35B-210673439286@gmail.com>
 <CAAkaFLVr47Viuet6-YWe8qguA0c655cojK3t2i=z+DKevjp+eA@mail.gmail.com>
Message-ID: <CAMhVi-RALPniqW+C5AMaMGVpYBEJpdG1V4W84Of+=u7JVZyeqQ@mail.gmail.com>

@sebastian: thanks. Indeed, I am aware of this problem.

I developed something here:
https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b0799dc2ce but
realized that the performance of prediction is pretty lame when there are
around 100-150 columns used as the input.
Do you have some ideas how to speed this up?

Regards,
Georg

Joel Nothman <joel.nothman at gmail.com> schrieb am So., 6. Aug. 2017 um
00:49 Uhr:

> We are working on CategoricalEncoder in
> https://github.com/scikit-learn/scikit-learn/pull/9151 to help users more
> with this kind of thing. Feedback and testing is welcome.
>
> On 6 August 2017 at 02:13, Sebastian Raschka <se.raschka at gmail.com> wrote:
>
>> Hi, Georg,
>>
>> I bring this up every time here on the mailing list :), and you probably
>> aware of this issue, but it makes a difference whether your categorical
>> data is nominal or ordinal. For instance if you have an ordinal variable
>> like with values like {small, medium, large} you probably want to encode it
>> as {1, 2, 3} or {1, 20, 100} or whatever is appropriate based on your
>> domain knowledge regarding the variable. If you have sth like {blue, red,
>> green} it may make more sense to do a one-hot encoding so that the
>> classifier doesn't assume  a relationship between the variables like blue >
>> red > green or sth like that.
>>
>> Now, the DictVectorizer and OneHotEncoder are both doing one hot
>> encoding. The LabelEncoder does convert a variable to integer values, but
>> if you have sth like {small, medium, large}, it wouldn't know the order (if
>> that's an ordinal variable) and it would just assign arbitrary integers in
>> increasing order. Thus, if you are dealing ordinal variables, there's no
>> way around doing this manually; for example you could create mapping
>> dictionaries for that (most conveniently done in pandas).
>>
>> Best,
>> Sebastian
>>
>> > On Aug 5, 2017, at 5:10 AM, Georg Heiler <georg.kf.heiler at gmail.com>
>> wrote:
>> >
>> > Hi,
>> >
>> > the LabelEncooder is only meant for a single column i.e. target
>> variable. Is the DictVectorizeer or a manual chaining of multiple
>> LabelEncoders (one per categorical column) the desired way to get values
>> which can be fed into a subsequent classifier?
>> >
>> > Is there some way I have overlooked which works better and possibly
>> also can handle unseen values by applying most frequent imputation?
>> >
>> > regards,
>> > Georg
>> > _______________________________________________
>> > scikit-learn mailing list
>> > scikit-learn at python.org
>> > https://mail.python.org/mailman/listinfo/scikit-learn
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170806/5b226178/attachment-0001.html>

From se.raschka at gmail.com  Sun Aug  6 14:37:15 2017
From: se.raschka at gmail.com (Sebastian Raschka)
Date: Sun, 6 Aug 2017 14:37:15 -0400
Subject: [scikit-learn] transform categorical data to numerical
 representation
In-Reply-To: <CAMhVi-RALPniqW+C5AMaMGVpYBEJpdG1V4W84Of+=u7JVZyeqQ@mail.gmail.com>
References: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>
 <6D0BF22C-9ABA-4C2A-B35B-210673439286@gmail.com>
 <CAAkaFLVr47Viuet6-YWe8qguA0c655cojK3t2i=z+DKevjp+eA@mail.gmail.com>
 <CAMhVi-RALPniqW+C5AMaMGVpYBEJpdG1V4W84Of+=u7JVZyeqQ@mail.gmail.com>
Message-ID: <103609E5-E50B-4993-87F2-11661E7C7EB5@gmail.com>

> performance of prediction is pretty lame when there are around 100-150 columns used as the input.

you are talking about computational performance when you are calling the "transform" method? Have you done some profiling to find out where your bottle neck (in the for loop) is? Just one a very quick look, I think this

data.loc[~data[column].isin(fittedLabels), column] = str(replacementForUnseen)

is already very slow because fittedLabels is an array where you have O(n) lookup instead of an average O(1) by using a hash table. Or is the isin function converting it to a hashtable/set/dict?

In general, would it maybe help to use pandas' factorize? https://pandas.pydata.org/pandas-docs/stable/generated/pandas.factorize.html
For predict time, say you have only 1 example for prediction that needs to be converted, you could append prototypes of all possible values that could occur, do the transformation, and then only pass the 1 transformed sample to the classifier. I guess that could be even slow though ... 

Best,
Sebastian

> On Aug 6, 2017, at 6:30 AM, Georg Heiler <georg.kf.heiler at gmail.com> wrote:
> 
> @sebastian: thanks. Indeed, I am aware of this problem.
> 
> I developed something here: https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b0799dc2ce but realized that the performance of prediction is pretty lame when there are around 100-150 columns used as the input.
> Do you have some ideas how to speed this up?
> 
> Regards,
> Georg
> 
> Joel Nothman <joel.nothman at gmail.com> schrieb am So., 6. Aug. 2017 um 00:49 Uhr:
> We are working on CategoricalEncoder in https://github.com/scikit-learn/scikit-learn/pull/9151 to help users more with this kind of thing. Feedback and testing is welcome.
> 
> On 6 August 2017 at 02:13, Sebastian Raschka <se.raschka at gmail.com> wrote:
> Hi, Georg,
> 
> I bring this up every time here on the mailing list :), and you probably aware of this issue, but it makes a difference whether your categorical data is nominal or ordinal. For instance if you have an ordinal variable like with values like {small, medium, large} you probably want to encode it as {1, 2, 3} or {1, 20, 100} or whatever is appropriate based on your domain knowledge regarding the variable. If you have sth like {blue, red, green} it may make more sense to do a one-hot encoding so that the classifier doesn't assume  a relationship between the variables like blue > red > green or sth like that.
> 
> Now, the DictVectorizer and OneHotEncoder are both doing one hot encoding. The LabelEncoder does convert a variable to integer values, but if you have sth like {small, medium, large}, it wouldn't know the order (if that's an ordinal variable) and it would just assign arbitrary integers in increasing order. Thus, if you are dealing ordinal variables, there's no way around doing this manually; for example you could create mapping dictionaries for that (most conveniently done in pandas).
> 
> Best,
> Sebastian
> 
> > On Aug 5, 2017, at 5:10 AM, Georg Heiler <georg.kf.heiler at gmail.com> wrote:
> >
> > Hi,
> >
> > the LabelEncooder is only meant for a single column i.e. target variable. Is the DictVectorizeer or a manual chaining of multiple LabelEncoders (one per categorical column) the desired way to get values which can be fed into a subsequent classifier?
> >
> > Is there some way I have overlooked which works better and possibly also can handle unseen values by applying most frequent imputation?
> >
> > regards,
> > Georg
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
> 
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
> 
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn


From georg.kf.heiler at gmail.com  Mon Aug  7 02:40:18 2017
From: georg.kf.heiler at gmail.com (Georg Heiler)
Date: Mon, 07 Aug 2017 06:40:18 +0000
Subject: [scikit-learn] transform categorical data to numerical
 representation
In-Reply-To: <103609E5-E50B-4993-87F2-11661E7C7EB5@gmail.com>
References: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>
 <6D0BF22C-9ABA-4C2A-B35B-210673439286@gmail.com>
 <CAAkaFLVr47Viuet6-YWe8qguA0c655cojK3t2i=z+DKevjp+eA@mail.gmail.com>
 <CAMhVi-RALPniqW+C5AMaMGVpYBEJpdG1V4W84Of+=u7JVZyeqQ@mail.gmail.com>
 <103609E5-E50B-4993-87F2-11661E7C7EB5@gmail.com>
Message-ID: <CAMhVi-Qs0ha2xhUFnLCTj11jPkK_dwKx6=9btK8UEVwF9xMN8w@mail.gmail.com>

I will need to look into factorize. Here is the result from profiling the
transform method on a single new observation
https://codereview.stackexchange.com/q/171622/132999


Best Georg
Sebastian Raschka <se.raschka at gmail.com> schrieb am So. 6. Aug. 2017 um
20:39:

> > performance of prediction is pretty lame when there are around 100-150
> columns used as the input.
>
> you are talking about computational performance when you are calling the
> "transform" method? Have you done some profiling to find out where your
> bottle neck (in the for loop) is? Just one a very quick look, I think this
>
> data.loc[~data[column].isin(fittedLabels), column] =
> str(replacementForUnseen)
>
> is already very slow because fittedLabels is an array where you have O(n)
> lookup instead of an average O(1) by using a hash table. Or is the isin
> function converting it to a hashtable/set/dict?
>
> In general, would it maybe help to use pandas' factorize?
> https://pandas.pydata.org/pandas-docs/stable/generated/pandas.factorize.html
> For predict time, say you have only 1 example for prediction that needs to
> be converted, you could append prototypes of all possible values that could
> occur, do the transformation, and then only pass the 1 transformed sample
> to the classifier. I guess that could be even slow though ...
>
> Best,
> Sebastian
>
> > On Aug 6, 2017, at 6:30 AM, Georg Heiler <georg.kf.heiler at gmail.com>
> wrote:
> >
> > @sebastian: thanks. Indeed, I am aware of this problem.
> >
> > I developed something here:
> https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b0799dc2ce but
> realized that the performance of prediction is pretty lame when there are
> around 100-150 columns used as the input.
> > Do you have some ideas how to speed this up?
> >
> > Regards,
> > Georg
> >
> > Joel Nothman <joel.nothman at gmail.com> schrieb am So., 6. Aug. 2017 um
> 00:49 Uhr:
> > We are working on CategoricalEncoder in
> https://github.com/scikit-learn/scikit-learn/pull/9151 to help users more
> with this kind of thing. Feedback and testing is welcome.
> >
> > On 6 August 2017 at 02:13, Sebastian Raschka <se.raschka at gmail.com>
> wrote:
> > Hi, Georg,
> >
> > I bring this up every time here on the mailing list :), and you probably
> aware of this issue, but it makes a difference whether your categorical
> data is nominal or ordinal. For instance if you have an ordinal variable
> like with values like {small, medium, large} you probably want to encode it
> as {1, 2, 3} or {1, 20, 100} or whatever is appropriate based on your
> domain knowledge regarding the variable. If you have sth like {blue, red,
> green} it may make more sense to do a one-hot encoding so that the
> classifier doesn't assume  a relationship between the variables like blue >
> red > green or sth like that.
> >
> > Now, the DictVectorizer and OneHotEncoder are both doing one hot
> encoding. The LabelEncoder does convert a variable to integer values, but
> if you have sth like {small, medium, large}, it wouldn't know the order (if
> that's an ordinal variable) and it would just assign arbitrary integers in
> increasing order. Thus, if you are dealing ordinal variables, there's no
> way around doing this manually; for example you could create mapping
> dictionaries for that (most conveniently done in pandas).
> >
> > Best,
> > Sebastian
> >
> > > On Aug 5, 2017, at 5:10 AM, Georg Heiler <georg.kf.heiler at gmail.com>
> wrote:
> > >
> > > Hi,
> > >
> > > the LabelEncooder is only meant for a single column i.e. target
> variable. Is the DictVectorizeer or a manual chaining of multiple
> LabelEncoders (one per categorical column) the desired way to get values
> which can be fed into a subsequent classifier?
> > >
> > > Is there some way I have overlooked which works better and possibly
> also can handle unseen values by applying most frequent imputation?
> > >
> > > regards,
> > > Georg
> > > _______________________________________________
> > > scikit-learn mailing list
> > > scikit-learn at python.org
> > > https://mail.python.org/mailman/listinfo/scikit-learn
> >
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
> >
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170807/419bc1b0/attachment.html>

From georg.kf.heiler at gmail.com  Mon Aug  7 02:41:39 2017
From: georg.kf.heiler at gmail.com (Georg Heiler)
Date: Mon, 07 Aug 2017 06:41:39 +0000
Subject: [scikit-learn] transform categorical data to numerical
 representation
In-Reply-To: <CAMhVi-Qs0ha2xhUFnLCTj11jPkK_dwKx6=9btK8UEVwF9xMN8w@mail.gmail.com>
References: <CAMhVi-QpAriKd8Yyrr-wf7HOyiyP1xUdqyqs3U4HVqniRE92Hw@mail.gmail.com>
 <6D0BF22C-9ABA-4C2A-B35B-210673439286@gmail.com>
 <CAAkaFLVr47Viuet6-YWe8qguA0c655cojK3t2i=z+DKevjp+eA@mail.gmail.com>
 <CAMhVi-RALPniqW+C5AMaMGVpYBEJpdG1V4W84Of+=u7JVZyeqQ@mail.gmail.com>
 <103609E5-E50B-4993-87F2-11661E7C7EB5@gmail.com>
 <CAMhVi-Qs0ha2xhUFnLCTj11jPkK_dwKx6=9btK8UEVwF9xMN8w@mail.gmail.com>
Message-ID: <CAMhVi-RZBgaNdi2H7LKuh-847-jZbP+KS+GkZZTQkr64yw3iBA@mail.gmail.com>

To my understanding pandas.factorize only works for the static case where
no unseen variables can occur.
Georg Heiler <georg.kf.heiler at gmail.com> schrieb am Mo. 7. Aug. 2017 um
08:40:

> I will need to look into factorize. Here is the result from profiling the
> transform method on a single new observation
> https://codereview.stackexchange.com/q/171622/132999
>
>
> Best Georg
> Sebastian Raschka <se.raschka at gmail.com> schrieb am So. 6. Aug. 2017 um
> 20:39:
>
>> > performance of prediction is pretty lame when there are around 100-150
>> columns used as the input.
>>
>> you are talking about computational performance when you are calling the
>> "transform" method? Have you done some profiling to find out where your
>> bottle neck (in the for loop) is? Just one a very quick look, I think this
>>
>> data.loc[~data[column].isin(fittedLabels), column] =
>> str(replacementForUnseen)
>>
>> is already very slow because fittedLabels is an array where you have O(n)
>> lookup instead of an average O(1) by using a hash table. Or is the isin
>> function converting it to a hashtable/set/dict?
>>
>> In general, would it maybe help to use pandas' factorize?
>> https://pandas.pydata.org/pandas-docs/stable/generated/pandas.factorize.html
>> For predict time, say you have only 1 example for prediction that needs
>> to be converted, you could append prototypes of all possible values that
>> could occur, do the transformation, and then only pass the 1 transformed
>> sample to the classifier. I guess that could be even slow though ...
>>
>> Best,
>> Sebastian
>>
>> > On Aug 6, 2017, at 6:30 AM, Georg Heiler <georg.kf.heiler at gmail.com>
>> wrote:
>> >
>> > @sebastian: thanks. Indeed, I am aware of this problem.
>> >
>> > I developed something here:
>> https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b0799dc2ce but
>> realized that the performance of prediction is pretty lame when there are
>> around 100-150 columns used as the input.
>> > Do you have some ideas how to speed this up?
>> >
>> > Regards,
>> > Georg
>> >
>> > Joel Nothman <joel.nothman at gmail.com> schrieb am So., 6. Aug. 2017 um
>> 00:49 Uhr:
>> > We are working on CategoricalEncoder in
>> https://github.com/scikit-learn/scikit-learn/pull/9151 to help users
>> more with this kind of thing. Feedback and testing is welcome.
>> >
>> > On 6 August 2017 at 02:13, Sebastian Raschka <se.raschka at gmail.com>
>> wrote:
>> > Hi, Georg,
>> >
>> > I bring this up every time here on the mailing list :), and you
>> probably aware of this issue, but it makes a difference whether your
>> categorical data is nominal or ordinal. For instance if you have an ordinal
>> variable like with values like {small, medium, large} you probably want to
>> encode it as {1, 2, 3} or {1, 20, 100} or whatever is appropriate based on
>> your domain knowledge regarding the variable. If you have sth like {blue,
>> red, green} it may make more sense to do a one-hot encoding so that the
>> classifier doesn't assume  a relationship between the variables like blue >
>> red > green or sth like that.
>> >
>> > Now, the DictVectorizer and OneHotEncoder are both doing one hot
>> encoding. The LabelEncoder does convert a variable to integer values, but
>> if you have sth like {small, medium, large}, it wouldn't know the order (if
>> that's an ordinal variable) and it would just assign arbitrary integers in
>> increasing order. Thus, if you are dealing ordinal variables, there's no
>> way around doing this manually; for example you could create mapping
>> dictionaries for that (most conveniently done in pandas).
>> >
>> > Best,
>> > Sebastian
>> >
>> > > On Aug 5, 2017, at 5:10 AM, Georg Heiler <georg.kf.heiler at gmail.com>
>> wrote:
>> > >
>> > > Hi,
>> > >
>> > > the LabelEncooder is only meant for a single column i.e. target
>> variable. Is the DictVectorizeer or a manual chaining of multiple
>> LabelEncoders (one per categorical column) the desired way to get values
>> which can be fed into a subsequent classifier?
>> > >
>> > > Is there some way I have overlooked which works better and possibly
>> also can handle unseen values by applying most frequent imputation?
>> > >
>> > > regards,
>> > > Georg
>> > > _______________________________________________
>> > > scikit-learn mailing list
>> > > scikit-learn at python.org
>> > > https://mail.python.org/mailman/listinfo/scikit-learn
>> >
>> > _______________________________________________
>> > scikit-learn mailing list
>> > scikit-learn at python.org
>> > https://mail.python.org/mailman/listinfo/scikit-learn
>> >
>> > _______________________________________________
>> > scikit-learn mailing list
>> > scikit-learn at python.org
>> > https://mail.python.org/mailman/listinfo/scikit-learn
>> > _______________________________________________
>> > scikit-learn mailing list
>> > scikit-learn at python.org
>> > https://mail.python.org/mailman/listinfo/scikit-learn
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170807/96027774/attachment-0001.html>

From andre.nascimento.melo at gmail.com  Thu Aug 10 09:55:22 2017
From: andre.nascimento.melo at gmail.com (=?UTF-8?Q?Andr=C3=A9_Melo?=)
Date: Thu, 10 Aug 2017 15:55:22 +0200
Subject: [scikit-learn] Truncated svd not working for complex matrices
Message-ID: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>

Hello all,

I'm trying to use the randomized version of scikit-learn's
TruncatedSVD (although I'm actually calling the internal function
randomized_svd to get the actual u, s, v matrices). While it is
working fine for real matrices, for complex matrices I can't get back
the original matrix even though the singular values are exactly
correct:

>>> import numpy as np
>>> from sklearn.utils.extmath import randomized_svd
>>> N = 3
>>> a = np.random.rand(N, N)*(1 + 1j)
>>> u1, s1, v1 = np.linalg.svd(a)
>>> u2, s2, v2 = randomized_svd(a, n_components=N, n_iter=7)
>>> np.allclose(s1, s2)
True
>>> np.allclose(a, u1.dot(np.diag(s1)).dot(v1))
True
>>> np.allclose(a, u2.dot(np.diag(s2)).dot(v2))
False

Any idea what could be wrong?

Thank you!

Best regards,
Andre Melo

From olivier.grisel at ensta.org  Thu Aug 10 10:13:16 2017
From: olivier.grisel at ensta.org (Olivier Grisel)
Date: Thu, 10 Aug 2017 16:13:16 +0200
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
Message-ID: <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>

I have no idea whether the randomized SVD method is supposed to work for
complex data or not (from a mathematical point of view). I think that all
scikit-learn estimators assume real data (or integer data for class labels)
and our input validation utilities will cast numeric values to float64 by
default. This might be the cause of your problem. Have a look at the source
code to confirm. The reference to the paper can also be found in the
docstring of those functions.

-- 
Olivier
?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170810/1041cdc6/attachment.html>

From andre.nascimento.melo at gmail.com  Thu Aug 10 10:56:43 2017
From: andre.nascimento.melo at gmail.com (=?UTF-8?Q?Andr=C3=A9_Melo?=)
Date: Thu, 10 Aug 2017 16:56:43 +0200
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
 <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
Message-ID: <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>

Hi Olivier,

Thank you very much for your reply. I was convinced it couldn't be a
fundamental mathematical issue because the singular values were coming
out exactly right, so it had to be a problem with the way complex
values were being handled.

I decided to look at the source code and it turns out the problem is
when the following transformation is applied:

U = np.dot(Q, Uhat)

Replacing this by

U = np.dot(Q.conj(), Uhat)

solves the issue! Should I report this on github?

On 10 August 2017 at 16:13, Olivier Grisel <olivier.grisel at ensta.org> wrote:
> I have no idea whether the randomized SVD method is supposed to work for
> complex data or not (from a mathematical point of view). I think that all
> scikit-learn estimators assume real data (or integer data for class labels)
> and our input validation utilities will cast numeric values to float64 by
> default. This might be the cause of your problem. Have a look at the source
> code to confirm. The reference to the paper can also be found in the
> docstring of those functions.
>
> --
> Olivier
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>

From andre.nascimento.melo at gmail.com  Thu Aug 10 11:08:09 2017
From: andre.nascimento.melo at gmail.com (=?UTF-8?Q?Andr=C3=A9_Melo?=)
Date: Thu, 10 Aug 2017 17:08:09 +0200
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
 <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
 <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>
Message-ID: <CAHZgJrhBaVgKcYDfM_-kPCOTrJtZSf4UFVxPE-bZDxGVim4qJA@mail.gmail.com>

Actually, it makes more sense to change

    B = safe_sparse_dot(Q.T, M)

To
    B = safe_sparse_dot(Q.T.conj(), M)

On 10 August 2017 at 16:56, Andr? Melo <andre.nascimento.melo at gmail.com> wrote:
> Hi Olivier,
>
> Thank you very much for your reply. I was convinced it couldn't be a
> fundamental mathematical issue because the singular values were coming
> out exactly right, so it had to be a problem with the way complex
> values were being handled.
>
> I decided to look at the source code and it turns out the problem is
> when the following transformation is applied:
>
> U = np.dot(Q, Uhat)
>
> Replacing this by
>
> U = np.dot(Q.conj(), Uhat)
>
> solves the issue! Should I report this on github?
>
> On 10 August 2017 at 16:13, Olivier Grisel <olivier.grisel at ensta.org> wrote:
>> I have no idea whether the randomized SVD method is supposed to work for
>> complex data or not (from a mathematical point of view). I think that all
>> scikit-learn estimators assume real data (or integer data for class labels)
>> and our input validation utilities will cast numeric values to float64 by
>> default. This might be the cause of your problem. Have a look at the source
>> code to confirm. The reference to the paper can also be found in the
>> docstring of those functions.
>>
>> --
>> Olivier
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>

From joel.nothman at gmail.com  Thu Aug 10 23:41:47 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Fri, 11 Aug 2017 13:41:47 +1000
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <CAHZgJrhBaVgKcYDfM_-kPCOTrJtZSf4UFVxPE-bZDxGVim4qJA@mail.gmail.com>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
 <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
 <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>
 <CAHZgJrhBaVgKcYDfM_-kPCOTrJtZSf4UFVxPE-bZDxGVim4qJA@mail.gmail.com>
Message-ID: <CAAkaFLXk0VY7Frf8OrS3Wv-=8F=PX7Ptt+vQS6rXnaFLL_Zmkg@mail.gmail.com>

Should we be more explicitly forbidding complex data in most estimators,
and perhaps allow it in a few where it is tested (particularly
decomposition)?

On 11 August 2017 at 01:08, Andr? Melo <andre.nascimento.melo at gmail.com>
wrote:

> Actually, it makes more sense to change
>
>     B = safe_sparse_dot(Q.T, M)
>
> To
>     B = safe_sparse_dot(Q.T.conj(), M)
>
> On 10 August 2017 at 16:56, Andr? Melo <andre.nascimento.melo at gmail.com>
> wrote:
> > Hi Olivier,
> >
> > Thank you very much for your reply. I was convinced it couldn't be a
> > fundamental mathematical issue because the singular values were coming
> > out exactly right, so it had to be a problem with the way complex
> > values were being handled.
> >
> > I decided to look at the source code and it turns out the problem is
> > when the following transformation is applied:
> >
> > U = np.dot(Q, Uhat)
> >
> > Replacing this by
> >
> > U = np.dot(Q.conj(), Uhat)
> >
> > solves the issue! Should I report this on github?
> >
> > On 10 August 2017 at 16:13, Olivier Grisel <olivier.grisel at ensta.org>
> wrote:
> >> I have no idea whether the randomized SVD method is supposed to work for
> >> complex data or not (from a mathematical point of view). I think that
> all
> >> scikit-learn estimators assume real data (or integer data for class
> labels)
> >> and our input validation utilities will cast numeric values to float64
> by
> >> default. This might be the cause of your problem. Have a look at the
> source
> >> code to confirm. The reference to the paper can also be found in the
> >> docstring of those functions.
> >>
> >> --
> >> Olivier
> >>
> >> _______________________________________________
> >> scikit-learn mailing list
> >> scikit-learn at python.org
> >> https://mail.python.org/mailman/listinfo/scikit-learn
> >>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/610d0562/attachment.html>

From drraph at gmail.com  Fri Aug 11 03:16:59 2017
From: drraph at gmail.com (Raphael C)
Date: Fri, 11 Aug 2017 09:16:59 +0200
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <CAAkaFLXk0VY7Frf8OrS3Wv-=8F=PX7Ptt+vQS6rXnaFLL_Zmkg@mail.gmail.com>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
 <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
 <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>
 <CAHZgJrhBaVgKcYDfM_-kPCOTrJtZSf4UFVxPE-bZDxGVim4qJA@mail.gmail.com>
 <CAAkaFLXk0VY7Frf8OrS3Wv-=8F=PX7Ptt+vQS6rXnaFLL_Zmkg@mail.gmail.com>
Message-ID: <CAFHc1QaqoTp7Me0rx30o1qxEXaELxv2k7H9b7YjR8cG8wk0b0A@mail.gmail.com>

Although the first priority should be correctness (in implementation
and documentation) and it makes sense to explicitly test for inputs
for which code will give the wrong answer, it would be great if we
could support complex data types, especially where it is very little
extra work.

Raphael

On 11 August 2017 at 05:41, Joel Nothman <joel.nothman at gmail.com> wrote:
> Should we be more explicitly forbidding complex data in most estimators, and
> perhaps allow it in a few where it is tested (particularly decomposition)?
>
> On 11 August 2017 at 01:08, Andr? Melo <andre.nascimento.melo at gmail.com>
> wrote:
>>
>> Actually, it makes more sense to change
>>
>>     B = safe_sparse_dot(Q.T, M)
>>
>> To
>>     B = safe_sparse_dot(Q.T.conj(), M)
>>
>> On 10 August 2017 at 16:56, Andr? Melo <andre.nascimento.melo at gmail.com>
>> wrote:
>> > Hi Olivier,
>> >
>> > Thank you very much for your reply. I was convinced it couldn't be a
>> > fundamental mathematical issue because the singular values were coming
>> > out exactly right, so it had to be a problem with the way complex
>> > values were being handled.
>> >
>> > I decided to look at the source code and it turns out the problem is
>> > when the following transformation is applied:
>> >
>> > U = np.dot(Q, Uhat)
>> >
>> > Replacing this by
>> >
>> > U = np.dot(Q.conj(), Uhat)
>> >
>> > solves the issue! Should I report this on github?
>> >
>> > On 10 August 2017 at 16:13, Olivier Grisel <olivier.grisel at ensta.org>
>> > wrote:
>> >> I have no idea whether the randomized SVD method is supposed to work
>> >> for
>> >> complex data or not (from a mathematical point of view). I think that
>> >> all
>> >> scikit-learn estimators assume real data (or integer data for class
>> >> labels)
>> >> and our input validation utilities will cast numeric values to float64
>> >> by
>> >> default. This might be the cause of your problem. Have a look at the
>> >> source
>> >> code to confirm. The reference to the paper can also be found in the
>> >> docstring of those functions.
>> >>
>> >> --
>> >> Olivier
>> >>
>> >> _______________________________________________
>> >> scikit-learn mailing list
>> >> scikit-learn at python.org
>> >> https://mail.python.org/mailman/listinfo/scikit-learn
>> >>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>

From sambarnett95 at gmail.com  Fri Aug 11 06:16:50 2017
From: sambarnett95 at gmail.com (Sam Barnett)
Date: Fri, 11 Aug 2017 11:16:50 +0100
Subject: [scikit-learn] Overflow Error with Cross-Validation (but not
 normally fitting the data)
Message-ID: <CACWGMfomO_1ignL-1kfi9t0YHjUOmdCdA_cGspjrtECaX8ZfBg@mail.gmail.com>

To all,

I am working on a scikit-learn estimator that performs a version of SVC
with a custom kernel. Unfortunately, I have been presented with a problem:
when running a grid search (or even using the cross_val_score function), my
estimator encounters an overflow error when evaluating my kernel
(specifically, in an array multiplication operation). What is particularly
strange about this is that, when I train the estimator on the whole
dataset, this error does not occur. In other words: the problem only
appears to occur when the data is split into folds. Is this something that
has been seen before? How ought I fix this?

I have attached the source code below (in particular, see the notebook for
how the problem arises).

Best,
Sam
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/fea2af24/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kernelsqizer.py
Type: text/x-python-script
Size: 2592 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/fea2af24/attachment-0003.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SeqSVC Toy Data Tests.ipynb
Type: application/octet-stream
Size: 6613 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/fea2af24/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: seqsvc.py
Type: text/x-python-script
Size: 11023 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/fea2af24/attachment-0004.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: timeseriestools.py
Type: text/x-python-script
Size: 1419 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/fea2af24/attachment-0005.bin>

From t3kcit at gmail.com  Fri Aug 11 12:37:12 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Fri, 11 Aug 2017 12:37:12 -0400
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <CAFHc1QaqoTp7Me0rx30o1qxEXaELxv2k7H9b7YjR8cG8wk0b0A@mail.gmail.com>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
 <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
 <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>
 <CAHZgJrhBaVgKcYDfM_-kPCOTrJtZSf4UFVxPE-bZDxGVim4qJA@mail.gmail.com>
 <CAAkaFLXk0VY7Frf8OrS3Wv-=8F=PX7Ptt+vQS6rXnaFLL_Zmkg@mail.gmail.com>
 <CAFHc1QaqoTp7Me0rx30o1qxEXaELxv2k7H9b7YjR8cG8wk0b0A@mail.gmail.com>
Message-ID: <d42e51e4-9045-a586-25a6-5e4b934652d5@gmail.com>

I opened https://github.com/scikit-learn/scikit-learn/issues/9528

I suggest to first error everywhere and then fix those for which it seems
easy and worth it, as Joel said, probably mostly in decomposition.

Though adding support even in a few places seems like dangerous feature 
creep.

On 08/11/2017 03:16 AM, Raphael C wrote:
> Although the first priority should be correctness (in implementation
> and documentation) and it makes sense to explicitly test for inputs
> for which code will give the wrong answer, it would be great if we
> could support complex data types, especially where it is very little
> extra work.
>
> Raphael
>
> On 11 August 2017 at 05:41, Joel Nothman <joel.nothman at gmail.com> wrote:
>> Should we be more explicitly forbidding complex data in most estimators, and
>> perhaps allow it in a few where it is tested (particularly decomposition)?
>>
>> On 11 August 2017 at 01:08, Andr? Melo <andre.nascimento.melo at gmail.com>
>> wrote:
>>> Actually, it makes more sense to change
>>>
>>>      B = safe_sparse_dot(Q.T, M)
>>>
>>> To
>>>      B = safe_sparse_dot(Q.T.conj(), M)
>>>
>>> On 10 August 2017 at 16:56, Andr? Melo <andre.nascimento.melo at gmail.com>
>>> wrote:
>>>> Hi Olivier,
>>>>
>>>> Thank you very much for your reply. I was convinced it couldn't be a
>>>> fundamental mathematical issue because the singular values were coming
>>>> out exactly right, so it had to be a problem with the way complex
>>>> values were being handled.
>>>>
>>>> I decided to look at the source code and it turns out the problem is
>>>> when the following transformation is applied:
>>>>
>>>> U = np.dot(Q, Uhat)
>>>>
>>>> Replacing this by
>>>>
>>>> U = np.dot(Q.conj(), Uhat)
>>>>
>>>> solves the issue! Should I report this on github?
>>>>
>>>> On 10 August 2017 at 16:13, Olivier Grisel <olivier.grisel at ensta.org>
>>>> wrote:
>>>>> I have no idea whether the randomized SVD method is supposed to work
>>>>> for
>>>>> complex data or not (from a mathematical point of view). I think that
>>>>> all
>>>>> scikit-learn estimators assume real data (or integer data for class
>>>>> labels)
>>>>> and our input validation utilities will cast numeric values to float64
>>>>> by
>>>>> default. This might be the cause of your problem. Have a look at the
>>>>> source
>>>>> code to confirm. The reference to the paper can also be found in the
>>>>> docstring of those functions.
>>>>>
>>>>> --
>>>>> Olivier
>>>>>
>>>>> _______________________________________________
>>>>> scikit-learn mailing list
>>>>> scikit-learn at python.org
>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn


From gael.varoquaux at normalesup.org  Fri Aug 11 12:45:31 2017
From: gael.varoquaux at normalesup.org (Gael Varoquaux)
Date: Fri, 11 Aug 2017 18:45:31 +0200
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <d42e51e4-9045-a586-25a6-5e4b934652d5@gmail.com>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
 <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
 <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>
 <CAHZgJrhBaVgKcYDfM_-kPCOTrJtZSf4UFVxPE-bZDxGVim4qJA@mail.gmail.com>
 <CAAkaFLXk0VY7Frf8OrS3Wv-=8F=PX7Ptt+vQS6rXnaFLL_Zmkg@mail.gmail.com>
 <CAFHc1QaqoTp7Me0rx30o1qxEXaELxv2k7H9b7YjR8cG8wk0b0A@mail.gmail.com>
 <d42e51e4-9045-a586-25a6-5e4b934652d5@gmail.com>
Message-ID: <20170811164531.GB3756445@phare.normalesup.org>

On Fri, Aug 11, 2017 at 12:37:12PM -0400, Andreas Mueller wrote:
> I opened https://github.com/scikit-learn/scikit-learn/issues/9528

> I suggest to first error everywhere and then fix those for which it seems
> easy and worth it, as Joel said, probably mostly in decomposition.

> Though adding support even in a few places seems like dangerous feature
> creep.

I am trying to predent that I am offline and in vacations, so I shouldn't
answer. But I do have a clear cut opinion here.

I believe that we should decide _not_ to support complex data everywhere.
The reason is that the support for complex data will always be incomplete
and risks being buggy. Indeed, complex data is very infrequent in machine
learning (unlike with signal processing). Hence, it will recieve little
usage. In addition, many machine learning algorithms cannot easily be
adapted to complex data.

To manage user expectation and to ensure quality of the codebase, let us
error on complex data.

Should we move this discussion on the issue opened by Andy?

Ga?l


> On 08/11/2017 03:16 AM, Raphael C wrote:
> >Although the first priority should be correctness (in implementation
> >and documentation) and it makes sense to explicitly test for inputs
> >for which code will give the wrong answer, it would be great if we
> >could support complex data types, especially where it is very little
> >extra work.

> >Raphael

> >On 11 August 2017 at 05:41, Joel Nothman <joel.nothman at gmail.com> wrote:
> >>Should we be more explicitly forbidding complex data in most estimators, and
> >>perhaps allow it in a few where it is tested (particularly decomposition)?

> >>On 11 August 2017 at 01:08, Andr? Melo <andre.nascimento.melo at gmail.com>
> >>wrote:
> >>>Actually, it makes more sense to change

> >>>     B = safe_sparse_dot(Q.T, M)

> >>>To
> >>>     B = safe_sparse_dot(Q.T.conj(), M)

> >>>On 10 August 2017 at 16:56, Andr? Melo <andre.nascimento.melo at gmail.com>
> >>>wrote:
> >>>>Hi Olivier,

> >>>>Thank you very much for your reply. I was convinced it couldn't be a
> >>>>fundamental mathematical issue because the singular values were coming
> >>>>out exactly right, so it had to be a problem with the way complex
> >>>>values were being handled.

> >>>>I decided to look at the source code and it turns out the problem is
> >>>>when the following transformation is applied:

> >>>>U = np.dot(Q, Uhat)

> >>>>Replacing this by

> >>>>U = np.dot(Q.conj(), Uhat)

> >>>>solves the issue! Should I report this on github?

> >>>>On 10 August 2017 at 16:13, Olivier Grisel <olivier.grisel at ensta.org>
> >>>>wrote:
> >>>>>I have no idea whether the randomized SVD method is supposed to work
> >>>>>for
> >>>>>complex data or not (from a mathematical point of view). I think that
> >>>>>all
> >>>>>scikit-learn estimators assume real data (or integer data for class
> >>>>>labels)
> >>>>>and our input validation utilities will cast numeric values to float64
> >>>>>by
> >>>>>default. This might be the cause of your problem. Have a look at the
> >>>>>source
> >>>>>code to confirm. The reference to the paper can also be found in the
> >>>>>docstring of those functions.

> >>>>>--
> >>>>>Olivier

> >>>>>_______________________________________________
> >>>>>scikit-learn mailing list
> >>>>>scikit-learn at python.org
> >>>>>https://mail.python.org/mailman/listinfo/scikit-learn

> >>>_______________________________________________
> >>>scikit-learn mailing list
> >>>scikit-learn at python.org
> >>>https://mail.python.org/mailman/listinfo/scikit-learn


> >>_______________________________________________
> >>scikit-learn mailing list
> >>scikit-learn at python.org
> >>https://mail.python.org/mailman/listinfo/scikit-learn

> >_______________________________________________
> >scikit-learn mailing list
> >scikit-learn at python.org
> >https://mail.python.org/mailman/listinfo/scikit-learn

> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-- 
    Gael Varoquaux
    Researcher, INRIA Parietal
    NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
    Phone:  ++ 33-1-69-08-79-68
    http://gael-varoquaux.info            http://twitter.com/GaelVaroquaux

From olivier.grisel at ensta.org  Fri Aug 11 17:49:13 2017
From: olivier.grisel at ensta.org (Olivier Grisel)
Date: Fri, 11 Aug 2017 23:49:13 +0200
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
Message-ID: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>

Grab it with pip or conda !

Quoting the release highlights from the website:

We are excited to release a number of great new features including
neighbors.LocalOutlierFactor for anomaly detection,
preprocessing.QuantileTransformer for robust feature transformation, and
the multioutput.ClassifierChain meta-estimator to simply account for
dependencies between classes in multilabel problems. We have some new
algorithms in existing estimators, such as multiplicative update in
decomposition.NMF and multinomial linear_model.LogisticRegression with L1
loss (use solver='saga').

Cross validation is now able to return the results from multiple metric
evaluations. The new model_selection.cross_validate can return many scores
on the test data as well as training set performance and timings, and we
have extended the scoring and refit parameters for grid/randomized search
to handle multiple metrics.

You can also learn faster. For instance, the new option to cache
transformations in pipeline.Pipeline makes grid search over pipelines
including slow transformations much more efficient. And you can predict
faster: if you?re sure you know what you?re doing, you can turn off
validating that the input is finite using config_context.

We?ve made some important fixes too. We?ve fixed a longstanding
implementation error in metrics.average_precision_score, so please be
cautious with prior results reported from that function. A number of errors
in the manifold.TSNE implementation have been fixed, particularly in the
default Barnes-Hut approximation. semi_supervised.LabelSpreading and
semi_supervised.LabelPropagation have had substantial fixes.
LabelPropagation was previously broken. LabelSpreading should now correctly
respect its alpha parameter.

Please see the full changelog at:

http://scikit-learn.org/0.19/whats_new.html#version-0-19

Notably some models have changed behaviors (bug fixes) and some methods or
parameters part of the public API have been deprecated.

A big thank you to anyone who made this release possible and Joel in
particular.

--
Olivier
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/5aaa3603/attachment.html>

From t3kcit at gmail.com  Fri Aug 11 17:57:03 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Fri, 11 Aug 2017 17:57:03 -0400
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
Message-ID: <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>

Thank you everybody for making the release possible, in particular 
Olivier and Joel :)

Wohoo!


From fabian.sippl at gmx.net  Fri Aug 11 17:57:32 2017
From: fabian.sippl at gmx.net (fabian.sippl at gmx.net)
Date: Fri, 11 Aug 2017 23:57:32 +0200
Subject: [scikit-learn] Question-Early Stopping MLPClassifer
 RandomizedSearchCV
Message-ID: <trinity-d41235f5-2dbd-41a5-9d4e-2b027fa8ac50-1502488652625@3c-app-gmx-bs75>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170811/6cd5afa3/attachment.html>

From g.lemaitre58 at gmail.com  Fri Aug 11 18:16:07 2017
From: g.lemaitre58 at gmail.com (=?UTF-8?Q?Guillaume_Lema=C3=AEtre?=)
Date: Sat, 12 Aug 2017 00:16:07 +0200
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
 <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>
Message-ID: <CACDxx9h0ZOr+ikoTDz6ArfDbUgZnmqsYBc4E0TcAU3Jyj9FvTg@mail.gmail.com>

Congrats guys!!!!

On 11 August 2017 at 23:57, Andreas Mueller <t3kcit at gmail.com> wrote:

> Thank you everybody for making the release possible, in particular Olivier
> and Joel :)
>
> Wohoo!
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>



-- 
Guillaume Lemaitre
INRIA Saclay - Parietal team
Center for Data Science Paris-Saclay
https://glemaitre.github.io/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170812/407f6b8b/attachment.html>

From gael.varoquaux at normalesup.org  Sat Aug 12 01:14:49 2017
From: gael.varoquaux at normalesup.org (Gael Varoquaux)
Date: Sat, 12 Aug 2017 07:14:49 +0200
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <CACDxx9h0ZOr+ikoTDz6ArfDbUgZnmqsYBc4E0TcAU3Jyj9FvTg@mail.gmail.com>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
 <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>
 <CACDxx9h0ZOr+ikoTDz6ArfDbUgZnmqsYBc4E0TcAU3Jyj9FvTg@mail.gmail.com>
Message-ID: <20170812051449.GD3225585@phare.normalesup.org>

Hurray, thank you everybody. This is a good one! (as always).

Ga?l

On Sat, Aug 12, 2017 at 12:16:07AM +0200, Guillaume Lema?tre wrote:
> Congrats guys!!!!

> On 11 August 2017 at 23:57, Andreas Mueller <t3kcit at gmail.com> wrote:

>     Thank you everybody for making the release possible, in particular Olivier
>     and Joel :)

>     Wohoo!

>     _______________________________________________
>     scikit-learn mailing list
>     scikit-learn at python.org
>     https://mail.python.org/mailman/listinfo/scikit-learn
-- 
    Gael Varoquaux
    Researcher, INRIA Parietal
    NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
    Phone:  ++ 33-1-69-08-79-68
    http://gael-varoquaux.info            http://twitter.com/GaelVaroquaux

From se.raschka at gmail.com  Sat Aug 12 01:19:41 2017
From: se.raschka at gmail.com (Sebastian Raschka)
Date: Sat, 12 Aug 2017 01:19:41 -0400
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <20170812051449.GD3225585@phare.normalesup.org>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
 <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>
 <CACDxx9h0ZOr+ikoTDz6ArfDbUgZnmqsYBc4E0TcAU3Jyj9FvTg@mail.gmail.com>
 <20170812051449.GD3225585@phare.normalesup.org>
Message-ID: <E47A6654-A38B-4155-AE30-B579C9E2CDBF@gmail.com>

Yay, as an avid user, thanks to all the developers! This is a great release indeed -- no breaking changes (at least for my code base) and so many improvements and additions (that I need to check out in detail) :)


> On Aug 12, 2017, at 1:14 AM, Gael Varoquaux <gael.varoquaux at normalesup.org> wrote:
> 
> Hurray, thank you everybody. This is a good one! (as always).
> 
> Ga?l
> 
> On Sat, Aug 12, 2017 at 12:16:07AM +0200, Guillaume Lema?tre wrote:
>> Congrats guys!!!!
> 
>> On 11 August 2017 at 23:57, Andreas Mueller <t3kcit at gmail.com> wrote:
> 
>>    Thank you everybody for making the release possible, in particular Olivier
>>    and Joel :)
> 
>>    Wohoo!
> 
>>    _______________________________________________
>>    scikit-learn mailing list
>>    scikit-learn at python.org
>>    https://mail.python.org/mailman/listinfo/scikit-learn
> -- 
>    Gael Varoquaux
>    Researcher, INRIA Parietal
>    NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
>    Phone:  ++ 33-1-69-08-79-68
>    http://gael-varoquaux.info            http://twitter.com/GaelVaroquaux
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn


From raga.markely at gmail.com  Sat Aug 12 01:32:38 2017
From: raga.markely at gmail.com (Raga Markely)
Date: Sat, 12 Aug 2017 01:32:38 -0400
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <E47A6654-A38B-4155-AE30-B579C9E2CDBF@gmail.com>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
 <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>
 <CACDxx9h0ZOr+ikoTDz6ArfDbUgZnmqsYBc4E0TcAU3Jyj9FvTg@mail.gmail.com>
 <20170812051449.GD3225585@phare.normalesup.org>
 <E47A6654-A38B-4155-AE30-B579C9E2CDBF@gmail.com>
Message-ID: <CAOLKFqt+zjYHZMiEcgGGE_ZG-Vev3hx18hJOawk5bQd3nMZhRQ@mail.gmail.com>

Thanks a lot for all the hard work and congratz!

Best,
Raga

On Aug 12, 2017 1:21 AM, "Sebastian Raschka" <se.raschka at gmail.com> wrote:

> Yay, as an avid user, thanks to all the developers! This is a great
> release indeed -- no breaking changes (at least for my code base) and so
> many improvements and additions (that I need to check out in detail) :)
>
>
> > On Aug 12, 2017, at 1:14 AM, Gael Varoquaux <
> gael.varoquaux at normalesup.org> wrote:
> >
> > Hurray, thank you everybody. This is a good one! (as always).
> >
> > Ga?l
> >
> > On Sat, Aug 12, 2017 at 12:16:07AM +0200, Guillaume Lema?tre wrote:
> >> Congrats guys!!!!
> >
> >> On 11 August 2017 at 23:57, Andreas Mueller <t3kcit at gmail.com> wrote:
> >
> >>    Thank you everybody for making the release possible, in particular
> Olivier
> >>    and Joel :)
> >
> >>    Wohoo!
> >
> >>    _______________________________________________
> >>    scikit-learn mailing list
> >>    scikit-learn at python.org
> >>    https://mail.python.org/mailman/listinfo/scikit-learn
> > --
> >    Gael Varoquaux
> >    Researcher, INRIA Parietal
> >    NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
> >    Phone:  ++ 33-1-69-08-79-68
> >    http://gael-varoquaux.info            http://twitter.com/
> GaelVaroquaux
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170812/bcf3830b/attachment-0001.html>

From ashimb9 at gmail.com  Sat Aug 12 02:32:57 2017
From: ashimb9 at gmail.com (Ashim Bhattarai)
Date: Sat, 12 Aug 2017 01:32:57 -0500
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <CAOLKFqt+zjYHZMiEcgGGE_ZG-Vev3hx18hJOawk5bQd3nMZhRQ@mail.gmail.com>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
 <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>
 <CACDxx9h0ZOr+ikoTDz6ArfDbUgZnmqsYBc4E0TcAU3Jyj9FvTg@mail.gmail.com>
 <20170812051449.GD3225585@phare.normalesup.org>
 <E47A6654-A38B-4155-AE30-B579C9E2CDBF@gmail.com>
 <CAOLKFqt+zjYHZMiEcgGGE_ZG-Vev3hx18hJOawk5bQd3nMZhRQ@mail.gmail.com>
Message-ID: <CAJhGTCW5FCgtddfYb9tSp_nx78hQsMye7xJukrwqwvdz6oprGA@mail.gmail.com>

Yes, thank you everyone!

On Sat, Aug 12, 2017 at 12:32 AM, Raga Markely <raga.markely at gmail.com>
wrote:

> Thanks a lot for all the hard work and congratz!
>
> Best,
> Raga
>
> On Aug 12, 2017 1:21 AM, "Sebastian Raschka" <se.raschka at gmail.com> wrote:
>
>> Yay, as an avid user, thanks to all the developers! This is a great
>> release indeed -- no breaking changes (at least for my code base) and so
>> many improvements and additions (that I need to check out in detail) :)
>>
>>
>> > On Aug 12, 2017, at 1:14 AM, Gael Varoquaux <
>> gael.varoquaux at normalesup.org> wrote:
>> >
>> > Hurray, thank you everybody. This is a good one! (as always).
>> >
>> > Ga?l
>> >
>> > On Sat, Aug 12, 2017 at 12:16:07AM +0200, Guillaume Lema?tre wrote:
>> >> Congrats guys!!!!
>> >
>> >> On 11 August 2017 at 23:57, Andreas Mueller <t3kcit at gmail.com> wrote:
>> >
>> >>    Thank you everybody for making the release possible, in particular
>> Olivier
>> >>    and Joel :)
>> >
>> >>    Wohoo!
>> >
>> >>    _______________________________________________
>> >>    scikit-learn mailing list
>> >>    scikit-learn at python.org
>> >>    https://mail.python.org/mailman/listinfo/scikit-learn
>> > --
>> >    Gael Varoquaux
>> >    Researcher, INRIA Parietal
>> >    NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
>> >    Phone:  ++ 33-1-69-08-79-68
>> >    http://gael-varoquaux.info            http://twitter.com/GaelVaroqua
>> ux
>> > _______________________________________________
>> > scikit-learn mailing list
>> > scikit-learn at python.org
>> > https://mail.python.org/mailman/listinfo/scikit-learn
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170812/107df6c5/attachment.html>

From bertrand.thirion at inria.fr  Sat Aug 12 04:50:43 2017
From: bertrand.thirion at inria.fr (bthirion)
Date: Sat, 12 Aug 2017 10:50:43 +0200
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
Message-ID: <539f18e9-8e59-7294-bd3e-26ae24928f3f@inria.fr>

Congratulations for all these improvements and for orchestrating the 
release !
Bertrand


On 11/08/2017 23:49, Olivier Grisel wrote:
> Grab it with pip or conda !
>
> Quoting the release highlights from the website:
>
> We are excited to release a number of great new features including 
> neighbors.LocalOutlierFactor for anomaly detection, 
> preprocessing.QuantileTransformer for robust feature transformation, 
> and the multioutput.ClassifierChain meta-estimator to simply account 
> for dependencies between classes in multilabel problems. We have some 
> new algorithms in existing estimators, such as multiplicative update 
> in decomposition.NMF and multinomial linear_model.LogisticRegression 
> with L1 loss (use solver='saga').
>
> Cross validation is now able to return the results from multiple 
> metric evaluations. The new model_selection.cross_validate can return 
> many scores on the test data as well as training set performance and 
> timings, and we have extended the scoring and refit parameters for 
> grid/randomized search to handle multiple metrics.
>
> You can also learn faster. For instance, the new option to cache 
> transformations in pipeline.Pipeline makes grid search over pipelines 
> including slow transformations much more efficient. And you can 
> predict faster: if you?re sure you know what you?re doing, you can 
> turn off validating that the input is finite using config_context.
>
> We?ve made some important fixes too. We?ve fixed a longstanding 
> implementation error in metrics.average_precision_score, so please be 
> cautious with prior results reported from that function. A number of 
> errors in the manifold.TSNE implementation have been fixed, 
> particularly in the default Barnes-Hut approximation. 
> semi_supervised.LabelSpreading and semi_supervised.LabelPropagation 
> have had substantial fixes. LabelPropagation was previously broken. 
> LabelSpreading should now correctly respect its alpha parameter.
>
> Please see the full changelog at:
>
> http://scikit-learn.org/0.19/whats_new.html#version-0-19
>
> Notably some models have changed behaviors (bug fixes) and some 
> methods or parameters part of the public API have been deprecated.
>
> A big thank you to anyone who made this release possible and Joel in 
> particular.
>
> --
> Olivier
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170812/84c0e80d/attachment.html>

From valerio.maggio at gmail.com  Sat Aug 12 05:18:08 2017
From: valerio.maggio at gmail.com (Valerio Maggio)
Date: Sat, 12 Aug 2017 09:18:08 +0000
Subject: [scikit-learn] scikit-learn 0.19.0 is out!
In-Reply-To: <E47A6654-A38B-4155-AE30-B579C9E2CDBF@gmail.com>
References: <CAFvE7K7=h9=V2h3HtohXcmCFOO4AAa74ZKOAD1eYXgeDW6tU_w@mail.gmail.com>
 <77c26ae0-808a-3d87-1912-c634edc1fb7c@gmail.com>
 <CACDxx9h0ZOr+ikoTDz6ArfDbUgZnmqsYBc4E0TcAU3Jyj9FvTg@mail.gmail.com>
 <20170812051449.GD3225585@phare.normalesup.org>
 <E47A6654-A38B-4155-AE30-B579C9E2CDBF@gmail.com>
Message-ID: <CAEt-TDxuy2AYrAJHBc8EtLG--XRKyiNyg=FDR=uL-vZ4mStCzQ@mail.gmail.com>

On Sat, 12 Aug 2017 at 07:20, Sebastian Raschka <se.raschka at gmail.com>
wrote:

> Yay, as an avid user, thanks to all the developers! This is a great
> release indeed -- no breaking changes (at least for my code base) and so
> many improvements and additions (that I need to check out in detail) :)


Quoting Sebastian: totally agree!!
+1

Thanks a lot for this super new release and for all these improvements.

Cheers
Valerio


>
>
> > On Aug 12, 2017, at 1:14 AM, Gael Varoquaux <
> gael.varoquaux at normalesup.org> wrote:
> >
> > Hurray, thank you everybody. This is a good one! (as always).
> >
> > Ga?l
> >
> > On Sat, Aug 12, 2017 at 12:16:07AM +0200, Guillaume Lema?tre wrote:
> >> Congrats guys!!!!
> >
> >> On 11 August 2017 at 23:57, Andreas Mueller <t3kcit at gmail.com> wrote:
> >
> >>    Thank you everybody for making the release possible, in particular
> Olivier
> >>    and Joel :)
> >
> >>    Wohoo!
> >
> >>    _______________________________________________
> >>    scikit-learn mailing list
> >>    scikit-learn at python.org
> >>    https://mail.python.org/mailman/listinfo/scikit-learn
> > --
> >    Gael Varoquaux
> >    Researcher, INRIA Parietal
> >    NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
> >    Phone:  ++ 33-1-69-08-79-68
> >    http://gael-varoquaux.info
> http://twitter.com/GaelVaroquaux
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-- 
# valerio
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170812/9f37b5c8/attachment-0001.html>

From alexandre.gramfort at telecom-paristech.fr  Sat Aug 12 08:11:13 2017
From: alexandre.gramfort at telecom-paristech.fr (Alexandre Gramfort)
Date: Sat, 12 Aug 2017 14:11:13 +0200
Subject: [scikit-learn] Truncated svd not working for complex matrices
In-Reply-To: <20170811164531.GB3756445@phare.normalesup.org>
References: <CAHZgJrimsj51q6PmhGGBi0RSGc6PL9tbk8QqM173bD22xqOHEg@mail.gmail.com>
 <CAFvE7K40tpMbtW=RUB0k2h0y=DBte9TgvRzRRDP_vjOrRd_cLA@mail.gmail.com>
 <CAHZgJri1wwDWUN4-gyLxe53MLjVYKsjzHO-Zcs7N8v5OaULzOw@mail.gmail.com>
 <CAHZgJrhBaVgKcYDfM_-kPCOTrJtZSf4UFVxPE-bZDxGVim4qJA@mail.gmail.com>
 <CAAkaFLXk0VY7Frf8OrS3Wv-=8F=PX7Ptt+vQS6rXnaFLL_Zmkg@mail.gmail.com>
 <CAFHc1QaqoTp7Me0rx30o1qxEXaELxv2k7H9b7YjR8cG8wk0b0A@mail.gmail.com>
 <d42e51e4-9045-a586-25a6-5e4b934652d5@gmail.com>
 <20170811164531.GB3756445@phare.normalesup.org>
Message-ID: <CADeotZqXddr48h61Y7qQsqYRwBFHh74f0qEjOg8u2=N7KA-t1g@mail.gmail.com>

I agree with Ga?l on this. If you want to support complex values just
copy the estimators / functions you want and maintain them in a
separate package. +1 to error when complex are passed.

From b.noushin7 at gmail.com  Sun Aug 13 12:16:53 2017
From: b.noushin7 at gmail.com (Ariani A)
Date: Sun, 13 Aug 2017 12:16:53 -0400
Subject: [scikit-learn] No module named crluster.hierarchical
Message-ID: <CAEUHNkb5Fq4qdADxCkgPF+YRZtNRCGWLgQSmRurnCpRGtCVp6w@mail.gmail.com>

Dear all,

I am writing this import:

    from sklearn.crluster.hierarchical import (_hc_cut, _TREE_BUILDERS,
                                          linkage_tree)
But it gives this error:
ImportError: No module named crluster.hierarchical

Any clue?
Best regards,
-Noushin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170813/38c5e01b/attachment.html>

From zephyr14 at gmail.com  Sun Aug 13 12:20:27 2017
From: zephyr14 at gmail.com (Vlad Niculae)
Date: Sun, 13 Aug 2017 12:20:27 -0400
Subject: [scikit-learn] No module named crluster.hierarchical
In-Reply-To: <CAEUHNkb5Fq4qdADxCkgPF+YRZtNRCGWLgQSmRurnCpRGtCVp6w@mail.gmail.com>
References: <CAEUHNkb5Fq4qdADxCkgPF+YRZtNRCGWLgQSmRurnCpRGtCVp6w@mail.gmail.com>
Message-ID: <CAFJw_eGScaqJVb=qUb=+voS3T5ZKtV8AZTRUH5k4rH7rcYVikA@mail.gmail.com>

Looks like you're misspelling the word "cluster".

Yours,
Vlad

On Aug 13, 2017 12:19 PM, "Ariani A" <b.noushin7 at gmail.com> wrote:

> Dear all,
>
> I am writing this import:
>
>     from sklearn.crluster.hierarchical import (_hc_cut, _TREE_BUILDERS,
>                                           linkage_tree)
> But it gives this error:
> ImportError: No module named crluster.hierarchical
>
> Any clue?
> Best regards,
> -Noushin
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170813/482066e7/attachment.html>

From b.noushin7 at gmail.com  Sun Aug 13 12:25:49 2017
From: b.noushin7 at gmail.com (Ariani A)
Date: Sun, 13 Aug 2017 12:25:49 -0400
Subject: [scikit-learn] No module named crluster.hierarchical
In-Reply-To: <CAFJw_eGScaqJVb=qUb=+voS3T5ZKtV8AZTRUH5k4rH7rcYVikA@mail.gmail.com>
References: <CAEUHNkb5Fq4qdADxCkgPF+YRZtNRCGWLgQSmRurnCpRGtCVp6w@mail.gmail.com>
 <CAFJw_eGScaqJVb=qUb=+voS3T5ZKtV8AZTRUH5k4rH7rcYVikA@mail.gmail.com>
Message-ID: <CAEUHNkaqeH+nQFdXyMkEeQbiJu5t=5k5Uba+dY+pNpf68yjR=g@mail.gmail.com>

Thank you so much!

On Sun, Aug 13, 2017 at 12:20 PM, Vlad Niculae <zephyr14 at gmail.com> wrote:

> Looks like you're misspelling the word "cluster".
>
> Yours,
> Vlad
>
> On Aug 13, 2017 12:19 PM, "Ariani A" <b.noushin7 at gmail.com> wrote:
>
>> Dear all,
>>
>> I am writing this import:
>>
>>     from sklearn.crluster.hierarchical import (_hc_cut, _TREE_BUILDERS,
>>                                           linkage_tree)
>> But it gives this error:
>> ImportError: No module named crluster.hierarchical
>>
>> Any clue?
>> Best regards,
>> -Noushin
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170813/346523cc/attachment.html>

From t3kcit at gmail.com  Mon Aug 14 10:05:56 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Mon, 14 Aug 2017 10:05:56 -0400
Subject: [scikit-learn] Question-Early Stopping MLPClassifer
 RandomizedSearchCV
In-Reply-To: <trinity-d41235f5-2dbd-41a5-9d4e-2b027fa8ac50-1502488652625@3c-app-gmx-bs75>
References: <trinity-d41235f5-2dbd-41a5-9d4e-2b027fa8ac50-1502488652625@3c-app-gmx-bs75>
Message-ID: <d6368d15-e5be-df27-b0f9-43c15ac9209d@gmail.com>

Yes, you understood correctly.
You can see the implementation in the code:
https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neural_network/multilayer_perceptron.py#L491

It calls ``train_test_split``, so it's a random subset of the data. 
Currently the API doesn't allow providing your own validation set.
What is the use-case for that?

Andy

On 08/11/2017 05:57 PM, fabian.sippl at gmx.net wrote:
> Hello Scikit-Learn Team,
> I?ve got a question concerning the implementation of Early Stopping in 
> MLPClassifier. I am using it in combination with RandomizedSearchCV. 
> The fraction used for validation in early stopping is set with the 
> parameter validation_fraction of MLPClassifier. How is the validaton 
> set extracted from the training set ? Does the function simply take 
> the last X % from the training set ? Is there a possibility to 
> manually set this validation set ?
> I wonder whether I correctly understand the functionality: The neural 
> net is trained on the training data and the performance is evaluated 
> after every epoch on the validation set (which is internally selected 
> by the MLPClassifer)? If the Net stops training, the performance on 
> the left out data (Parameter "cv" in RandomizedSearch) is determined ?
> Thank you very much for your help !
> Kind Regards,
> Fabian Sippl
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170814/cf86ff8c/attachment.html>

From georg.kf.heiler at gmail.com  Wed Aug 16 07:28:21 2017
From: georg.kf.heiler at gmail.com (Georg Heiler)
Date: Wed, 16 Aug 2017 11:28:21 +0000
Subject: [scikit-learn] caching transformers during hyper parameter
 optimization
Message-ID: <CAMhVi-RjgQK_T-d88hARLbgYc_u31NNT6LXTP7ZtmeCbdcV98w@mail.gmail.com>

There is a new option in the pipeline:
http://scikit-learn.org/stable/modules/pipeline.html#pipeline-cache
How can I use this to also store the transformed data as I only want to
compute the last step i.e. estimator during hyper parameter tuning and not
the transform methods of the clean steps?

Is there a possibility to apply this for crossvalidation? I would want to
see all the folds precomputed and stored to disk in a folder.

Regards,
Georg
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170816/1ceb0a9c/attachment.html>

From joel.nothman at gmail.com  Wed Aug 16 07:51:19 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Wed, 16 Aug 2017 21:51:19 +1000
Subject: [scikit-learn] caching transformers during hyper parameter
 optimization
In-Reply-To: <CAMhVi-RjgQK_T-d88hARLbgYc_u31NNT6LXTP7ZtmeCbdcV98w@mail.gmail.com>
References: <CAMhVi-RjgQK_T-d88hARLbgYc_u31NNT6LXTP7ZtmeCbdcV98w@mail.gmail.com>
Message-ID: <CAAkaFLVVEV=ZRA=CKUubXhXvTWaDSjnXDZW9qU_jJepGqmer0A@mail.gmail.com>

We certainly considered this over the many years that Pipeline caching has
been in the pipeline. Storing the fitted model means we can do both a
fit_transform and a transform on new data, and in many cases takes away the
pain point of CV over pipelines where downstream steps are varied.

What transformer are you using where the transform is costly? Or is it more
a matter of you wanting to store the transformed data at each step?

There are custom ways to do this sort of thing generically with a mixin if
you really want.

On 16 August 2017 at 21:28, Georg Heiler <georg.kf.heiler at gmail.com> wrote:

> There is a new option in the pipeline: http://scikit-learn.
> org/stable/modules/pipeline.html#pipeline-cache
> How can I use this to also store the transformed data as I only want to
> compute the last step i.e. estimator during hyper parameter tuning and not
> the transform methods of the clean steps?
>
> Is there a possibility to apply this for crossvalidation? I would want to
> see all the folds precomputed and stored to disk in a folder.
>
> Regards,
> Georg
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170816/5ff856bb/attachment.html>

From georg.kf.heiler at gmail.com  Wed Aug 16 12:53:28 2017
From: georg.kf.heiler at gmail.com (Georg Heiler)
Date: Wed, 16 Aug 2017 16:53:28 +0000
Subject: [scikit-learn] caching transformers during hyper parameter
 optimization
In-Reply-To: <CAAkaFLVVEV=ZRA=CKUubXhXvTWaDSjnXDZW9qU_jJepGqmer0A@mail.gmail.com>
References: <CAMhVi-RjgQK_T-d88hARLbgYc_u31NNT6LXTP7ZtmeCbdcV98w@mail.gmail.com>
 <CAAkaFLVVEV=ZRA=CKUubXhXvTWaDSjnXDZW9qU_jJepGqmer0A@mail.gmail.com>
Message-ID: <CAMhVi-R=zu-QzgLYXFpNat+5EaNs+fp5_Qe-QRFycbLLQaU=gg@mail.gmail.com>

Data cleaning @ enrichment

Could you link an example for a mixing?

Currently this is a bit if a mess with custom pickle persistence in a big
for loop and custom transformers

Thanks.
Georg
Joel Nothman <joel.nothman at gmail.com> schrieb am Mi. 16. Aug. 2017 um 13:51:

> We certainly considered this over the many years that Pipeline caching has
> been in the pipeline. Storing the fitted model means we can do both a
> fit_transform and a transform on new data, and in many cases takes away the
> pain point of CV over pipelines where downstream steps are varied.
>
> What transformer are you using where the transform is costly? Or is it
> more a matter of you wanting to store the transformed data at each step?
>
> There are custom ways to do this sort of thing generically with a mixin if
> you really want.
>
> On 16 August 2017 at 21:28, Georg Heiler <georg.kf.heiler at gmail.com>
> wrote:
>
>> There is a new option in the pipeline:
>> http://scikit-learn.org/stable/modules/pipeline.html#pipeline-cache
>> How can I use this to also store the transformed data as I only want to
>> compute the last step i.e. estimator during hyper parameter tuning and not
>> the transform methods of the clean steps?
>>
>> Is there a possibility to apply this for crossvalidation? I would want to
>> see all the folds precomputed and stored to disk in a folder.
>>
>> Regards,
>> Georg
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170816/9590f6fb/attachment.html>

From joel.nothman at gmail.com  Wed Aug 16 21:15:03 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Thu, 17 Aug 2017 11:15:03 +1000
Subject: [scikit-learn] caching transformers during hyper parameter
 optimization
In-Reply-To: <CAMhVi-R=zu-QzgLYXFpNat+5EaNs+fp5_Qe-QRFycbLLQaU=gg@mail.gmail.com>
References: <CAMhVi-RjgQK_T-d88hARLbgYc_u31NNT6LXTP7ZtmeCbdcV98w@mail.gmail.com>
 <CAAkaFLVVEV=ZRA=CKUubXhXvTWaDSjnXDZW9qU_jJepGqmer0A@mail.gmail.com>
 <CAMhVi-R=zu-QzgLYXFpNat+5EaNs+fp5_Qe-QRFycbLLQaU=gg@mail.gmail.com>
Message-ID: <CAAkaFLWmOaUjcdt1H2iofydknecZJOfRQNr4BriGwk78q2TNaw@mail.gmail.com>

Now this isn't the best example, because joblib.Memory isn't going to be
very fast at dumping a list of strings, but I hope you can get the idea
from https://gist.github.com/jnothman/019d594d197c98a3d6192fa0cb19c850


On 17 August 2017 at 02:53, Georg Heiler <georg.kf.heiler at gmail.com> wrote:

> Data cleaning @ enrichment
>
> Could you link an example for a mixing?
>
> Currently this is a bit if a mess with custom pickle persistence in a big
> for loop and custom transformers
>
> Thanks.
> Georg
> Joel Nothman <joel.nothman at gmail.com> schrieb am Mi. 16. Aug. 2017 um
> 13:51:
>
>> We certainly considered this over the many years that Pipeline caching
>> has been in the pipeline. Storing the fitted model means we can do both a
>> fit_transform and a transform on new data, and in many cases takes away the
>> pain point of CV over pipelines where downstream steps are varied.
>>
>> What transformer are you using where the transform is costly? Or is it
>> more a matter of you wanting to store the transformed data at each step?
>>
>> There are custom ways to do this sort of thing generically with a mixin
>> if you really want.
>>
>> On 16 August 2017 at 21:28, Georg Heiler <georg.kf.heiler at gmail.com>
>> wrote:
>>
>>> There is a new option in the pipeline: http://scikit-learn.
>>> org/stable/modules/pipeline.html#pipeline-cache
>>> How can I use this to also store the transformed data as I only want to
>>> compute the last step i.e. estimator during hyper parameter tuning and not
>>> the transform methods of the clean steps?
>>>
>>> Is there a possibility to apply this for crossvalidation? I would want
>>> to see all the folds precomputed and stored to disk in a folder.
>>>
>>> Regards,
>>> Georg
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/ea6224c8/attachment.html>

From sambarnett95 at gmail.com  Thu Aug 17 05:22:21 2017
From: sambarnett95 at gmail.com (Sam Barnett)
Date: Thu, 17 Aug 2017 10:22:21 +0100
Subject: [scikit-learn] Malformed input for
 SVC(kernel='precomputed').predict()
Message-ID: <CACWGMfq30hSJ+iVZDkQA6b+39uJF6sCYJ4V4DMeyd2nyU4Xw0g@mail.gmail.com>

I am rolling classifier based on SVC which computes a custom Gram matrix
and runs this through the SVC classifier with kernel = 'precomputed'. While
this works fine with the fit method, I face a dilemma with the predict
method, shown here:


    def predict(self, X):
        """Run the predict method of the previously-instantiated SVM
        classifier, returning the predicted classes for test set X."""

        # Check is fit had been called
        check_is_fitted(self, ['X_', 'y_'])

        # Input validation
        X = check_array(X)

        cut_off = self.cut_ord_pair[0]
        order = self.cut_ord_pair[1]

        X_gram = seq_kernel_free(X, self.X_, \
        pri_kernel=kernselect(self.kernel, self.coef0, self.gamma,
self.degree, self.scale), \
        cut_off=cut_off, order=order)

        X_gram = np.nan_to_num(X_gram)

        return self.ord_svc_.predict(X_gram)

This will run on any dataset just fine. However, it fails the
check_estimator test. Specifically, when trying to raise an error for
malformed input on predict (in check_classifiers_train), it says that a
ValueError is not raised. Yet if I change the order of X and self.X_ in
seq_kernel_free (which computes the [n_samples_train, n_samples_test] Gram
matrix), it passes the check_estimator test yet fails to run the predict
method.

How do I resolve both issues simultaneously?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/8194a94c/attachment-0001.html>

From georg.kf.heiler at gmail.com  Thu Aug 17 07:50:33 2017
From: georg.kf.heiler at gmail.com (Georg Heiler)
Date: Thu, 17 Aug 2017 11:50:33 +0000
Subject: [scikit-learn] Categorical handling
Message-ID: <CAMhVi-RWWMhcjCmHr11-=5hb3+_GTG5UZnfHZz0-nvDBD+ZH=w@mail.gmail.com>

Hi,

how can I properly handle categorical values in scikit-learn?
https://stackoverflow.com/questions/45727934/pandas-categories-new-levels?noredirect=1#comment78424496_45727934


goals

   - scikit-learn syle fit/transform methods to encode labels of
   categorical features of X
   - should handle unseen labels
   - should be faster than running a label encoder manually for each fold
   and manually checking if the label already was seen in the training data
   i.e. what I currently do (
   https://stackoverflow.com/questions/45727934/pandas-categories-new-levels?noredirect=1#comment78424496_45727934
which
   links to https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b0799dc2ce
   )
   - only some columns are categorical, and only these should be converted


Regards,
Georg
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/0a2d2c9b/attachment.html>

From t3kcit at gmail.com  Thu Aug 17 11:03:16 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Thu, 17 Aug 2017 11:03:16 -0400
Subject: [scikit-learn] Malformed input for
 SVC(kernel='precomputed').predict()
In-Reply-To: <CACWGMfq30hSJ+iVZDkQA6b+39uJF6sCYJ4V4DMeyd2nyU4Xw0g@mail.gmail.com>
References: <CACWGMfq30hSJ+iVZDkQA6b+39uJF6sCYJ4V4DMeyd2nyU4Xw0g@mail.gmail.com>
Message-ID: <0d66111e-3e35-869b-9753-1f20cb118216@gmail.com>

Hi Sam.

Can you say which test fails exactly and where (i.e. give traceback)?
The estimator checks are currently quite strict with respect to raising 
helpful error messages.
That doesn't mean your estimator is broken (necessarily).
With a precomputed gram matrix, I expect the shape of X in predict to be 
(n_samples_test, n_samples_train), right?
Does you estimator have a _pairwise attribute? (It should to work with 
cross-validation, I'm not sure if it's
used in the estimator checks right now, but it should).

Your feedback will help making check_estimator be more robust. I don't 
think it's tested with anything that requires
"precomputed" kernels.

Thanks

Andy

On 08/17/2017 05:22 AM, Sam Barnett wrote:
> I am rolling classifier based on SVC which computes a custom Gram 
> matrix and runs this through the SVC classifier with kernel = 
> 'precomputed'. While this works fine with the fit method, I face a 
> dilemma with the predict method, shown here:
>
>
> def predict(self, X):
> """Run the predict method of the previously-instantiated SVM
>       classifier, returning the predicted classes for test set X."""
>
> # Check is fit had been called
> check_is_fitted(self, ['X_', 'y_'])
>
>         # Input validation
>         X = check_array(X)
>
>         cut_off = self.cut_ord_pair[0]
>         order = self.cut_ord_pair[1]
>
>         X_gram = seq_kernel_free(X, self.X_, \
> pri_kernel=kernselect(self.kernel, self.coef0, self.gamma, 
> self.degree, self.scale), \
> cut_off=cut_off, order=order)
>
>         X_gram = np.nan_to_num(X_gram)
>
>         return self.ord_svc_.predict(X_gram)
>
> This will run on any dataset just fine. However, it fails the 
> check_estimator test. Specifically, when trying to raise an error for 
> malformed input on predict (in check_classifiers_train), it says that 
> a ValueError is not raised. Yet if I change the order of X and self.X_ 
> in seq_kernel_free (which computes the [n_samples_train, 
> n_samples_test] Gram matrix), it passes the check_estimator test yet 
> fails to run the predict method.
>
> How do I resolve both issues simultaneously?
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/c0f8fee3/attachment.html>

From t3kcit at gmail.com  Thu Aug 17 11:11:49 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Thu, 17 Aug 2017 11:11:49 -0400
Subject: [scikit-learn] Categorical handling
In-Reply-To: <CAMhVi-RWWMhcjCmHr11-=5hb3+_GTG5UZnfHZz0-nvDBD+ZH=w@mail.gmail.com>
References: <CAMhVi-RWWMhcjCmHr11-=5hb3+_GTG5UZnfHZz0-nvDBD+ZH=w@mail.gmail.com>
Message-ID: <dbaebe05-8236-4415-b3b3-56acddaae26f@gmail.com>

Hi Georg.
Unfortunately this is not entirely trivial right now, but will be fixed by
https://github.com/scikit-learn/scikit-learn/pull/9151
and
https://github.com/scikit-learn/scikit-learn/pull/9012
which will be in the next release (0.20).

LabelBinarizer is probably the best work-around for now, and selecting 
columns can be done (awkwardly)
like in this example: 
http://scikit-learn.org/dev/auto_examples/hetero_feature_union.html#sphx-glr-auto-examples-hetero-feature-union-py

Best,
Andy

On 08/17/2017 07:50 AM, Georg Heiler wrote:
> Hi,
>
> how can I properly handle categorical values in scikit-learn?
> https://stackoverflow.com/questions/45727934/pandas-categories-new-levels?noredirect=1#comment78424496_45727934 
>
>
> goals
>
>   * scikit-learn syle fit/transform methods to encode labels of
>     categorical features of X
>   * should handle unseen labels
>   * should be faster than running a label encoder manually for each
>     fold and manually checking if the label already was seen in the
>     training data i.e. what I currently do
>     (https://stackoverflow.com/questions/45727934/pandas-categories-new-levels?noredirect=1#comment78424496_45727934 which
>     links to
>     https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b0799dc2ce)
>   * only some columns are categorical, and only these should be converted
>
>
> Regards,
> Georg
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/9b5f4f24/attachment-0001.html>

From joel.nothman at gmail.com  Thu Aug 17 11:26:13 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Fri, 18 Aug 2017 01:26:13 +1000
Subject: [scikit-learn] Categorical handling
In-Reply-To: <dbaebe05-8236-4415-b3b3-56acddaae26f@gmail.com>
References: <CAMhVi-RWWMhcjCmHr11-=5hb3+_GTG5UZnfHZz0-nvDBD+ZH=w@mail.gmail.com>
 <dbaebe05-8236-4415-b3b3-56acddaae26f@gmail.com>
Message-ID: <CAAkaFLW_BfM4BV-JKutXYVpDc_H5Gpewc2aEFw2DH+sNdNOCdQ@mail.gmail.com>

I don't consider LabelBinarizer the best workaround.

Given a Pandas dataframe df, I'd use:

DictVectorizer().fit_transform(df.to_dict(orient='records'))

which will handle encoding strings with one-hot and numerical features as
column vectors. Or:

class PandasVectorizer(DictVectorizer):
    def fit(self, x, y=None):
        return super(PandasVectorizer, self).fit(x.to_dict('records'))
    def fit_transform(self, x, y=None):
        return super(PandasVectorizer,
self).fit_transform(x.to_dict('records'))
    def transform(self, x):
        return super(PandasVectorizer, self).transform(x.to_dict('records'))


On 18 August 2017 at 01:11, Andreas Mueller <t3kcit at gmail.com> wrote:

> Hi Georg.
> Unfortunately this is not entirely trivial right now, but will be fixed by
> https://github.com/scikit-learn/scikit-learn/pull/9151
> and
> https://github.com/scikit-learn/scikit-learn/pull/9012
> which will be in the next release (0.20).
>
> LabelBinarizer is probably the best work-around for now, and selecting
> columns can be done (awkwardly)
> like in this example: http://scikit-learn.org/dev/
> auto_examples/hetero_feature_union.html#sphx-glr-auto-
> examples-hetero-feature-union-py
>
> Best,
> Andy
>
>
> On 08/17/2017 07:50 AM, Georg Heiler wrote:
>
> Hi,
>
> how can I properly handle categorical values in scikit-learn?
> https://stackoverflow.com/questions/45727934/pandas-categories-new-levels?
> noredirect=1#comment78424496_45727934
>
> goals
>
>    - scikit-learn syle fit/transform methods to encode labels of
>    categorical features of X
>    - should handle unseen labels
>    - should be faster than running a label encoder manually for each fold
>    and manually checking if the label already was seen in the training data
>    i.e. what I currently do (https://stackoverflow.com/
>    questions/45727934/pandas-categories-new-levels?
>    noredirect=1#comment78424496_45727934
>    <https://stackoverflow.com/questions/45727934/pandas-categories-new-levels?noredirect=1#comment78424496_45727934> which
>    links to https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b0799dc2
>    ce)
>    - only some columns are categorical, and only these should be converted
>
>
> Regards,
> Georg
>
>
> _______________________________________________
> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170818/15305614/attachment.html>

From joel.nothman at gmail.com  Thu Aug 17 11:27:43 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Fri, 18 Aug 2017 01:27:43 +1000
Subject: [scikit-learn] Categorical handling
In-Reply-To: <CAAkaFLW_BfM4BV-JKutXYVpDc_H5Gpewc2aEFw2DH+sNdNOCdQ@mail.gmail.com>
References: <CAMhVi-RWWMhcjCmHr11-=5hb3+_GTG5UZnfHZz0-nvDBD+ZH=w@mail.gmail.com>
 <dbaebe05-8236-4415-b3b3-56acddaae26f@gmail.com>
 <CAAkaFLW_BfM4BV-JKutXYVpDc_H5Gpewc2aEFw2DH+sNdNOCdQ@mail.gmail.com>
Message-ID: <CAAkaFLUo0Nu+y5KysNy0B8NhN2WELwhJSifpNkpjAj68CJo2Cg@mail.gmail.com>

gist at https://gist.github.com/jnothman/a75bac778c1eb9661017555249e50379

On 18 August 2017 at 01:26, Joel Nothman <joel.nothman at gmail.com> wrote:

> I don't consider LabelBinarizer the best workaround.
>
> Given a Pandas dataframe df, I'd use:
>
> DictVectorizer().fit_transform(df.to_dict(orient='records'))
>
> which will handle encoding strings with one-hot and numerical features as
> column vectors. Or:
>
> class PandasVectorizer(DictVectorizer):
>     def fit(self, x, y=None):
>         return super(PandasVectorizer, self).fit(x.to_dict('records'))
>     def fit_transform(self, x, y=None):
>         return super(PandasVectorizer, self).fit_transform(x.to_dict(
> 'records'))
>     def transform(self, x):
>         return super(PandasVectorizer, self).transform(x.to_dict('
> records'))
>
>
> On 18 August 2017 at 01:11, Andreas Mueller <t3kcit at gmail.com> wrote:
>
>> Hi Georg.
>> Unfortunately this is not entirely trivial right now, but will be fixed by
>> https://github.com/scikit-learn/scikit-learn/pull/9151
>> and
>> https://github.com/scikit-learn/scikit-learn/pull/9012
>> which will be in the next release (0.20).
>>
>> LabelBinarizer is probably the best work-around for now, and selecting
>> columns can be done (awkwardly)
>> like in this example: http://scikit-learn.org/dev/au
>> to_examples/hetero_feature_union.html#sphx-glr-auto-examples
>> -hetero-feature-union-py
>>
>> Best,
>> Andy
>>
>>
>> On 08/17/2017 07:50 AM, Georg Heiler wrote:
>>
>> Hi,
>>
>> how can I properly handle categorical values in scikit-learn?
>> https://stackoverflow.com/questions/45727934/pandas-categori
>> es-new-levels?noredirect=1#comment78424496_45727934
>>
>> goals
>>
>>    - scikit-learn syle fit/transform methods to encode labels of
>>    categorical features of X
>>    - should handle unseen labels
>>    - should be faster than running a label encoder manually for each
>>    fold and manually checking if the label already was seen in the training
>>    data i.e. what I currently do (https://stackoverflow.com/que
>>    stions/45727934/pandas-categories-new-levels?noredirect=1#
>>    comment78424496_45727934
>>    <https://stackoverflow.com/questions/45727934/pandas-categories-new-levels?noredirect=1#comment78424496_45727934> which
>>    links to https://gist.github.com/geoHeil/5caff5236b4850d673b2c9b07
>>    99dc2ce)
>>    - only some columns are categorical, and only these should be
>>    converted
>>
>>
>> Regards,
>> Georg
>>
>>
>> _______________________________________________
>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170818/746b804e/attachment.html>

From sambarnett95 at gmail.com  Thu Aug 17 13:21:24 2017
From: sambarnett95 at gmail.com (Sam Barnett)
Date: Thu, 17 Aug 2017 18:21:24 +0100
Subject: [scikit-learn] Malformed input for
 SVC(kernel='precomputed').predict()
In-Reply-To: <0d66111e-3e35-869b-9753-1f20cb118216@gmail.com>
References: <CACWGMfq30hSJ+iVZDkQA6b+39uJF6sCYJ4V4DMeyd2nyU4Xw0g@mail.gmail.com>
 <0d66111e-3e35-869b-9753-1f20cb118216@gmail.com>
Message-ID: <CACWGMfriHTDVdZ3gxPd6uOkm8qySXH19cBA=ar-G+wnPmS07UA@mail.gmail.com>

Hi Andy,

Please find attached a Jupyter notebook showing exactly where the problem
appears.

Best,
Sam

On Thu, Aug 17, 2017 at 4:03 PM, Andreas Mueller <t3kcit at gmail.com> wrote:

> Hi Sam.
>
> Can you say which test fails exactly and where (i.e. give traceback)?
> The estimator checks are currently quite strict with respect to raising
> helpful error messages.
> That doesn't mean your estimator is broken (necessarily).
> With a precomputed gram matrix, I expect the shape of X in predict to be
> (n_samples_test, n_samples_train), right?
> Does you estimator have a _pairwise attribute? (It should to work with
> cross-validation, I'm not sure if it's
> used in the estimator checks right now, but it should).
>
> Your feedback will help making check_estimator be more robust. I don't
> think it's tested with anything that requires
> "precomputed" kernels.
>
> Thanks
>
> Andy
>
>
> On 08/17/2017 05:22 AM, Sam Barnett wrote:
>
> I am rolling classifier based on SVC which computes a custom Gram matrix
> and runs this through the SVC classifier with kernel = 'precomputed'. While
> this works fine with the fit method, I face a dilemma with the predict
> method, shown here:
>
>
>     def predict(self, X):
>         """Run the predict method of the previously-instantiated SVM
>         classifier, returning the predicted classes for test set X."""
>
>         # Check is fit had been called
>         check_is_fitted(self, ['X_', 'y_'])
>
>         # Input validation
>         X = check_array(X)
>
>         cut_off = self.cut_ord_pair[0]
>         order = self.cut_ord_pair[1]
>
>         X_gram = seq_kernel_free(X, self.X_, \
>         pri_kernel=kernselect(self.kernel, self.coef0, self.gamma,
> self.degree, self.scale), \
>         cut_off=cut_off, order=order)
>
>         X_gram = np.nan_to_num(X_gram)
>
>         return self.ord_svc_.predict(X_gram)
>
> This will run on any dataset just fine. However, it fails the
> check_estimator test. Specifically, when trying to raise an error for
> malformed input on predict (in check_classifiers_train), it says that a
> ValueError is not raised. Yet if I change the order of X and self.X_ in
> seq_kernel_free (which computes the [n_samples_train, n_samples_test] Gram
> matrix), it passes the check_estimator test yet fails to run the predict
> method.
>
> How do I resolve both issues simultaneously?
>
>
>
> _______________________________________________
> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/ac538adf/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kernelsqizer.py
Type: text/x-python-script
Size: 5142 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/ac538adf/attachment-0003.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SeqSVC Check Estimator Test.ipynb
Type: application/octet-stream
Size: 9321 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/ac538adf/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: seqsvc_v2.py
Type: text/x-python-script
Size: 10890 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/ac538adf/attachment-0004.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: timeseriestools.py
Type: text/x-python-script
Size: 1419 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/ac538adf/attachment-0005.bin>

From mcapizzi at email.arizona.edu  Thu Aug 17 14:46:50 2017
From: mcapizzi at email.arizona.edu (Michael Capizzi)
Date: Thu, 17 Aug 2017 11:46:50 -0700
Subject: [scikit-learn] any interest in incorporating a new Transformer?
Message-ID: <CAAX6ikFJciCVrR9RfRSAbRcuGWbDOLBtAgkJJui3EDTdDUCLLw@mail.gmail.com>

Hi all -

Forgive me if this is the wrong place for posting this question, but I'd
like to inquire about the community's interest in incorporating a new
Transformer into the code base.

This paper ( https://nlp.stanford.edu/pubs/sidaw12_simple_sentiment.pdf )
is a "classic" in Natural Language Processing and is often times used as a
very competitive baseline.  TL;DR it transforms a traditional count-based
feature space into the conditional probabilities of a `Naive Bayes`
classifier.  These transformed features can then be used to train any
linear classifier.  The paper focuses on `SVM`.

The attached notebook has an example of the custom `Transformer` I built
along with a custom `Classifier` to utilize this `Transformer` in a
`multiclass` case (as the feature space transformation differs depending on
the label).

If there is interest in the community for the inclusion of this
`Transformer` and `Classifier`, I'd happily go through the official process
of a `pull-request`, etc.

-Michael
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/2cdb076a/attachment-0002.html>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170817/2cdb076a/attachment-0003.html>

From joel.nothman at gmail.com  Sat Aug 19 05:47:06 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Sat, 19 Aug 2017 19:47:06 +1000
Subject: [scikit-learn] any interest in incorporating a new Transformer?
In-Reply-To: <CAAX6ikFJciCVrR9RfRSAbRcuGWbDOLBtAgkJJui3EDTdDUCLLw@mail.gmail.com>
References: <CAAX6ikFJciCVrR9RfRSAbRcuGWbDOLBtAgkJJui3EDTdDUCLLw@mail.gmail.com>
Message-ID: <CAAkaFLWTRKbKJt7PDNab3=vbsaq4hjn1Z92SefvouGNU9B+rNQ@mail.gmail.com>

this is the right place to ask, but I'd be more interested to see a
scikit-learn-compatible implementation available, perhaps in
scikit-learn-contrib more than to see it part of the main package...

On 19 Aug 2017 2:13 am, "Michael Capizzi" <mcapizzi at email.arizona.edu>
wrote:

> Hi all -
>
> Forgive me if this is the wrong place for posting this question, but I'd
> like to inquire about the community's interest in incorporating a new
> Transformer into the code base.
>
> This paper ( https://nlp.stanford.edu/pubs/sidaw12_simple_sentiment.pdf )
> is a "classic" in Natural Language Processing and is often times used as a
> very competitive baseline.  TL;DR it transforms a traditional count-based
> feature space into the conditional probabilities of a `Naive Bayes`
> classifier.  These transformed features can then be used to train any
> linear classifier.  The paper focuses on `SVM`.
>
> The attached notebook has an example of the custom `Transformer` I built
> along with a custom `Classifier` to utilize this `Transformer` in a
> `multiclass` case (as the feature space transformation differs depending on
> the label).
>
> If there is interest in the community for the inclusion of this
> `Transformer` and `Classifier`, I'd happily go through the official process
> of a `pull-request`, etc.
>
> -Michael
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170819/faf2b8c9/attachment.html>

From mcapizzi at email.arizona.edu  Sat Aug 19 18:36:21 2017
From: mcapizzi at email.arizona.edu (Michael Capizzi)
Date: Sat, 19 Aug 2017 15:36:21 -0700
Subject: [scikit-learn] any interest in incorporating a new Transformer?
In-Reply-To: <CAAkaFLWTRKbKJt7PDNab3=vbsaq4hjn1Z92SefvouGNU9B+rNQ@mail.gmail.com>
References: <CAAX6ikFJciCVrR9RfRSAbRcuGWbDOLBtAgkJJui3EDTdDUCLLw@mail.gmail.com>
 <CAAkaFLWTRKbKJt7PDNab3=vbsaq4hjn1Z92SefvouGNU9B+rNQ@mail.gmail.com>
Message-ID: <CAAX6ikESc2scpJhUm1pXMLvr5Jm-OtgEbFMjZW7oroNLq3H9wA@mail.gmail.com>

Thanks @joel -

I wasn?t aware of scikit-learn-contrib. Is this what you?re referring to?
https://github.com/scikit-learn-contrib/scikit-learn-contrib

If so, I don?t see any existing projects that this would fit into; could I
start a new one in a pull-request?

-M
?

On Sat, Aug 19, 2017 at 2:47 AM, Joel Nothman <joel.nothman at gmail.com>
wrote:

> this is the right place to ask, but I'd be more interested to see a
> scikit-learn-compatible implementation available, perhaps in
> scikit-learn-contrib more than to see it part of the main package...
>
> On 19 Aug 2017 2:13 am, "Michael Capizzi" <mcapizzi at email.arizona.edu>
> wrote:
>
>> Hi all -
>>
>> Forgive me if this is the wrong place for posting this question, but I'd
>> like to inquire about the community's interest in incorporating a new
>> Transformer into the code base.
>>
>> This paper ( https://nlp.stanford.edu/pubs/sidaw12_simple_sentiment.pdf )
>> is a "classic" in Natural Language Processing and is often times used as a
>> very competitive baseline.  TL;DR it transforms a traditional count-based
>> feature space into the conditional probabilities of a `Naive Bayes`
>> classifier.  These transformed features can then be used to train any
>> linear classifier.  The paper focuses on `SVM`.
>>
>> The attached notebook has an example of the custom `Transformer` I built
>> along with a custom `Classifier` to utilize this `Transformer` in a
>> `multiclass` case (as the feature space transformation differs depending on
>> the label).
>>
>> If there is interest in the community for the inclusion of this
>> `Transformer` and `Classifier`, I'd happily go through the official process
>> of a `pull-request`, etc.
>>
>> -Michael
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170819/4e3bfb11/attachment.html>

From joel.nothman at gmail.com  Sun Aug 20 08:28:44 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Sun, 20 Aug 2017 22:28:44 +1000
Subject: [scikit-learn] any interest in incorporating a new Transformer?
In-Reply-To: <CAAX6ikESc2scpJhUm1pXMLvr5Jm-OtgEbFMjZW7oroNLq3H9wA@mail.gmail.com>
References: <CAAX6ikFJciCVrR9RfRSAbRcuGWbDOLBtAgkJJui3EDTdDUCLLw@mail.gmail.com>
 <CAAkaFLWTRKbKJt7PDNab3=vbsaq4hjn1Z92SefvouGNU9B+rNQ@mail.gmail.com>
 <CAAX6ikESc2scpJhUm1pXMLvr5Jm-OtgEbFMjZW7oroNLq3H9wA@mail.gmail.com>
Message-ID: <CAAkaFLW01e_E+wvavdE+iPhnkYwvn+n3S8R+YrMipQJ8m1sowg@mail.gmail.com>

The idea is to take the template (
https://github.com/scikit-learn-contrib/project-template), build, test and
document your estimator(s), and offer it to be housed within
scikit-learn-contrib.

On 20 August 2017 at 08:36, Michael Capizzi <mcapizzi at email.arizona.edu>
wrote:

> Thanks @joel -
>
> I wasn?t aware of scikit-learn-contrib. Is this what you?re referring to?
> https://github.com/scikit-learn-contrib/scikit-learn-contrib
>
> If so, I don?t see any existing projects that this would fit into; could I
> start a new one in a pull-request?
>
> -M
> ?
>
> On Sat, Aug 19, 2017 at 2:47 AM, Joel Nothman <joel.nothman at gmail.com>
> wrote:
>
>> this is the right place to ask, but I'd be more interested to see a
>> scikit-learn-compatible implementation available, perhaps in
>> scikit-learn-contrib more than to see it part of the main package...
>>
>> On 19 Aug 2017 2:13 am, "Michael Capizzi" <mcapizzi at email.arizona.edu>
>> wrote:
>>
>>> Hi all -
>>>
>>> Forgive me if this is the wrong place for posting this question, but I'd
>>> like to inquire about the community's interest in incorporating a new
>>> Transformer into the code base.
>>>
>>> This paper ( https://nlp.stanford.edu/pubs/sidaw12_simple_sentiment.pdf )
>>> is a "classic" in Natural Language Processing and is often times used as a
>>> very competitive baseline.  TL;DR it transforms a traditional count-based
>>> feature space into the conditional probabilities of a `Naive Bayes`
>>> classifier.  These transformed features can then be used to train any
>>> linear classifier.  The paper focuses on `SVM`.
>>>
>>> The attached notebook has an example of the custom `Transformer` I built
>>> along with a custom `Classifier` to utilize this `Transformer` in a
>>> `multiclass` case (as the feature space transformation differs depending on
>>> the label).
>>>
>>> If there is interest in the community for the inclusion of this
>>> `Transformer` and `Classifier`, I'd happily go through the official process
>>> of a `pull-request`, etc.
>>>
>>> -Michael
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170820/5d8ced86/attachment.html>

From s.atasever at gmail.com  Mon Aug 21 10:11:25 2017
From: s.atasever at gmail.com (Sema Atasever)
Date: Mon, 21 Aug 2017 17:11:25 +0300
Subject: [scikit-learn] How can i write the birch prediction results to the
 file
Message-ID: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>

Dear scikit-learn developers,

I have a text file where the columns represent the 22 features and the rows
represent the amino asid . (you can see in the attachment)


I want to apply hierarchical clustering to this database usign
*sklearn.cluster.Birch
algorithm.*

There are too many prediction results and it is not possible to see them on
the screen.
How can i write the birch prediction results to the file?

I would appreciate if you could advise on some methods.
Thanks.

*Birch Codes:*
from sklearn.cluster import Birch
import numpy as np

X=np.loadtxt(open("C:\class1.txt", "rb"), delimiter=";")

brc = Birch(branching_factor=50, n_clusters=None,
threshold=0.5,compute_labels=True,copy=True)

brc.fit(X)

centroids = brc.subcluster_centers_

labels = brc.subcluster_labels_
n_clusters = np.unique(labels).size
brc.predict(X)

print("\n brc.predict(X)")
print(brc.predict(X))
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170821/c0f2bf84/attachment-0001.html>
-------------- next part --------------
0.1877;0.386705;0.242412;0.175513;0.109395;0.5;0;0.244492;0.25402;0.501485;0.08978;0.30011;0.610105;0.399205;0.20793;0.392865;0.183304;0.450634;0.181905;0.119275;0.441045;0.15519
0.175513;0.395115;0.244492;0.25402;0.20793;0.5;0;0.183304;0.450634;0.36606;0.085465;0.30111;0.61342;0.39889;0.181905;0.419205;0.196001;0.284247;0.15519;0.116745;0.508985;0.120175
0.096681;0.508985;0.231754;0.22298;0.122875;0.982014;0;0.239757;0.362949;0.397294;0.14375;0.30115;0.5551;0.5607;0.1547;0.2846;0.238086;0.365937;0.155065;0.112465;0.58629;0.114685
0.22298;0.54238;0.239757;0.362949;0.1547;0.5;0.976014;0.238086;0.365937;0.395978;0.11412;0.302565;0.58332;0.591975;0.155065;0.25296;0.234884;0.292462;0.114685;0.08463;0.46578;0.03225
0.362949;0.5607;0.238086;0.365937;0.155065;0.952574;0;0.234884;0.292462;0.472657;0.112465;0.262565;0.624975;0.58629;0.114685;0.29903;0.185031;0.109242;0.03225;0.061355;0.30821;0.01616
0.362957;0.305915;0.101388;0.502346;0.193875;0.5;0;0.092671;0.540357;0.36697;0.042315;0.565685;0.391995;0.23211;0.203875;0.564015;0.075362;0.544133;0.20143;0.039395;0.177705;0.17322
0.502346;0.253195;0.092671;0.540357;0.203875;0.982014;0;0.075362;0.544133;0.380503;0.0419;0.58211;0.37599;0.180535;0.20143;0.61803;0.07361;0.518633;0.17322;0.040375;0.25297;0.31168
0.540357;0.23211;0.075362;0.544133;0.20143;0.119203;0;0.07361;0.518633;0.40776;0.039395;0.5865;0.37411;0.177705;0.17322;0.64908;0.099123;0.540488;0.31168;0.039185;0.297655;0.30087
0.544133;0.180535;0.07361;0.518633;0.17322;0.952574;0;0.099123;0.540488;0.36039;0.040375;0.52303;0.436595;0.25297;0.31168;0.435355;0.137794;0.547889;0.30087;0.035415;0.30434;0.295825
0.518633;0.177705;0.099123;0.540488;0.31168;0.5;0.976014;0.137794;0.547889;0.314322;0.039185;0.513795;0.447025;0.297655;0.30087;0.401485;0.137281;0.490202;0.295825;0.02889;0.29;0.23639
0.540488;0.25297;0.137794;0.547889;0.30087;0.952574;0;0.137281;0.490202;0.372519;0.035415;0.45937;0.50522;0.30434;0.295825;0.399835;0.130684;0.374903;0.23639;0.018825;0.24624;0.125565
0.547889;0.297655;0.137281;0.490202;0.295825;0.268941;0;0.130684;0.374903;0.494415;0.02889;0.20978;0.76133;0.29;0.23639;0.473615;0.089682;0.161964;0.125565;0.01771;0.254795;0.057755
0.235259;0.30101;0.114724;0.370166;0.17405;0.982014;0;0.11486;0.25665;0.628491;0.006885;0.096835;0.89629;0.33127;0.151235;0.51749;0.103503;0.216193;0.07958;0.01084;0.37686;0.02358
0.16566;0.37686;0.213938;0.173933;0.00739;0.119203;0;0.256584;0.252658;0.490756;0.020135;0.1854;0.79446;0.69126;0.020875;0.287865;0.288979;0.309277;0.03181;0.022485;0.82069;0.035805
0.173933;0.57118;0.256584;0.252658;0.020875;0.731059;0;0.288979;0.309277;0.401748;0.02228;0.24233;0.735395;0.835785;0.03181;0.13241;0.284203;0.319514;0.035805;0.0206;0.862155;0.034595
0.252658;0.69126;0.288979;0.309277;0.03181;0.268941;0;0.284203;0.319514;0.396281;0.022485;0.24967;0.72784;0.82069;0.035805;0.143505;0.29731;0.315762;0.034595;0.015885;0.81669;0.02245
0.309277;0.835785;0.284203;0.319514;0.035805;0.731059;0;0.29731;0.315762;0.38693;0.0206;0.230605;0.7488;0.862155;0.034595;0.10325;0.281889;0.206203;0.02245;0.00778;0.72839;0.012685
0.319514;0.82069;0.29731;0.315762;0.034595;0.119203;0;0.281889;0.206203;0.511907;0.015885;0.167175;0.816935;0.81669;0.02245;0.16086;0.288807;0.137009;0.012685;0.034445;0.592785;0.00568
0.060841;0.607885;0.283237;0.126276;0.03798;0.119203;0;0.280505;0.224908;0.494589;0.1794;0.120365;0.700235;0.65246;0.07413;0.273415;0.285956;0.228121;0.076115;0.23053;0.589895;0.063305
0.126276;0.66572;0.280505;0.224908;0.07413;0.982014;0;0.285956;0.228121;0.485922;0.2269;0.122395;0.65071;0.619125;0.076115;0.304755;0.278271;0.250223;0.063305;0.206875;0.373615;0.037035
0.224908;0.65246;0.285956;0.228121;0.076115;0.119203;0;0.278271;0.250223;0.471506;0.23053;0.104645;0.664825;0.589895;0.063305;0.3468;0.198532;0.132265;0.037035;0.213415;0.413025;0.04198
0.228121;0.619125;0.278271;0.250223;0.063305;0.047426;0;0.198532;0.132265;0.669203;0.206875;0.021325;0.7718;0.373615;0.037035;0.58935;0.214268;0.220627;0.04198;0.213965;0.49312;0.049615
0.250223;0.589895;0.198532;0.132265;0.037035;0.268941;0;0.214268;0.220627;0.565108;0.213415;0.01544;0.77115;0.413025;0.04198;0.545;0.270003;0.192249;0.049615;0.18857;0.45497;0.034465
0.132265;0.373615;0.214268;0.220627;0.04198;0.880797;0;0.270003;0.192249;0.537747;0.213965;0.014525;0.771505;0.49312;0.049615;0.457265;0.276269;0.146924;0.034465;0.14891;0.454565;0.03569
0.146924;0.45497;0.254095;0.136505;0.03569;0.731059;0;0.239351;0.203895;0.556755;0.135675;0.019835;0.84449;0.547205;0.034755;0.418045;0.225775;0.151238;0.019;0.03194;0.45914;0.01009
0.044534;0.537675;0.219391;0.148928;0.031155;0.119203;0;0.233709;0.110772;0.655518;0.085805;0.00849;0.9057;0.605745;0.033535;0.36072;0.223357;0.163415;0.069595;0.09819;0.359975;0.07867
0.148928;0.605765;0.233709;0.110772;0.033535;0.268941;0.976014;0.223357;0.163415;0.613229;0.093765;0.069415;0.83682;0.567775;0.069595;0.36263;0.155202;0.247704;0.07867;0.091425;0.29815;0.095765
0.110772;0.605745;0.223357;0.163415;0.069595;0.119203;0;0.155202;0.247704;0.597093;0.09819;0.18264;0.71917;0.359975;0.07867;0.561355;0.131984;0.322047;0.095765;0.09255;0.283705;0.30327
0.163415;0.567775;0.155202;0.247704;0.07867;0.268941;0;0.131984;0.322047;0.545971;0.091425;0.228625;0.67995;0.29815;0.095765;0.60609;0.127338;0.533373;0.30327;0.093525;0.285145;0.439565
0.247704;0.359975;0.131984;0.322047;0.095765;0.119203;0;0.127338;0.533373;0.339288;0.09255;0.47722;0.43023;0.283705;0.30327;0.413025;0.127969;0.661972;0.439565;0.09255;0.28159;0.519385
0.322047;0.29815;0.127338;0.533373;0.30327;0.880797;0;0.127969;0.661972;0.21006;0.093525;0.60772;0.29876;0.285145;0.439565;0.27529;0.126322;0.676301;0.519385;0.087105;0.261645;0.567635
0.533373;0.283705;0.127969;0.661972;0.439565;0.5;0;0.126322;0.676301;0.197377;0.09255;0.68466;0.22279;0.28159;0.519385;0.199025;0.148321;0.693539;0.567635;0.07853;0.200985;0.52805
0.661972;0.285145;0.126322;0.676301;0.519385;0.952574;0;0.148321;0.693539;0.158141;0.087105;0.69281;0.22009;0.261645;0.567635;0.17072;0.124622;0.695229;0.52805;0.066455;0.18649;0.347945
0.676301;0.28159;0.148321;0.693539;0.567635;0.268941;0;0.124622;0.695229;0.180149;0.07853;0.67916;0.24231;0.200985;0.52805;0.270965;0.115651;0.449274;0.347945;0.062215;0.181915;0.127245
0.051303;0.166175;0.097584;0.010879;0.020105;0.047426;0;0.09575;0.18132;0.722928;0.04626;0.02684;0.926895;0.23601;0.02507;0.73892;0.132114;0.3721;0.14666;0.093275;0.319155;0.157535
0.010879;0.21263;0.09575;0.18132;0.02507;0.119203;0;0.132114;0.3721;0.495786;0.06582;0.190955;0.743225;0.32563;0.14666;0.52771;0.139137;0.47708;0.157535;0.10699;0.333655;0.2156
0.18132;0.23601;0.132114;0.3721;0.14666;0.268941;0;0.139137;0.47708;0.383782;0.093275;0.28367;0.623055;0.319155;0.157535;0.52331;0.148548;0.534862;0.2156;0.139145;0.36626;0.24109
0.3721;0.32563;0.139137;0.47708;0.157535;0.119203;0;0.148548;0.534862;0.316592;0.10699;0.398985;0.494025;0.333655;0.2156;0.45075;0.170184;0.549754;0.24109;0.1508;0.367415;0.24485
0.47708;0.319155;0.148548;0.534862;0.2156;0.952574;0;0.170184;0.549754;0.280064;0.139145;0.51996;0.340895;0.36626;0.24109;0.392655;0.174615;0.551945;0.24485;0.163915;0.47935;0.3
0.534862;0.333655;0.170184;0.549754;0.24109;0.268941;0;0.174615;0.551945;0.273441;0.1508;0.533245;0.315955;0.367415;0.24485;0.38774;0.216316;0.520662;0.3;0.16551;0.544115;0.33293
0.549754;0.36626;0.174615;0.551945;0.24485;0.5;0.976014;0.216316;0.520662;0.263024;0.163915;0.56274;0.273345;0.47935;0.3;0.220655;0.238555;0.581823;0.33293;0.165685;0.566745;0.33161
0.551945;0.367415;0.216316;0.520662;0.3;0.5;0;0.238555;0.581823;0.17962;0.16551;0.59629;0.2382;0.544115;0.33293;0.12295;0.246606;0.584443;0.33161;0.14973;0.552795;0.32441
0.520662;0.47935;0.238555;0.581823;0.33293;0.731059;0;0.246606;0.584443;0.168951;0.165685;0.56978;0.264535;0.566745;0.33161;0.101645;0.268405;0.373439;0.32441;0.12076;0.510895;0.198525
0.581823;0.544115;0.246606;0.584443;0.33161;0.119203;0;0.268405;0.373439;0.358159;0.14973;0.510535;0.339735;0.552795;0.32441;0.122805;0.264856;0.21281;0.198525;0.11001;0.48784;0.056955
0.075419;0.33633;0.134668;0.268703;0.08415;0.5;0.976014;0.16702;0.432621;0.40036;0.14939;0.171745;0.678865;0.33464;0.160175;0.505185;0.163303;0.478866;0.306115;0.174145;0.29995;0.379215
0.268703;0.329135;0.16702;0.432621;0.160175;0.119203;0;0.163303;0.478866;0.35783;0.171445;0.251265;0.57729;0.304375;0.306115;0.38951;0.162333;0.554726;0.379215;0.181885;0.30061;0.477735
0.432621;0.33464;0.163303;0.478866;0.306115;0.268941;0;0.162333;0.554726;0.282941;0.174145;0.423855;0.402;0.29995;0.379215;0.320835;0.165015;0.638838;0.477735;0.182815;0.29999;0.61531
0.478866;0.304375;0.162333;0.554726;0.379215;0.5;0;0.165015;0.638838;0.196149;0.181885;0.601345;0.216775;0.30061;0.477735;0.221655;0.165118;0.772105;0.61531;0.182125;0.293595;0.6517
0.554726;0.29995;0.165015;0.638838;0.477735;0.268941;0;0.165118;0.772105;0.062777;0.182815;0.726105;0.09108;0.29999;0.61531;0.0847;0.163001;0.794721;0.6517;0.181555;0.24602;0.64613
0.638838;0.30061;0.165118;0.772105;0.61531;0.5;0;0.163001;0.794721;0.042276;0.182125;0.75903;0.05884;0.293595;0.6517;0.054705;0.147314;0.701928;0.64613;0.177455;0.244545;0.60125
0.772105;0.29999;0.163001;0.794721;0.6517;0.982014;0;0.147314;0.701928;0.150756;0.181555;0.718695;0.099745;0.24602;0.64613;0.10785;0.149118;0.632085;0.60125;0.16804;0.12363;0.512945
0.794721;0.293595;0.147314;0.701928;0.64613;0.5;0;0.149118;0.632085;0.218799;0.177455;0.6506;0.171945;0.244545;0.60125;0.15421;0.115354;0.508349;0.512945;0.096545;0.04381;0.110115
0.701928;0.24602;0.149118;0.632085;0.60125;0.982014;0;0.115354;0.508349;0.376297;0.16804;0.48107;0.35089;0.12363;0.512945;0.363425;0.070178;0.112402;0.110115;0;0;0
0.007272;0.147985;0.134302;0.013124;0.022915;0.731059;0.963771;0.181377;0.016925;0.801699;0.167775;0.026965;0.80526;0.374755;0.02221;0.60304;0.233115;0.012332;0.01381;0.276805;0.617895;0.00969
0.002356;0.94317;0.872384;0.00263;0.005535;0.002473;0;0.487121;0.006569;0.506309;0.831585;0.001145;0.16727;0.628115;0.0169;0.35498;0.58298;0.003255;0.0052;0.38921;0.4108;0.00115
0.323931;0.00121;0.079323;0.408448;0.590075;0.982014;0.863303;0.073819;0.903539;0.022644;0.22018;0.767645;0.01218;0.001205;0.997195;0.0016;0.073449;0.905625;0.998335;0.20652;0.00066;0.998185
0.408448;0.00122;0.073819;0.903539;0.997195;0.982014;0.590492;0.073449;0.905625;0.020927;0.219095;0.772555;0.00835;0.00118;0.998335;0.000485;0.069084;0.901119;0.998185;0.20752;0.00062;0.9915
0.903539;0.001205;0.073449;0.905625;0.998335;0.982014;0.53743;0.069084;0.901119;0.029799;0.20652;0.75918;0.0343;0.00066;0.998185;0.00116;0.069404;0.882532;0.9915;0.2028;0.000595;0.3302
0.905625;0.00118;0.069084;0.901119;0.998185;0.017986;0;0.069404;0.882532;0.048063;0.20752;0.690115;0.102365;0.00062;0.9915;0.007875;0.067822;0.239705;0.3302;0.026295;0.00087;0.18147
0.239705;0.000595;0.009079;0.114279;0.18147;0.000911;0.040195;0.014765;0.381576;0.603657;0.02538;0.144105;0.83051;0.001575;0.069035;0.92939;0.018693;0.168116;0.25333;0.053385;0.005235;0.443395
0.243523;0.005235;0.03053;0.450328;0.809155;0.731059;0.60659;0.031525;0.767024;0.20145;0.056355;0.566095;0.37755;0.02088;0.804525;0.174595;0.031348;0.215145;0.47567;0.6854;0.199265;0.176545
0.450328;0.00817;0.031525;0.767024;0.804525;0.000335;0;0.031348;0.215145;0.753507;0.052715;0.00412;0.94317;0.023915;0.47567;0.50041;0.436929;0.093897;0.176545;0.71705;0.57;0.01179
0.023184;0.76356;0.253804;0.037863;0.00155;0.006693;0.216531;0.230198;0.076535;0.693269;0.0537;0.0015;0.9448;0.539185;0.014415;0.446405;0.203502;0.264082;0.026925;0.073625;0.68837;0.003265
0.037863;0.577295;0.230198;0.076535;0.014415;0.5;0.275878;0.203502;0.264082;0.532418;0.061525;0.00279;0.93569;0.50012;0.026925;0.472955;0.296627;0.008366;0.003265;0.110415;0.871655;0.00024
0.004486;0.00001;0.011028;0.064004;0.03695;0.047426;0.343666;0.013581;0.561416;0.425003;0.000005;0.179595;0.8204;0;0.68652;0.31348;0.015574;0.968942;0.994335;0.00004;0.00001;0.99976
0.064004;0;0.013581;0.561416;0.68652;0.731059;0;0.015574;0.968942;0.015486;0.00004;0.978415;0.02155;0.00001;0.994335;0.005655;0.01673;0.980562;0.99976;0.00004;0.00001;0.99957
0.561416;0;0.015574;0.968942;0.994335;0.982014;0.761514;0.01673;0.980562;0.002708;0.00004;0.99704;0.00292;0.00001;0.99976;0.00023;0.019454;0.979457;0.99957;0.00002;0.00001;0.929785
0.968942;0.00001;0.01673;0.980562;0.99976;0.952574;0.365632;0.019454;0.979457;0.001089;0.00004;0.998405;0.00155;0.00001;0.99957;0.000425;0.017822;0.93131;0.929785;0.000005;0.000465;0.692255
0.980562;0.00001;0.019454;0.979457;0.99957;0.268941;0.699517;0.017822;0.93131;0.050869;0.00002;0.99017;0.009805;0.00001;0.929785;0.070215;0.015362;0.553587;0.692255;0.000025;0.01079;0.146955
0.979457;0.00001;0.017822;0.93131;0.929785;0.002473;0.260186;0.015362;0.553587;0.431048;0.000005;0.486165;0.513825;0.000465;0.692255;0.307275;0.017397;0.175649;0.146955;0.00018;0.05644;0.02807
0.107791;0.38941;0.360409;0.094321;0.120315;0.017986;0.579812;0.282293;0.32822;0.389485;0.200125;0.054475;0.745395;0.398155;0.395415;0.20643;0.225211;0.490728;0.499605;0.08699;0.379415;0.55187
0.094321;0.3975;0.282293;0.32822;0.395415;0.993307;0.595801;0.225211;0.490728;0.284059;0.12206;0.230095;0.64784;0.392625;0.499605;0.10777;0.205975;0.563051;0.55187;0.08514;0.354385;0.511085
0.32822;0.398155;0.225211;0.490728;0.499605;0.952574;0.332922;0.205975;0.563051;0.230975;0.08699;0.41396;0.49905;0.379415;0.55187;0.068715;0.1634;0.433684;0.511085;0.47228;0.34851;0.374965
0.490728;0.392625;0.205975;0.563051;0.55187;0.017986;0.527223;0.1634;0.433684;0.402916;0.08514;0.40595;0.50891;0.354385;0.511085;0.13453;0.456107;0.285092;0.374965;0.455275;0.34145;0.28912
0.033213;0.469805;0.526536;0.150858;0.22959;0.982014;0.598688;0.098255;0.838856;0.062889;0.28257;0.61279;0.104645;0.00239;0.97829;0.019315;0.098046;0.889193;0.99112;0.277465;0.001315;0.996565
0.150858;0.46981;0.098255;0.838856;0.97829;0.731059;0.394365;0.098046;0.889193;0.012762;0.2797;0.706605;0.013695;0.00208;0.99112;0.0068;0.096238;0.890941;0.996565;0.276265;0.0003;0.91948
0.838856;0.00239;0.098046;0.889193;0.99112;0.017986;0.686111;0.096238;0.890941;0.012821;0.277465;0.70809;0.014445;0.001315;0.996565;0.00212;0.0955;0.866052;0.91948;0.26696;0.00061;0.45372
0.889193;0.00208;0.096238;0.890941;0.996565;0.999089;0.444233;0.0955;0.866052;0.038445;0.276265;0.706325;0.017405;0.0003;0.91948;0.080215;0.0925;0.590819;0.45372;0.062765;0.009245;0.31457
0.890941;0.001315;0.0955;0.866052;0.91948;0.002473;0.49525;0.0925;0.590819;0.316681;0.26696;0.41425;0.31879;0.00061;0.45372;0.54567;0.027319;0.123275;0.31457;0.033165;0.036845;0.01395
0.240744;0.00019;0.009187;0.594416;0.516895;0.880797;0.703913;0.009794;0.955948;0.03426;0.013205;0.949855;0.03694;0.00112;0.990655;0.00823;0.012711;0.959844;0.994915;0.008995;0.001515;0.995435
0.594416;0.0003;0.009794;0.955948;0.990655;0.952574;0.790344;0.012711;0.959844;0.027449;0.01517;0.95863;0.02621;0.00151;0.994915;0.003575;0.010517;0.927037;0.995435;0.0085;0.001485;0.779545
0.955948;0.00112;0.012711;0.959844;0.994915;0.000335;0.50575;0.010517;0.927037;0.062446;0.008995;0.96307;0.027935;0.001515;0.995435;0.00305;0.007897;0.812989;0.779545;0.00562;0.016965;0.349095
0.959844;0.00151;0.010517;0.927037;0.995435;0.997527;0.610639;0.007897;0.812989;0.179113;0.0085;0.93467;0.05683;0.001485;0.779545;0.218965;0.015394;0.155837;0.349095;0.00065;0.03528;0.16833
0.077914;0.009575;0.020339;0.218886;0.444115;0.731059;0.167842;0.021978;0.884858;0.093164;0.006755;0.839665;0.15358;0.0087;0.959875;0.031425;0.018916;0.898364;0.97293;0.007105;0.0035;0.965875
0.218886;0.009035;0.021978;0.884858;0.959875;0.982014;0.06037;0.018916;0.898364;0.082718;0.007855;0.83419;0.15795;0.00363;0.97293;0.02344;0.023723;0.841787;0.965875;0.009;0.004505;0.00242
0.884858;0.0087;0.018916;0.898364;0.97293;0.982014;0.765846;0.023723;0.841787;0.134486;0.007105;0.82348;0.16941;0.0035;0.965875;0.03062;0.027247;0.052629;0.00242;0.00919;0.006095;0.00003
0.004944;0.100865;0.024145;0.007634;0.007485;0.006693;0.551319;0.046341;0.151277;0.802382;0.03534;0.02668;0.93798;0.00609;0.07053;0.92338;0.086738;0.196522;0.250195;0.16302;0.00652;0.13455
0.015566;0.09013;0.004389;0.480567;0.35378;0.982014;0.484005;0.004502;0.979457;0.016039;0.00294;0.97179;0.025265;0.000625;0.978285;0.02109;0.005214;0.9933;0.99777;0.00297;0.001305;0.983505
0.480567;0.00052;0.004502;0.979457;0.978285;0.880797;0.172359;0.005214;0.9933;0.001487;0.00292;0.993815;0.003265;0.00119;0.99777;0.00104;0.004616;0.98871;0.983505;0.002665;0.00132;0.79871
0.979457;0.000625;0.005214;0.9933;0.99777;0.5;0.34299;0.004616;0.98871;0.006674;0.00297;0.992355;0.004675;0.001305;0.983505;0.01519;0.001798;0.905601;0.79871;0.00107;0.01199;0.239855
0.9933;0.00119;0.004616;0.98871;0.983505;0.952574;0.647713;0.001798;0.905601;0.092603;0.002665;0.984255;0.013085;0.00132;0.79871;0.19997;0.004971;0.228885;0.239855;0.00185;0.037485;0.00845
0.013341;0.122755;0.030508;0.07802;0.194175;0.047426;0.21032;0.029174;0.574614;0.396212;0.002405;0.59517;0.40243;0.071635;0.45088;0.47748;0.026203;0.89989;0.75755;0.00216;0.073185;0.85247
0.07802;0.0772;0.029174;0.574614;0.45088;0.047426;0.064888;0.026203;0.89989;0.073907;0.002355;0.95994;0.037705;0.076105;0.75755;0.166345;0.025164;0.929438;0.85247;0.002325;0.05591;0.632825
0.574614;0.071635;0.026203;0.89989;0.75755;0.731059;0.04319;0.025164;0.929438;0.0454;0.00216;0.9674;0.03044;0.073185;0.85247;0.07435;0.019463;0.616041;0.632825;0.011645;0.05171;0.227105
0.042094;0.002055;0.005219;0.473903;0.734625;0.997527;0.484005;0.002131;0.965516;0.032355;0.00068;0.974585;0.02474;0;0.951055;0.048945;0.00212;0.989401;0.99982;0.00053;0;0.999605
0.473903;0.00075;0.002131;0.965516;0.951055;0.952574;0.169103;0.00212;0.989401;0.008477;0.00065;0.996385;0.002965;0;0.99982;0.000175;0.002081;0.98797;0.999605;0.00047;0;0.996755
0.965516;0;0.00212;0.989401;0.99982;0.047426;0.843169;0.002081;0.98797;0.009951;0.00053;0.99298;0.006495;0;0.999605;0.000395;0.004674;0.984851;0.996755;0.00035;0.00001;0.535795
0.989401;0;0.002081;0.98797;0.999605;0.952574;0.024602;0.004674;0.984851;0.010475;0.00047;0.97303;0.0265;0;0.996755;0.003245;0.004609;0.656645;0.535795;0.000355;0.000045;0.03902
0.98797;0;0.004674;0.984851;0.996755;0.268941;0.419945;0.004609;0.656645;0.338749;0.00035;0.46422;0.535435;0.00001;0.535795;0.4642;0.003507;0.343683;0.03902;0.00176;0.01635;0.01423
0.984851;0;0.004609;0.656645;0.535795;0.002473;0.039658;0.003507;0.343683;0.652811;0.000355;0.155335;0.84431;0.000045;0.03902;0.960935;0.183935;0.034452;0.01423;0.035615;0.78056;0.000785
0.014009;0.00047;0.009229;0.040883;0.026625;0.952574;0.306613;0.011681;0.598624;0.389694;0.012395;0.585185;0.40242;0.00001;0.696755;0.303235;0.008672;0.895889;0.933615;0.01287;0.00001;0.99989
0.040883;0;0.011681;0.598624;0.696755;0.982014;0.093215;0.008672;0.895889;0.095439;0.01273;0.774725;0.212545;0.00001;0.933615;0.066375;0.008736;0.983715;0.99989;0.013095;0.00001;0.999865
0.598624;0.00001;0.008672;0.895889;0.933615;0.731059;0.275279;0.008736;0.983715;0.007549;0.01287;0.97109;0.01604;0.00001;0.99989;0.0001;0.010853;0.981604;0.999865;0.01246;0.00001;0.9977
0.895889;0.00001;0.008736;0.983715;0.99989;0.993307;0.152775;0.010853;0.981604;0.007543;0.013095;0.96446;0.022445;0.00001;0.999865;0.000125;0.010275;0.978682;0.9977;0.01166;0.00001;0.970805
0.983715;0.00001;0.010853;0.981604;0.999865;0.997527;0.015413;0.010275;0.978682;0.011043;0.01246;0.959105;0.028435;0.00001;0.9977;0.00229;0.010127;0.960651;0.970805;0.008425;0.00004;0.77845
0.981604;0.00001;0.010275;0.978682;0.9977;0.268941;0.039925;0.010127;0.960651;0.029226;0.01166;0.93147;0.056875;0.00001;0.970805;0.02919;0.006985;0.835344;0.77845;0.053325;0.00772;0.082345
0.978682;0.00001;0.010127;0.960651;0.970805;0.017986;0.016367;0.006985;0.835344;0.157669;0.008425;0.75382;0.237755;0.00004;0.77845;0.221505;0.024582;0.139714;0.082345;0.066775;0.018615;0.015495
0.139714;0.00772;0.030512;0.01276;0.015495;0.268941;0.618456;0.049878;0.135499;0.814621;0.07008;0.156935;0.77298;0.05353;0.04505;0.90142;0.114052;0.296745;0.05371;0.293185;0.294375;0.074375
0.176029;0.68548;0.401316;0.183996;0.07982;0.119203;0.296713;0.193638;0.177673;0.628689;0.002005;0.162875;0.83512;0.48434;0.08495;0.43071;0.114346;0.071748;0.05965;0.013005;0.366555;0.034925
0.021529;0.00028;0.050754;0.078146;0.009675;0.880797;0.594355;0.048512;0.360439;0.591047;0.007765;0.334275;0.65796;0.00147;0.059045;0.93948;0.048455;0.804422;0.78814;0.006715;0.00103;0.974055
0.078146;0.00136;0.048512;0.360439;0.059045;0.006693;0.706615;0.048455;0.804422;0.147125;0.006485;0.76785;0.22567;0.001875;0.78814;0.209985;0.044872;0.892974;0.974055;0.003485;0.00058;0.98263
0.360439;0.00147;0.048455;0.804422;0.78814;0.5;0.451899;0.044872;0.892974;0.062152;0.006715;0.84881;0.14447;0.00103;0.974055;0.024915;0.039232;0.855282;0.98263;0.004985;0.000635;0.811015
0.804422;0.001875;0.044872;0.892974;0.974055;0.119203;0.20751;0.039232;0.855282;0.105485;0.003485;0.813185;0.183325;0.00058;0.98263;0.01679;0.034006;0.780605;0.811015;0.006285;0.000695;0.018085
0.892974;0.00103;0.039232;0.855282;0.98263;0.997527;0.653849;0.034006;0.780605;0.185391;0.004985;0.80955;0.18547;0.000635;0.811015;0.18835;0.018621;0.145666;0.018085;0.01499;0.006995;0.00853
0.780605;0.000635;0.018621;0.145666;0.018085;0.047426;0.123035;0.033272;0.055623;0.911104;0.01499;0.004405;0.9806;0.006995;0.00853;0.984475;0.926382;0.000096;0.00006;0.862105;0.97443;0
0.007829;0.01228;0.022858;0.142897;0.07904;0.993307;0.715042;0.017755;0.872298;0.109947;0.001625;0.864975;0.1334;0.00005;0.92823;0.07172;0.017965;0.904778;0.994825;0.00183;0.00005;0.998615
0.142897;0.00003;0.017755;0.872298;0.92823;0.993307;0.153164;0.017965;0.904778;0.077258;0.00187;0.89422;0.10391;0.000055;0.994825;0.005125;0.018068;0.894045;0.998615;0.001555;0.000045;0.99505
0.872298;0.00005;0.017965;0.904778;0.994825;0.952574;0.14944;0.018068;0.894045;0.087885;0.00183;0.85784;0.140325;0.00005;0.998615;0.001335;0.017802;0.85711;0.99505;0.001475;0.000055;0.96293
0.904778;0.000055;0.018068;0.894045;0.998615;0.006693;0.076775;0.017802;0.85711;0.125088;0.001555;0.75029;0.24816;0.000045;0.99505;0.0049;0.026358;0.808248;0.96293;0.001195;0.00005;0.935395
0.894045;0.00005;0.017802;0.85711;0.99505;0.982014;0.615147;0.026358;0.808248;0.165392;0.001475;0.63985;0.35867;0.000055;0.96293;0.037015;0.027138;0.788297;0.935395;0.001285;0.00011;0.818915
0.85711;0.000045;0.026358;0.808248;0.96293;0.731059;0.112645;0.027138;0.788297;0.184565;0.001195;0.608475;0.39033;0.00005;0.935395;0.064555;0.028716;0.448114;0.818915;0.002385;0.000315;0.11658
0.116165;0.017385;0.03848;0.224348;0.274045;0.047426;0.338721;0.040265;0.428855;0.53088;0.078305;0.207855;0.71384;0.002225;0.649855;0.34792;0.073885;0.716853;0.931955;0.14737;0.00159;0.9665
0.224348;0.001155;0.040265;0.428855;0.649855;0.731059;0;0.073885;0.716853;0.209262;0.14609;0.50175;0.35216;0.00168;0.931955;0.066365;0.07448;0.773778;0.9665;0.14594;0.00159;0.987575
0.428855;0.002225;0.073885;0.716853;0.931955;0.047426;0.498;0.07448;0.773778;0.151743;0.14737;0.581055;0.27158;0.00159;0.9665;0.031905;0.073765;0.804605;0.987575;0.14915;0.00156;0.99152
0.716853;0.00168;0.07448;0.773778;0.9665;0.119203;0;0.073765;0.804605;0.12163;0.14594;0.621635;0.23243;0.00159;0.987575;0.01083;0.075355;0.813645;0.99152;0.147985;0.001585;0.983875
0.773778;0.00159;0.073765;0.804605;0.987575;0.119203;0.636453;0.075355;0.813645;0.110997;0.14915;0.63577;0.21508;0.00156;0.99152;0.006915;0.074785;0.8062;0.983875;0.14784;0.001565;0.962955
0.804605;0.00159;0.075355;0.813645;0.99152;0.731059;0.548596;0.074785;0.8062;0.119013;0.147985;0.628525;0.223485;0.001585;0.983875;0.01454;0.074703;0.752547;0.962955;0.137805;0.001155;0.918225
0.813645;0.00156;0.074785;0.8062;0.983875;0.047426;0;0.074703;0.752547;0.172745;0.14784;0.54214;0.31002;0.001565;0.962955;0.03547;0.06948;0.684555;0.918225;0.130225;0.001145;0.51806
0.8062;0.001585;0.074703;0.752547;0.962955;0.047426;0.14381;0.06948;0.684555;0.245965;0.137805;0.450885;0.411305;0.001155;0.918225;0.080625;0.065685;0.404043;0.51806;0.128785;0.00185;0.042
0.752547;0.001565;0.06948;0.684555;0.918225;0.119203;0.129431;0.065685;0.404043;0.530273;0.130225;0.290025;0.57975;0.001145;0.51806;0.480795;0.150575;0.063305;0.042;0.12209;0.00157;0.0188
0.060093;0.00144;0.078702;0.152268;0.08523;0.119203;0.140035;0.088707;0.549999;0.361296;0.265105;0.285135;0.44976;0.00032;0.55573;0.443955;0.096459;0.764338;0.965885;0.31945;0.00036;0.9953
0.152268;0.00061;0.088707;0.549999;0.55573;0.5;0;0.096459;0.764338;0.139203;0.28843;0.508745;0.202825;0.000345;0.965885;0.03377;0.11368;0.834506;0.9953;0.327975;0.00037;0.999085
0.549999;0.00032;0.096459;0.764338;0.965885;0.731059;0;0.11368;0.834506;0.051818;0.31945;0.53838;0.14217;0.00036;0.9953;0.00435;0.116499;0.841604;0.999085;0.328085;0.000375;0.997895
0.764338;0.000345;0.11368;0.834506;0.9953;0.997527;0.65701;0.116499;0.841604;0.041897;0.327975;0.54747;0.12455;0.00037;0.999085;0.00055;0.116537;0.841211;0.997895;0.335305;0.0004;0.989895
0.834506;0.00036;0.116499;0.841604;0.999085;0.119203;0.391026;0.116537;0.841211;0.04225;0.328085;0.54748;0.124435;0.000375;0.997895;0.001725;0.118952;0.832061;0.989895;0.335335;0.002955;0.93527
0.841604;0.00037;0.116537;0.841211;0.997895;0.982014;0.13447;0.118952;0.832061;0.048987;0.335305;0.52803;0.13666;0.0004;0.989895;0.00971;0.119826;0.784089;0.93527;0.324525;0.002985;0.67902
0.841211;0.000375;0.118952;0.832061;0.989895;0.731059;0.04936;0.119826;0.784089;0.096085;0.335335;0.438775;0.22589;0.002955;0.93527;0.061775;0.116232;0.625686;0.67902;0.30304;0.00966;0.38157
0.832061;0.0004;0.119826;0.784089;0.93527;0.047426;0.603005;0.116232;0.625686;0.258084;0.324525;0.219815;0.455665;0.002985;0.67902;0.317995;0.111328;0.475871;0.38157;0.101425;0.012375;0.26555
0.784089;0.002955;0.116232;0.625686;0.67902;0.268941;0.919384;0.111328;0.475871;0.412801;0.30304;0.06792;0.62904;0.00966;0.38157;0.60877;0.044468;0.413109;0.26555;0.061365;0.016445;0.19436
0.625686;0.002985;0.111328;0.475871;0.38157;0.5;0.229347;0.044468;0.413109;0.542424;0.101425;0.021545;0.877035;0.012375;0.26555;0.722075;0.030979;0.126875;0.19436;0.065815;0.05808;0.212275
0.097891;0.006055;0.015399;0.270954;0.408025;0.017986;0.643136;0.038785;0.431346;0.52987;0.113775;0.01419;0.872035;0.00197;0.67323;0.324805;0.044187;0.754108;0.8614;0.14031;0.001935;0.986005
0.270954;0.003295;0.038785;0.431346;0.67323;0.047426;0.066546;0.044187;0.754108;0.201707;0.130035;0.57857;0.291405;0.00193;0.8614;0.136665;0.047613;0.853348;0.986005;0.14102;0.00193;0.99528
0.431346;0.00197;0.044187;0.754108;0.8614;0.997527;0.65971;0.047613;0.853348;0.099041;0.14031;0.751035;0.108655;0.001935;0.986005;0.012065;0.047848;0.87919;0.99528;0.14037;0.001875;0.996685
0.754108;0.00193;0.047613;0.853348;0.986005;0.119203;0.340964;0.047848;0.87919;0.072963;0.14102;0.819285;0.039695;0.00193;0.99528;0.00279;0.047614;0.882708;0.996685;0.137425;0.00185;0.991255
0.853348;0.001935;0.047848;0.87919;0.99528;0.993307;0.514746;0.047614;0.882708;0.069676;0.14037;0.82968;0.029945;0.001875;0.996685;0.00144;0.052807;0.931313;0.991255;0.13122;0.001795;0.79284
0.87919;0.00193;0.047614;0.882708;0.996685;0.119203;0.097176;0.052807;0.931313;0.01588;0.137425;0.82243;0.040145;0.00185;0.991255;0.006895;0.050863;0.856942;0.79284;0.11338;0.00155;0.281755
0.882708;0.001875;0.052807;0.931313;0.991255;0.731059;0.102569;0.050863;0.856942;0.092197;0.13122;0.79817;0.070615;0.001795;0.79284;0.205365;0.052922;0.49468;0.281755;0.13241;0.00213;0.106875
0.931313;0.00185;0.050863;0.856942;0.79284;0.5;0.219943;0.052922;0.49468;0.452397;0.11338;0.459345;0.427275;0.00155;0.281755;0.716695;0.072946;0.246973;0.106875;0.24225;0.006445;0.017445
0.422303;0.0467;0.242832;0.533152;0.672385;0.268941;0.353658;0.273956;0.489737;0.23631;0.68913;0.115855;0.195015;0.04516;0.686095;0.268755;0.275656;0.45225;0.67896;0.670705;0.06558;0.66292
0.533152;0.0542;0.273956;0.489737;0.686095;0.119203;0.121425;0.275656;0.45225;0.272094;0.68266;0.113375;0.203965;0.0471;0.67896;0.27394;0.277059;0.338904;0.66292;0.680625;0.076905;0.655
0.071286;0.0026;0.043449;0.085994;0.12254;0.982014;0.402033;0.046885;0.289706;0.663407;0.11354;0.05196;0.834495;0.00619;0.22014;0.77367;0.048872;0.398986;0.49162;0.10709;0.008075;0.47003
0.085994;0.00545;0.046885;0.289706;0.22014;0.119203;0;0.048872;0.398986;0.552142;0.117175;0.106975;0.77585;0.008585;0.49162;0.499795;0.045326;0.401411;0.47003;0.162745;0.00806;0.634195
0.289706;0.00619;0.048872;0.398986;0.49162;0.017986;0.203915;0.045326;0.401411;0.553261;0.10709;0.13501;0.7579;0.008075;0.47003;0.52189;0.06388;0.474307;0.634195;0.160665;0.006395;0.80011
0.398986;0.008585;0.045326;0.401411;0.47003;0.017986;0.151357;0.06388;0.474307;0.461813;0.162745;0.1743;0.662955;0.00806;0.634195;0.357745;0.05588;0.613278;0.80011;0.158275;0.00424;0.93948
0.401411;0.008075;0.06388;0.474307;0.634195;0.268941;0.290285;0.05588;0.613278;0.33084;0.160665;0.21489;0.62444;0.006395;0.80011;0.193495;0.054364;0.717171;0.93948;0.15457;0.002805;0.95151
0.474307;0.00806;0.05588;0.613278;0.80011;0.268941;0.036299;0.054364;0.717171;0.228465;0.158275;0.233305;0.608415;0.00424;0.93948;0.056285;0.052651;0.723596;0.95151;0.132385;0.002775;0.911885
0.613278;0.006395;0.054364;0.717171;0.93948;0.731059;0.16645;0.052651;0.723596;0.223751;0.15457;0.24055;0.604875;0.002805;0.95151;0.045685;0.045245;0.707611;0.911885;0.160115;0.002715;0.770025
0.717171;0.00424;0.052651;0.723596;0.95151;0.119203;0.13308;0.045245;0.707611;0.247145;0.132385;0.22493;0.642685;0.002775;0.911885;0.085345;0.054469;0.618433;0.770025;0.16997;0.0047;0.47753
0.051053;0.101195;0.150668;0.158469;0.2019;0.5;0.450909;0.145796;0.630752;0.22345;0.434895;0.1371;0.428005;0.00177;0.911335;0.08689;0.14953;0.821307;0.98787;0.45316;0.00173;0.997335
0.158469;0.028205;0.145796;0.630752;0.911335;0.731059;0.136933;0.14953;0.821307;0.029162;0.446275;0.497205;0.05652;0.00174;0.98787;0.010385;0.151821;0.828652;0.997335;0.460125;0.001725;0.99794
0.630752;0.00177;0.14953;0.821307;0.98787;0.982014;0.269729;0.151821;0.828652;0.019525;0.45316;0.509765;0.037075;0.00173;0.997335;0.00093;0.154141;0.826954;0.99794;0.45835;0.00173;0.99302
0.821307;0.00174;0.151821;0.828652;0.997335;0.5;0.108515;0.154141;0.826954;0.018903;0.460125;0.50403;0.035845;0.001725;0.99794;0.00033;0.153551;0.823262;0.99302;0.458205;0.001755;0.974995
0.828652;0.00173;0.154141;0.826954;0.99794;0.268941;0.222354;0.153551;0.823262;0.023185;0.45835;0.497875;0.04377;0.00173;0.99302;0.00525;0.153511;0.788578;0.974995;0.446895;0.00175;0.874255
0.826954;0.001725;0.153551;0.823262;0.99302;0.993307;0.333366;0.153511;0.788578;0.05791;0.458205;0.433535;0.108255;0.001755;0.974995;0.02325;0.149742;0.735143;0.874255;0.431925;0.001875;0.667015
0.823262;0.00173;0.153511;0.788578;0.974995;0.119203;0;0.149742;0.735143;0.115115;0.446895;0.35438;0.19873;0.00175;0.874255;0.12399;0.144802;0.593409;0.667015;0.441515;0.00318;0.549985
0.788578;0.001755;0.149742;0.735143;0.874255;0.006693;0.259802;0.144802;0.593409;0.261795;0.431925;0.2198;0.34828;0.001875;0.667015;0.33112;0.196798;0.31799;0.549985;0.441775;0.00472;0.246845
0.113596;0.02837;0.162614;0.095305;0.014125;0.006693;0.361929;0.172438;0.327639;0.499923;0.465305;0.005125;0.52957;0.02582;0.206915;0.767265;0.188168;0.538699;0.742505;0.499855;0.02211;0.87246
0.095305;0.026315;0.172438;0.327639;0.206915;0.880797;0.122389;0.188168;0.538699;0.273133;0.512485;0.023255;0.464265;0.021715;0.742505;0.235775;0.185034;0.588711;0.87246;0.33162;0.021495;0.95142
0.327639;0.02582;0.188168;0.538699;0.742505;0.731059;0.037943;0.185034;0.588711;0.226255;0.499855;0.057875;0.44227;0.02211;0.87246;0.10543;0.135797;0.622053;0.95142;0.329215;0.021505;0.97028
0.538699;0.021715;0.185034;0.588711;0.87246;0.017986;0.663515;0.135797;0.622053;0.242147;0.33162;0.057365;0.61101;0.021495;0.95142;0.02708;0.134969;0.625211;0.97028;0.26987;0.0179;0.94991
0.588711;0.02211;0.135797;0.622053;0.95142;0.047426;0.138715;0.134969;0.625211;0.23982;0.329215;0.07728;0.5935;0.021505;0.97028;0.00822;0.182267;0.563414;0.94991;0.245875;0.016375;0.947805
0.622053;0.021495;0.134969;0.625211;0.97028;0.993307;0.727901;0.182267;0.563414;0.25432;0.26987;0.087805;0.642325;0.0179;0.94991;0.032195;0.187958;0.561418;0.947805;0.2386;0.016385;0.82359
0.625211;0.021505;0.182267;0.563414;0.94991;0.5;0.233438;0.187958;0.561418;0.250623;0.245875;0.08392;0.6702;0.016375;0.947805;0.03582;0.193081;0.516911;0.82359;0.23652;0.016055;0.51205
0.563414;0.0179;0.187958;0.561418;0.947805;0.982014;0.049878;0.193081;0.516911;0.29001;0.2386;0.074615;0.68679;0.016385;0.82359;0.160025;0.19982;0.352548;0.51205;0.372485;0.37679;0.19802
0.561418;0.016375;0.193081;0.516911;0.82359;0.119203;0.218744;0.19982;0.352548;0.447633;0.23652;0.043435;0.72005;0.016055;0.51205;0.471895;0.560497;0.07015;0.19802;0.28225;0.41645;0.18791
0.019828;0.08949;0.215775;0.265667;0.12676;0.880797;0.418971;0.259977;0.371936;0.368087;0.74063;0.00884;0.25053;0.03791;0.427645;0.534445;0.260497;0.420113;0.6075;0.74345;0.018485;0.778055
0.265667;0.05189;0.259977;0.371936;0.427645;0.119203;0.448187;0.260497;0.420113;0.319387;0.753355;0.009505;0.237135;0.02698;0.6075;0.365515;0.254264;0.595104;0.778055;0.78367;0.01662;0.890445
0.371936;0.03791;0.260497;0.420113;0.6075;0.5;0.052501;0.254264;0.595104;0.150628;0.74345;0.008975;0.24757;0.018485;0.778055;0.203455;0.267025;0.633416;0.890445;0.757225;0.006625;0.929185
0.420113;0.02698;0.254264;0.595104;0.778055;0.006693;0;0.267025;0.633416;0.09956;0.78367;0.011375;0.20496;0.01662;0.890445;0.092935;0.254879;0.646856;0.929185;0.742555;0.00333;0.927105
0.595104;0.018485;0.267025;0.633416;0.890445;0.047426;0.092122;0.254879;0.646856;0.098265;0.757225;0.012955;0.22982;0.006625;0.929185;0.06419;0.24889;0.646329;0.927105;0.676945;0.003595;0.69577
0.633416;0.01662;0.254879;0.646856;0.929185;0.047426;0.146665;0.24889;0.646329;0.10478;0.742555;0.013455;0.24399;0.00333;0.927105;0.069565;0.227113;0.305733;0.69577;0.593765;0.0027;0.488335
0.646329;0.00333;0.227113;0.305733;0.69577;0.268941;0;0.199186;0.503665;0.297145;0.593765;0.02485;0.38138;0.0027;0.488335;0.50896;0.131762;0.078209;0.049025;0.32168;0.007045;0.042645
0.035033;0.014175;0.043143;0.233167;0.08489;0.5;0;0.045863;0.647938;0.306202;0.06641;0.847025;0.08657;0.025315;0.44885;0.525835;0.03823;0.715917;0.58478;0.063535;0.01961;0.560555
0.002482;0.00089;0.020875;0.080635;0.06993;0.017986;0.195604;0.053433;0.189943;0.756615;0.09584;0.158975;0.745175;0.011025;0.22091;0.768055;0.058365;0.407607;0.319535;0.120425;0.011065;0.41217
0.080635;0.002115;0.053433;0.189943;0.22091;0.119203;0.753803;0.058365;0.407607;0.534025;0.105485;0.49568;0.39883;0.011245;0.319535;0.66922;0.065745;0.549207;0.41217;0.13447;0.01083;0.905705
0.189943;0.011025;0.058365;0.407607;0.319535;0.119203;0.966431;0.065745;0.549207;0.38505;0.120425;0.686245;0.193335;0.011065;0.41217;0.576765;0.07265;0.848058;0.905705;0.13637;0.01158;0.920875
0.407607;0.011245;0.065745;0.549207;0.41217;0.993307;0;0.07265;0.848058;0.079298;0.13447;0.79041;0.07512;0.01083;0.905705;0.083475;0.073975;0.870518;0.920875;0.136195;0.011565;0.92844
0.549207;0.011065;0.07265;0.848058;0.905705;0.880797;0;0.073975;0.870518;0.055507;0.13637;0.82016;0.04347;0.01158;0.920875;0.067545;0.07388;0.871145;0.92844;0.13414;0.01149;0.92473
0.848058;0.01083;0.073975;0.870518;0.920875;0.017986;0;0.07388;0.871145;0.054977;0.136195;0.81385;0.049955;0.011565;0.92844;0.06;0.072815;0.815647;0.92473;0.00865;0.001595;0.75983
0.870518;0.01158;0.07388;0.871145;0.92844;0.119203;0;0.072815;0.815647;0.11154;0.13414;0.706565;0.159295;0.01149;0.92473;0.063785;0.005122;0.591782;0.75983;0.001765;0.000245;0.35225
0.64433;0.02993;0.034845;0.927845;0.942265;0.982014;0.878361;0.034937;0.95042;0.014645;0.036955;0.944375;0.01867;0.03292;0.956465;0.01062;0.03499;0.952723;0.95948;0.037215;0.03137;0.959705
0.927845;0.03281;0.034937;0.95042;0.956465;0.268941;0;0.03499;0.952723;0.012287;0.037235;0.945965;0.0168;0.032745;0.95948;0.007775;0.034293;0.95237;0.959705;0.02426;0.030195;0.955835
0.95042;0.03292;0.03499;0.952723;0.95948;0.952574;0;0.034293;0.95237;0.013337;0.037215;0.945035;0.01775;0.03137;0.959705;0.008925;0.027227;0.916058;0.955835;0.021095;0.0298;0.59007
0.952723;0.032745;0.034293;0.95237;0.959705;0.880797;0;0.027227;0.916058;0.056715;0.02426;0.87628;0.09946;0.030195;0.955835;0.01397;0.025447;0.696613;0.59007;0.01859;0.03282;0.580155
0.95237;0.03137;0.027227;0.916058;0.955835;0.047426;0;0.025447;0.696613;0.27794;0.021095;0.803155;0.17575;0.0298;0.59007;0.38013;0.025705;0.681028;0.580155;0.015215;0.04437;0.499615
0.916058;0.030195;0.025447;0.696613;0.59007;0.268941;0;0.025705;0.681028;0.293265;0.01859;0.7819;0.19951;0.03282;0.580155;0.38702;0.029792;0.552175;0.499615;0.01509;0.04618;0.51508
0.696613;0.0298;0.025705;0.681028;0.580155;0.119203;0.859483;0.029792;0.552175;0.418038;0.015215;0.604735;0.380055;0.04437;0.499615;0.45602;0.030635;0.549233;0.51508;0.008275;0.043585;0.534745
0.681028;0.03282;0.029792;0.552175;0.499615;0.268941;0;0.030635;0.549233;0.420133;0.01509;0.583385;0.401525;0.04618;0.51508;0.43874;0.02593;0.374918;0.534745;0.00138;0.004705;0.023925
0.552175;0.04437;0.030635;0.549233;0.51508;0.017986;0;0.02593;0.374918;0.59915;0.008275;0.21509;0.77663;0.043585;0.534745;0.42167;0.003042;0.023242;0.023925;0.00094;0.001175;0.000265
0.549233;0.04618;0.02593;0.374918;0.534745;0.047426;0;0.003042;0.023242;0.973713;0.00138;0.02256;0.976055;0.004705;0.023925;0.97137;0.001058;0.00251;0.000265;0.00138;0.00041;0.005515

From nicholdav at gmail.com  Mon Aug 21 10:38:57 2017
From: nicholdav at gmail.com (David Nicholson)
Date: Mon, 21 Aug 2017 10:38:57 -0400
Subject: [scikit-learn] How can i write the birch prediction results to
 the file
In-Reply-To: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>
References: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>
Message-ID: <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>

Hi Sema,

You can save using pickle from the Python standard library, or using the
joblib library which is a dependency of sklearn (so you have it already).

The sklearn docs show examples of saving models but it will work for your
predict results too:
http://scikit-learn.org/stable/modules/model_persistence.html

You'd just do something like:
import joblib
...
# your code here
...
birch_predict = brc.predict(X)
filename = 'predictions'
joblib.dump(birch_predict, filename)

And you can get the values back into memory with joblib.load

Hth
--David (list lurker)

On Aug 21, 2017 10:13, "Sema Atasever" <s.atasever at gmail.com> wrote:

Dear scikit-learn developers,

I have a text file where the columns represent the 22 features and the rows
represent the amino asid . (you can see in the attachment)


I want to apply hierarchical clustering to this database usign
*sklearn.cluster.Birch
algorithm.*

There are too many prediction results and it is not possible to see them on
the screen.
How can i write the birch prediction results to the file?

I would appreciate if you could advise on some methods.
Thanks.

*Birch Codes:*
from sklearn.cluster import Birch
import numpy as np

X=np.loadtxt(open("C:\class1.txt", "rb"), delimiter=";")

brc = Birch(branching_factor=50, n_clusters=None,
threshold=0.5,compute_labels=True,copy=True)

brc.fit(X)

centroids = brc.subcluster_centers_

labels = brc.subcluster_labels_
n_clusters = np.unique(labels).size
brc.predict(X)

print("\n brc.predict(X)")
print(brc.predict(X))

_______________________________________________
scikit-learn mailing list
scikit-learn at python.org
https://mail.python.org/mailman/listinfo/scikit-learn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170821/dd480f2e/attachment.html>

From nicholdav at gmail.com  Mon Aug 21 10:41:51 2017
From: nicholdav at gmail.com (David Nicholson)
Date: Mon, 21 Aug 2017 10:41:51 -0400
Subject: [scikit-learn] How can i write the birch prediction results to
 the file
In-Reply-To: <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>
References: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>
 <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>
Message-ID: <CAMabFbX3iAA_TnXUwon255foRTUgmyec-YgkkbmYSPPQxT2NTw@mail.gmail.com>

Ack, should've mentioned you can do:

from sklearn.externals import joblib

since it is a sklearn dependency. That way you won't need to install joblib
separately

On Aug 21, 2017 10:38, "David Nicholson" <nicholdav at gmail.com> wrote:

> Hi Sema,
>
> You can save using pickle from the Python standard library, or using the
> joblib library which is a dependency of sklearn (so you have it already).
>
> The sklearn docs show examples of saving models but it will work for your
> predict results too:
> http://scikit-learn.org/stable/modules/model_persistence.html
>
> You'd just do something like:
> import joblib
> ...
> # your code here
> ...
> birch_predict = brc.predict(X)
> filename = 'predictions'
> joblib.dump(birch_predict, filename)
>
> And you can get the values back into memory with joblib.load
>
> Hth
> --David (list lurker)
>
> On Aug 21, 2017 10:13, "Sema Atasever" <s.atasever at gmail.com> wrote:
>
> Dear scikit-learn developers,
>
> I have a text file where the columns represent the 22 features and the
> rows represent the amino asid . (you can see in the attachment)
>
>
> I want to apply hierarchical clustering to this database usign *sklearn.cluster.Birch
> algorithm.*
>
> There are too many prediction results and it is not possible to see them
> on the screen.
> How can i write the birch prediction results to the file?
>
> I would appreciate if you could advise on some methods.
> Thanks.
>
> *Birch Codes:*
> from sklearn.cluster import Birch
> import numpy as np
>
> X=np.loadtxt(open("C:\class1.txt", "rb"), delimiter=";")
>
> brc = Birch(branching_factor=50, n_clusters=None,
> threshold=0.5,compute_labels=True,copy=True)
>
> brc.fit(X)
>
> centroids = brc.subcluster_centers_
>
> labels = brc.subcluster_labels_
> n_clusters = np.unique(labels).size
> brc.predict(X)
>
> print("\n brc.predict(X)")
> print(brc.predict(X))
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170821/ba3cc581/attachment.html>

From s.atasever at gmail.com  Tue Aug 22 04:24:26 2017
From: s.atasever at gmail.com (Sema Atasever)
Date: Tue, 22 Aug 2017 11:24:26 +0300
Subject: [scikit-learn] How can i write the birch prediction results to
 the file
In-Reply-To: <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>
References: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>
 <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>
Message-ID: <CAAir+CpaQTWo_0PP6p_TTo920AF6RFQN-wx=Ht=uxoXGP6=C-A@mail.gmail.com>

Dear David,

"joblib.dump" produces a file format with npy extension so I can not open
the file with the notepad editor. I can not see the predictions results
inside the file.

Is there another way to save the prediction results in text format?

Thank you for your help.

On Mon, Aug 21, 2017 at 5:38 PM, David Nicholson <nicholdav at gmail.com>
wrote:

> Hi Sema,
>
> You can save using pickle from the Python standard library, or using the
> joblib library which is a dependency of sklearn (so you have it already).
>
> The sklearn docs show examples of saving models but it will work for your
> predict results too:
> http://scikit-learn.org/stable/modules/model_persistence.html
>
> You'd just do something like:
> import joblib
> ...
> # your code here
> ...
> birch_predict = brc.predict(X)
> filename = 'predictions'
> joblib.dump(birch_predict, filename)
>
> And you can get the values back into memory with joblib.load
>
> Hth
> --David (list lurker)
>
> On Aug 21, 2017 10:13, "Sema Atasever" <s.atasever at gmail.com> wrote:
>
> Dear scikit-learn developers,
>
> I have a text file where the columns represent the 22 features and the
> rows represent the amino asid . (you can see in the attachment)
>
>
> I want to apply hierarchical clustering to this database usign *sklearn.cluster.Birch
> algorithm.*
>
> There are too many prediction results and it is not possible to see them
> on the screen.
> How can i write the birch prediction results to the file?
>
> I would appreciate if you could advise on some methods.
> Thanks.
>
> *Birch Codes:*
> from sklearn.cluster import Birch
> import numpy as np
>
> X=np.loadtxt(open("C:\class1.txt", "rb"), delimiter=";")
>
> brc = Birch(branching_factor=50, n_clusters=None,
> threshold=0.5,compute_labels=True,copy=True)
>
> brc.fit(X)
>
> centroids = brc.subcluster_centers_
>
> labels = brc.subcluster_labels_
> n_clusters = np.unique(labels).size
> brc.predict(X)
>
> print("\n brc.predict(X)")
> print(brc.predict(X))
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170822/39df903c/attachment.html>

From rth.yurchak at gmail.com  Tue Aug 22 04:33:21 2017
From: rth.yurchak at gmail.com (Roman Yurchak)
Date: Tue, 22 Aug 2017 11:33:21 +0300
Subject: [scikit-learn] How can i write the birch prediction results to
 the file
In-Reply-To: <CAAir+CpaQTWo_0PP6p_TTo920AF6RFQN-wx=Ht=uxoXGP6=C-A@mail.gmail.com>
References: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>
 <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>
 <CAAir+CpaQTWo_0PP6p_TTo920AF6RFQN-wx=Ht=uxoXGP6=C-A@mail.gmail.com>
Message-ID: <433d5f58-4e2f-661e-6e54-bdd388a4c1ce@gmail.com>

Hello Sema,

On 22/08/17 11:24, Sema Atasever wrote:
 > "joblib.dump" produces a file format with npy extension so I can not 
open the file with the notepad editor. I can not see the predictions 
results inside the file.
>
> Is there another way to save the prediction results in text format?

Prediction results are just an array: you could use numpy.savetxt to 
save them in an ascii text format.

-- 
Roman


From aliozcan at gmail.com  Tue Aug 22 04:37:36 2017
From: aliozcan at gmail.com (Ali Ozcan)
Date: Tue, 22 Aug 2017 10:37:36 +0200
Subject: [scikit-learn] How can i write the birch prediction results to
 the file
In-Reply-To: <433d5f58-4e2f-661e-6e54-bdd388a4c1ce@gmail.com>
References: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>
 <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>
 <CAAir+CpaQTWo_0PP6p_TTo920AF6RFQN-wx=Ht=uxoXGP6=C-A@mail.gmail.com>
 <433d5f58-4e2f-661e-6e54-bdd388a4c1ce@gmail.com>
Message-ID: <CAMetXZ9zAN4kW3BD7WYSOEPuGdn7Nugukz4YD0+0yZzCVVBs8g@mail.gmail.com>

Sema, you can use this

import numpy as np

np.savetxt('birch_predict.csv', birch_predict, delimiter=',')


On Tue, Aug 22, 2017 at 10:33 AM, Roman Yurchak <rth.yurchak at gmail.com>
wrote:

> Hello Sema,
>
> On 22/08/17 11:24, Sema Atasever wrote:
> > "joblib.dump" produces a file format with npy extension so I can not
> open the file with the notepad editor. I can not see the predictions
> results inside the file.
>
>>
>> Is there another way to save the prediction results in text format?
>>
>
> Prediction results are just an array: you could use numpy.savetxt to save
> them in an ascii text format.
>
> --
> Roman
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>



-- 
Ali Ozcan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170822/b583982f/attachment.html>

From s.atasever at gmail.com  Tue Aug 22 04:48:59 2017
From: s.atasever at gmail.com (Sema Atasever)
Date: Tue, 22 Aug 2017 11:48:59 +0300
Subject: [scikit-learn] How can i write the birch prediction results to
 the file
In-Reply-To: <433d5f58-4e2f-661e-6e54-bdd388a4c1ce@gmail.com>
References: <CAAir+CrhB=NvJY4LCeUaJeOrN9WF13mTuVvkjMVwOozV097dFA@mail.gmail.com>
 <CAMabFbWwAvtx4SxdW=r3bD4B3K=g+V4LMT9AWrfOYY9MbhA9gA@mail.gmail.com>
 <CAAir+CpaQTWo_0PP6p_TTo920AF6RFQN-wx=Ht=uxoXGP6=C-A@mail.gmail.com>
 <433d5f58-4e2f-661e-6e54-bdd388a4c1ce@gmail.com>
Message-ID: <CAAir+CorRraFURVpr7Q+nhyw0WKQTojPqyVrN_u-p+f=rf8LGw@mail.gmail.com>

Dear Roman and Ali,

it did worked thanks for all your help.

Regards.

On Tue, Aug 22, 2017 at 11:33 AM, Roman Yurchak <rth.yurchak at gmail.com>
wrote:

> Hello Sema,
>
> On 22/08/17 11:24, Sema Atasever wrote:
> > "joblib.dump" produces a file format with npy extension so I can not
> open the file with the notepad editor. I can not see the predictions
> results inside the file.
>
>>
>> Is there another way to save the prediction results in text format?
>>
>
> Prediction results are just an array: you could use numpy.savetxt to save
> them in an ascii text format.
>
> --
> Roman
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170822/3944149c/attachment-0001.html>

From mcapizzi at email.arizona.edu  Tue Aug 22 14:52:04 2017
From: mcapizzi at email.arizona.edu (Michael Capizzi)
Date: Tue, 22 Aug 2017 11:52:04 -0700
Subject: [scikit-learn] any interest in incorporating a new Transformer?
In-Reply-To: <CAAkaFLW01e_E+wvavdE+iPhnkYwvn+n3S8R+YrMipQJ8m1sowg@mail.gmail.com>
References: <CAAX6ikFJciCVrR9RfRSAbRcuGWbDOLBtAgkJJui3EDTdDUCLLw@mail.gmail.com>
 <CAAkaFLWTRKbKJt7PDNab3=vbsaq4hjn1Z92SefvouGNU9B+rNQ@mail.gmail.com>
 <CAAX6ikESc2scpJhUm1pXMLvr5Jm-OtgEbFMjZW7oroNLq3H9wA@mail.gmail.com>
 <CAAkaFLW01e_E+wvavdE+iPhnkYwvn+n3S8R+YrMipQJ8m1sowg@mail.gmail.com>
Message-ID: <CAAX6ikEend3DdZmUR8SVfQLxnzv6bksFZZgLPehnm-ud8DKrnw@mail.gmail.com>

Thanks @joel, for the guidance.  I will get right on it, and hopefully have
something for public consumption soon!

-M

On Sun, Aug 20, 2017 at 5:28 AM, Joel Nothman <joel.nothman at gmail.com>
wrote:

> The idea is to take the template (https://github.com/scikit-
> learn-contrib/project-template), build, test and document your
> estimator(s), and offer it to be housed within scikit-learn-contrib.
>
> On 20 August 2017 at 08:36, Michael Capizzi <mcapizzi at email.arizona.edu>
> wrote:
>
>> Thanks @joel -
>>
>> I wasn?t aware of scikit-learn-contrib. Is this what you?re referring
>> to? https://github.com/scikit-learn-contrib/scikit-learn-contrib
>>
>> If so, I don?t see any existing projects that this would fit into; could
>> I start a new one in a pull-request?
>>
>> -M
>> ?
>>
>> On Sat, Aug 19, 2017 at 2:47 AM, Joel Nothman <joel.nothman at gmail.com>
>> wrote:
>>
>>> this is the right place to ask, but I'd be more interested to see a
>>> scikit-learn-compatible implementation available, perhaps in
>>> scikit-learn-contrib more than to see it part of the main package...
>>>
>>> On 19 Aug 2017 2:13 am, "Michael Capizzi" <mcapizzi at email.arizona.edu>
>>> wrote:
>>>
>>>> Hi all -
>>>>
>>>> Forgive me if this is the wrong place for posting this question, but
>>>> I'd like to inquire about the community's interest in incorporating a new
>>>> Transformer into the code base.
>>>>
>>>> This paper ( https://nlp.stanford.edu/pubs/sidaw12_simple_sentiment.pdf )
>>>> is a "classic" in Natural Language Processing and is often times used as a
>>>> very competitive baseline.  TL;DR it transforms a traditional count-based
>>>> feature space into the conditional probabilities of a `Naive Bayes`
>>>> classifier.  These transformed features can then be used to train any
>>>> linear classifier.  The paper focuses on `SVM`.
>>>>
>>>> The attached notebook has an example of the custom `Transformer` I
>>>> built along with a custom `Classifier` to utilize this `Transformer` in a
>>>> `multiclass` case (as the feature space transformation differs depending on
>>>> the label).
>>>>
>>>> If there is interest in the community for the inclusion of this
>>>> `Transformer` and `Classifier`, I'd happily go through the official process
>>>> of a `pull-request`, etc.
>>>>
>>>> -Michael
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170822/cca50c9a/attachment.html>

From s.atasever at gmail.com  Wed Aug 23 04:55:21 2017
From: s.atasever at gmail.com (Sema Atasever)
Date: Wed, 23 Aug 2017 11:55:21 +0300
Subject: [scikit-learn] Accessing Clustering Feature Tree in Birch
Message-ID: <CAAir+Cq+Nu6k9=B1YrNt9YbN8cTNQfV3m94SMsVbb16Fac3m1Q@mail.gmail.com>

Dear scikit-learn members,

Considering the "CF-tree" data structure :

- How can i *access Clustering Feature Tree* in Birch?

- For example, how many clusters are there in the hierarchy under the *root
node* and what are the data samples in this cluster?

- Can I get them separately for 3 trees?

Best.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170823/01071e01/attachment-0001.html>
-------------- next part --------------
from sklearn.cluster import Birch
from sklearn.externals import joblib
import numpy as np
import matplotlib.pyplot as plt

X=np.loadtxt(open("C:\dataset.txt", "rb"), delimiter=";")

brc = Birch(branching_factor=50, n_clusters=None, threshold=0.5,compute_labels=True,copy=True)

brc.fit(X)  

birch_predict=brc.predict(X)
print ("\nClustering_result:\n")
print (birch_predict)

np.savetxt('birch_predict_CLASS_0.csv', birch_predict,fmt="%i", delimiter=',')

-------------- next part --------------
0.008442;0.66962;0.742786;0.12921;0.035335;0.5;0;0.751871;0.164742;0.083387;0.52606;0.426925;0.047015;0.760345;0.051905;0.18775;0.678724;0.181262;0.09258;0.527265;0.708205;0.09418
0.12921;0.755105;0.751871;0.164742;0.051905;0.047426;0;0.678724;0.181262;0.140013;0.52578;0.436165;0.038055;0.74779;0.09258;0.15963;0.666024;0.180252;0.09418;0.528065;0.717575;0.076115
0.164742;0.760345;0.678724;0.181262;0.09258;0.993307;0;0.666024;0.180252;0.153723;0.527265;0.431535;0.0412;0.708205;0.09418;0.197615;0.669414;0.167021;0.076115;0.528445;0.71082;0.06787
0.181262;0.74779;0.666024;0.180252;0.09418;0.047426;0.389599;0.669414;0.167021;0.163565;0.528065;0.409905;0.06203;0.717575;0.076115;0.20631;0.663989;0.155896;0.06787;0.52745;0.72124;0.073745
0.180252;0.708205;0.669414;0.167021;0.076115;0.017986;0;0.663989;0.155896;0.180117;0.528445;0.38415;0.087405;0.71082;0.06787;0.221315;0.66713;0.159766;0.073745;0.523975;0.73143;0.088295
0.167021;0.717575;0.663989;0.155896;0.06787;0.268941;0;0.66713;0.159766;0.173103;0.52745;0.389885;0.08267;0.72124;0.073745;0.20501;0.665781;0.1704;0.088295;0.510225;0.70722;0.083135
0.155896;0.71082;0.66713;0.159766;0.073745;0.5;0;0.665781;0.1704;0.163819;0.523975;0.406555;0.069475;0.73143;0.088295;0.18027;0.642032;0.165493;0.083135;0.49496;0.67203;0.1121
0.309835;0.614635;0.59243;0.24016;0.14647;0.119203;0;0.63434;0.194778;0.17088;0.61152;0.290565;0.09791;0.65716;0.09899;0.24385;0.709535;0.130572;0.044365;0.645405;0.734585;0.024905
0.24016;0.581995;0.63434;0.194778;0.09899;0.017986;0;0.709535;0.130572;0.159893;0.61952;0.256325;0.124155;0.691135;0.044365;0.2645;0.732647;0.113563;0.024905;0.664725;0.83201;0.02481
0.194778;0.65716;0.709535;0.130572;0.044365;0.017986;0;0.732647;0.113563;0.15379;0.645405;0.22476;0.129835;0.734585;0.024905;0.24051;0.771562;0.113287;0.02481;0.666795;0.745095;0.024735
0.130572;0.691135;0.732647;0.113563;0.024905;0.5;0;0.771562;0.113287;0.115153;0.664725;0.224025;0.111255;0.83201;0.02481;0.14318;0.723874;0.121635;0.024735;0.658105;0.70706;0.023255
0.113563;0.734585;0.771562;0.113287;0.02481;0.268941;0;0.723874;0.121635;0.154492;0.666795;0.220035;0.11317;0.745095;0.024735;0.23017;0.708299;0.116688;0.023255;0.63076;0.684905;0.017255
0.113287;0.83201;0.723874;0.121635;0.024735;0.5;0.854954;0.708299;0.116688;0.175015;0.658105;0.206675;0.13522;0.70706;0.023255;0.26969;0.468897;0.088995;0.017255;0.180385;0.455915;0.007055
0.121635;0.745095;0.708299;0.116688;0.023255;0.5;0;0.468897;0.088995;0.442108;0.63076;0.158705;0.210535;0.684905;0.017255;0.29784;0.231702;0.169084;0.007055;0.009225;0.027455;0.0043
0.035705;0.02416;0.072467;0.015195;0.01129;0.119203;0.262697;0.790346;0.01195;0.197704;0.769445;0.000195;0.23036;0.668065;0.00242;0.329515;0.793541;0.011307;0.000645;0.784085;0.7004;0.00229
0.015195;0.03245;0.790346;0.01195;0.00242;0.017986;0;0.793541;0.011307;0.19515;0.764935;0.00004;0.235025;0.68216;0.000645;0.31719;0.686992;0.007612;0.00229;0.97409;0.79086;0.08215
0.01195;0.668065;0.793541;0.011307;0.000645;0.006693;0;0.686992;0.007612;0.305397;0.784085;0.00002;0.215895;0.7004;0.00229;0.29731;0.907967;0.034313;0.08215;0.97517;0.688985;0.075555
0.011307;0.68216;0.686992;0.007612;0.00229;0.952574;0;0.907967;0.034313;0.057718;0.97409;0.000265;0.025645;0.79086;0.08215;0.126985;0.875077;0.031757;0.075555;0.97471;0.653285;0.061705
0.007612;0.7004;0.907967;0.034313;0.08215;0.5;0.667078;0.875077;0.031757;0.093164;0.97517;0.000255;0.024575;0.688985;0.075555;0.235455;0.863024;0.027104;0.061705;0.980635;0.70174;0.01938
0.034313;0.79086;0.875077;0.031757;0.075555;0.017986;0;0.863024;0.027104;0.109872;0.97471;0.000145;0.025145;0.653285;0.061705;0.28501;0.881151;0.012997;0.01938;0.97968;0.715565;0.0084
0.031757;0.688985;0.863024;0.027104;0.061705;0.119203;0;0.881151;0.012997;0.105856;0.980635;0.00015;0.01922;0.70174;0.01938;0.278885;0.885441;0.009437;0.0084;0.984365;0.72635;0.00049
0.027104;0.653285;0.881151;0.012997;0.01938;0.119203;0;0.885441;0.009437;0.105121;0.97968;0.00045;0.019865;0.715565;0.0084;0.276035;0.891236;0.006388;0.00049;0.993995;0.938085;0.006585
0.012997;0.70174;0.885441;0.009437;0.0084;0.047426;0;0.891236;0.006388;0.102376;0.984365;0.00017;0.015465;0.72635;0.00049;0.27316;0.965025;0.008656;0.006585;0.994325;0.961415;0.01333
0.009437;0.715565;0.891236;0.006388;0.00049;0.880797;0.667078;0.965025;0.008656;0.026319;0.993995;0.00088;0.005125;0.938085;0.006585;0.05533;0.972911;0.010983;0.01333;0.99423;0.973995;0.015275
0.006388;0.72635;0.965025;0.008656;0.006585;0.731059;0.948778;0.972911;0.010983;0.016106;0.994325;0.001115;0.00456;0.961415;0.01333;0.025255;0.977073;0.011641;0.015275;0.992885;0.97204;0.011895
0.008656;0.938085;0.972911;0.010983;0.01333;0.119203;0;0.977073;0.011641;0.011284;0.99423;0.001145;0.004625;0.973995;0.015275;0.010725;0.975973;0.010504;0.011895;0.98109;0.960235;0.01008
0.010983;0.961415;0.977073;0.011641;0.015275;0.5;0;0.975973;0.010504;0.013521;0.992885;0.001115;0.005995;0.97204;0.011895;0.016065;0.968106;0.009809;0.01008;0.905975;0.947335;0.00057
0.011641;0.973995;0.975973;0.010504;0.011895;0.5;0;0.968106;0.009809;0.022084;0.98109;0.000845;0.018065;0.960235;0.01008;0.029685;0.938041;0.00688;0.00057;0.629465;0.614855;0.00049
0.010504;0.97204;0.968106;0.009809;0.01008;0.5;0.488002;0.938041;0.00688;0.05508;0.905975;0.000475;0.09355;0.947335;0.00057;0.052095;0.59958;0.007675;0.00049;0.398225;0.054915;0.00193
0.009809;0.960235;0.938041;0.00688;0.00057;0.047426;0.223393;0.59958;0.007675;0.392743;0.629465;0.00032;0.37021;0.614855;0.00049;0.384655;0.163412;0.014925;0.00193;0.084205;0.0059;0.013055
0.326985;0.07661;0.051845;0.139195;0.186475;0.268941;0;0.06022;0.097127;0.842653;0.080545;0.043865;0.875595;0.039895;0.15039;0.80971;0.088763;0.016698;0.020645;0.091805;0.080505;0.02658
0.139195;0.083335;0.06022;0.097127;0.15039;0.268941;0.779714;0.088763;0.016698;0.894545;0.09449;0.01275;0.892765;0.083035;0.020645;0.896325;0.086155;0.042025;0.02658;0.09471;0.06764;0.31128
0.097127;0.039895;0.088763;0.016698;0.020645;0.017986;0;0.086155;0.042025;0.87182;0.091805;0.05747;0.850725;0.080505;0.02658;0.892915;0.081175;0.298045;0.31128;0.050345;0.02021;0.737805
0.495558;0.185695;0.093198;0.530914;0.220625;0.997527;0.781427;0.099843;0.516794;0.383363;0.099525;0.451045;0.44943;0.199435;0.100475;0.70009;0.098953;0.203845;0.02881;0.10351;0.217985;0.017295
0.530914;0.183815;0.099843;0.516794;0.100475;0.119203;0.64497;0.098953;0.203845;0.697199;0.09697;0.582155;0.32087;0.19932;0.02881;0.771865;0.107355;0.213066;0.017295;0.099415;0.22013;0.045165
0.516794;0.199435;0.098953;0.203845;0.02881;0.997527;0.457106;0.107355;0.213066;0.679579;0.10351;0.621335;0.275155;0.217985;0.017295;0.76472;0.106705;0.218615;0.045165;0.09363;0.26663;0.04146
0.203845;0.19932;0.107355;0.213066;0.017295;0.047426;0;0.106705;0.218615;0.674683;0.099415;0.61011;0.290475;0.22013;0.045165;0.73471;0.120276;0.457291;0.04146;0.07761;0.304405;0.03027
0.213066;0.217985;0.106705;0.218615;0.045165;0.119203;0.18954;0.120276;0.457291;0.422436;0.09363;0.33155;0.574825;0.26663;0.04146;0.691915;0.127528;0.058846;0.03027;0.026315;0.65796;0.00464
0.218615;0.22013;0.120276;0.457291;0.04146;0.047426;0.390788;0.127528;0.058846;0.813624;0.07761;0.1457;0.77669;0.304405;0.03027;0.66532;0.342137;0.008417;0.00464;0.03819;0.84422;0.00579
0.003138;0.84422;0.654379;0.003161;0.000425;0.017986;0.414595;0.783778;0.00473;0.211496;0.39544;0.012155;0.592415;0.95703;0.001465;0.041505;0.914397;0.003594;0.00043;0.809;0.99335;0.000095
0.003161;0.91128;0.783778;0.00473;0.001465;0.5;0.732433;0.914397;0.003594;0.082006;0.757605;0.009785;0.232605;0.98672;0.00043;0.012845;0.933738;0.002311;0.000095;0.877555;0.997475;0.00003
0.00473;0.95703;0.914397;0.003594;0.00043;0.006693;0.799472;0.933738;0.002311;0.063946;0.809;0.00627;0.18472;0.99335;0.000095;0.00655;0.957965;0.002491;0.00003;0.94554;0.99902;0.000005
0.003594;0.98672;0.933738;0.002311;0.000095;0.006693;0.478014;0.957965;0.002491;0.039544;0.877555;0.006875;0.11557;0.997475;0.00003;0.002495;0.981142;0.003436;0.000005;0.958705;0.999125;0
0.002311;0.99335;0.957965;0.002491;0.00003;0.047426;0.39556;0.981142;0.003436;0.015424;0.94554;0.009735;0.044725;0.99902;0.000005;0.00098;0.985565;0.005279;0;0.95914;0.999945;0
0.002491;0.997475;0.981142;0.003436;0.000005;0.993307;0.907795;0.985565;0.005279;0.009157;0.958705;0.01527;0.02603;0.999125;0;0.000875;0.985984;0.005669;0;0.960575;0.999935;0
0.003436;0.99902;0.985565;0.005279;0;0.017986;0.695508;0.985984;0.005669;0.008347;0.95914;0.01644;0.02442;0.999945;0;0.000055;0.986459;0.005656;0;0.961995;0.99982;0
0.005279;0.999125;0.985984;0.005669;0;0.047426;0.925048;0.986459;0.005656;0.007886;0.960575;0.0164;0.023025;0.999935;0;0.000065;0.986894;0.005554;0;0.95715;0.99988;0
0.005669;0.999945;0.986459;0.005656;0;0.880797;0.319733;0.986894;0.005554;0.007552;0.961995;0.016095;0.02191;0.99982;0;0.00018;0.985299;0.005352;0;0.90035;0.999795;0
0.005656;0.999935;0.986894;0.005554;0;0.268941;0.863303;0.985299;0.005352;0.009349;0.95715;0.01549;0.02736;0.99988;0;0.00012;0.966326;0.005159;0;0.805855;0.999615;0
0.005554;0.99982;0.985299;0.005352;0;0.047426;0.601807;0.966326;0.005159;0.028516;0.90035;0.014895;0.08476;0.999795;0;0.000205;0.934732;0.003761;0;0.606475;0.99777;0.00001
0.003039;0.99777;0.670359;0.001451;0.00001;0.268941;0.586375;0.629746;0.000693;0.369563;0.011315;0.000355;0.988335;0.88131;0.00003;0.11866;0.23484;0.001218;0.000075;0.00434;0.43038;0.00258
0.001451;0.99192;0.629746;0.000693;0.00003;0.119203;0.433889;0.23484;0.001218;0.763943;0.00538;0.00105;0.99357;0.6911;0.000075;0.30883;0.219774;0.076059;0.00258;0.003685;0.10887;0.00114
0.000693;0.88131;0.23484;0.001218;0.000075;0.047426;0.458099;0.219774;0.076059;0.704164;0.00434;0.000995;0.99466;0.43038;0.00258;0.56704;0.112386;0.085661;0.00114;0.002;0.03147;0.002315
0.032934;0.22018;0.25514;0.028839;0.01624;0.119203;0.7191;0.603796;0.011174;0.38503;0.29032;0.02698;0.682695;0.56421;0.0017;0.434095;0.923163;0.003005;0.000175;0.889325;0.984285;0.000035
0.028839;0.349925;0.603796;0.011174;0.0017;0.047426;0.334033;0.923163;0.003005;0.073832;0.84408;0.004025;0.151895;0.943165;0.000175;0.05666;0.9378;0.002788;0.000035;0.955635;0.99905;0.00001
0.011174;0.56421;0.923163;0.003005;0.000175;0.5;0.629483;0.9378;0.002788;0.059412;0.889325;0.003655;0.10702;0.984285;0.000035;0.01568;0.948102;0.002001;0.00001;0.98219;0.999775;0.000005
0.003005;0.943165;0.9378;0.002788;0.000035;0.952574;0.584676;0.948102;0.002001;0.049899;0.955635;0.005635;0.03873;0.99905;0.00001;0.000945;0.957786;0.004236;0.000005;0.98412;0.999975;0
0.002788;0.984285;0.948102;0.002001;0.00001;0.952574;0.905081;0.957786;0.004236;0.037978;0.98219;0.01235;0.00546;0.999775;0.000005;0.00022;0.952973;0.012087;0;0.984635;0.999955;0
0.002001;0.99905;0.957786;0.004236;0.000005;0.017986;0.340964;0.952973;0.012087;0.03494;0.98412;0.013755;0.002125;0.999975;0;0.000025;0.951057;0.014627;0;0.98458;0.99991;0
0.004236;0.999775;0.952973;0.012087;0;0.880797;0.873028;0.951057;0.014627;0.034314;0.984635;0.01438;0.00098;0.999955;0;0.000045;0.950876;0.015671;0;0.98404;0.99996;0
0.012087;0.999975;0.951057;0.014627;0;0.880797;0.697622;0.950876;0.015671;0.033453;0.98458;0.014395;0.001025;0.99991;0;0.00009;0.960394;0.005822;0;0.982855;0.999905;0
0.014627;0.999955;0.950876;0.015671;0;0.952574;0.905253;0.960394;0.005822;0.033783;0.98404;0.014125;0.001835;0.99996;0;0.00004;0.957887;0.005889;0;0.97646;0.99974;0
0.015671;0.99991;0.960394;0.005822;0;0.047426;0.712795;0.957887;0.005889;0.036224;0.982855;0.013745;0.0034;0.999905;0;0.000095;0.956875;0.004227;0;0.955485;0.999375;0.000005
0.005822;0.99996;0.957887;0.005889;0;0.5;0.78279;0.956875;0.004227;0.038896;0.97646;0.008745;0.01479;0.99974;0;0.00026;0.92999;0.003049;0.000005;0.938535;0.99311;0.000005
0.005889;0.999905;0.956875;0.004227;0;0.119203;0.579568;0.92999;0.003049;0.066959;0.955485;0.005195;0.03932;0.999375;0.000005;0.000615;0.942257;0.002056;0.000005;0.87656;0.911585;0
0.004227;0.99974;0.92999;0.003049;0.000005;0.268941;0.759511;0.942257;0.002056;0.055687;0.938535;0.00228;0.059185;0.99311;0.000005;0.006885;0.745626;0.015394;0;0.653185;0.171825;0.00004
0.003049;0.999375;0.942257;0.002056;0.000005;0.119203;0.872917;0.745626;0.015394;0.23898;0.87656;0.00142;0.12202;0.911585;0;0.088415;0.309117;0.141822;0.00004;0.55808;0.00044;0.000785
0.144274;0.04801;0.146957;0.084234;0.11681;0.119203;0.450414;0.536664;0.04495;0.418386;0.195625;0.079095;0.72528;0.568975;0.04854;0.382485;0.64845;0.021258;0.03926;0.39226;0.90174;0.05751
0.084234;0.17759;0.536664;0.04495;0.04854;0.017986;0.147795;0.64845;0.021258;0.33029;0.311835;0.01742;0.670745;0.73025;0.03926;0.230485;0.761245;0.036791;0.05751;0.52422;0.929955;0.06594
0.04495;0.568975;0.64845;0.021258;0.03926;0.119203;0.746872;0.761245;0.036791;0.20196;0.39226;0.045655;0.562085;0.90174;0.05751;0.04074;0.815355;0.049582;0.06594;0.536225;0.930545;0.066465
0.021258;0.73025;0.761245;0.036791;0.05751;0.982014;0.609212;0.815355;0.049582;0.135061;0.52422;0.075575;0.400205;0.929955;0.06594;0.0041;0.818375;0.056409;0.066465;0.5574;0.930585;0.067005
0.036791;0.90174;0.815355;0.049582;0.06594;0.017986;0.608021;0.818375;0.056409;0.125216;0.536225;0.09544;0.368335;0.930545;0.066465;0.00299;0.826736;0.093697;0.067005;0.61885;0.930695;0.066395
0.049582;0.929955;0.818375;0.056409;0.066465;0.047426;0.207839;0.826736;0.093697;0.079569;0.5574;0.20671;0.23589;0.930585;0.067005;0.002415;0.847252;0.12085;0.066395;0.619115;0.928225;0.06536
0.056409;0.930545;0.826736;0.093697;0.067005;0.982014;0.896135;0.847252;0.12085;0.031899;0.61885;0.28877;0.092385;0.930695;0.066395;0.00291;0.788514;0.118014;0.06536;0.575755;0.919285;0.046905
0.093697;0.930585;0.847252;0.12085;0.066395;0.731059;0.721115;0.788514;0.118014;0.093474;0.619115;0.28828;0.092605;0.928225;0.06536;0.00642;0.772007;0.109543;0.046905;0.51972;0.86882;0.01736
0.12085;0.930695;0.788514;0.118014;0.06536;0.017986;0.418971;0.772007;0.109543;0.118451;0.575755;0.28133;0.142915;0.919285;0.046905;0.033815;0.734065;0.097927;0.01736;0.45373;0.409985;0.00429
0.118014;0.928225;0.772007;0.109543;0.046905;0.119203;0.375194;0.734065;0.097927;0.168009;0.51972;0.276015;0.204265;0.86882;0.01736;0.11382;0.347599;0.093625;0.00429;0.32548;0.002525;0.003945
0.109543;0.919285;0.734065;0.097927;0.01736;0.268941;0.080394;0.347599;0.093625;0.558776;0.45373;0.275445;0.270825;0.409985;0.00429;0.585725;0.162941;0.128565;0.003945;0.304175;0.00075;0.19228
0.015575;0.128615;0.18182;0.010233;0.006035;0.006693;0.628083;0.509804;0.002916;0.48728;0.564515;0.000295;0.43519;0.82232;0.000115;0.177565;0.836858;0.003037;0.00001;0.843905;0.956825;0.000005
0.010233;0.131695;0.509804;0.002916;0.000115;0.119203;0.665299;0.836858;0.003037;0.160105;0.642435;0.00077;0.356795;0.89494;0.00001;0.10505;0.924991;0.001083;0.000005;0.851295;0.999195;0
0.002916;0.82232;0.836858;0.003037;0.00001;0.993307;0.4985;0.924991;0.001083;0.073921;0.843905;0.00277;0.15332;0.956825;0.000005;0.04316;0.941577;0.000813;0;0.919575;0.999835;0
0.003037;0.89494;0.924991;0.001083;0.000005;0.119203;0.606828;0.941577;0.000813;0.057608;0.851295;0.001965;0.14674;0.999195;0;0.0008;0.964536;0.000653;0;0.952645;0.99983;0
0.001083;0.956825;0.941577;0.000813;0;0.017986;0.601088;0.964536;0.000653;0.034808;0.919575;0.001485;0.07893;0.999835;0;0.000165;0.975557;0.001343;0;0.95212;0.999935;0
0.000813;0.999195;0.964536;0.000653;0;0.047426;0.514996;0.975557;0.001343;0.023099;0.952645;0.003555;0.0438;0.99983;0;0.00017;0.977389;0.003606;0;0.94496;0.999805;0
0.000653;0.999835;0.975557;0.001343;0;0.880797;0.218744;0.977389;0.003606;0.019006;0.95212;0.010345;0.03754;0.999935;0;0.000065;0.974858;0.006419;0;0.937955;0.99909;0
0.001343;0.99983;0.977389;0.003606;0;0.047426;0.347964;0.974858;0.006419;0.018723;0.94496;0.01178;0.04326;0.999805;0;0.000195;0.973184;0.004053;0;0.90462;0.997515;0
0.003606;0.999935;0.974858;0.006419;0;0.047426;0.49725;0.973184;0.004053;0.022763;0.937955;0.011675;0.05037;0.99909;0;0.00091;0.960937;0.003776;0;0.799285;0.986475;0.000035
0.006419;0.999805;0.973184;0.004053;0;0.731059;0.426046;0.960937;0.003776;0.035285;0.90462;0.010795;0.08458;0.997515;0;0.002485;0.928221;0.003408;0.000035;0.72453;0.894125;0.00006
0.004053;0.99909;0.960937;0.003776;0;0.047426;0.725916;0.928221;0.003408;0.068373;0.799285;0.00964;0.19108;0.986475;0.000035;0.01349;0.842942;0.003839;0.00006;0.698595;0.739345;0.00018
0.003776;0.997515;0.928221;0.003408;0.000035;0.047426;0.299643;0.842942;0.003839;0.153217;0.72453;0.010775;0.264695;0.894125;0.00006;0.10581;0.650625;0.012722;0.00018;0.663635;0.65696;0.008015
0.003408;0.986475;0.842942;0.003839;0.00006;0.5;0.536187;0.650625;0.012722;0.336655;0.698595;0.03711;0.2643;0.739345;0.00018;0.260475;0.505169;0.043735;0.008015;0.630895;0.39294;0.00277
0.072568;0.06769;0.021446;0.038434;0.07939;0.047426;0;0.348958;0.036144;0.614896;0.077135;0.00615;0.91671;0.021545;0.07638;0.902075;0.348864;0.030603;0.034935;0.11489;0.024195;0.162845
0.038434;0.020485;0.348958;0.036144;0.07638;0.119203;0.803924;0.348864;0.030603;0.620533;0.085065;0.027595;0.88734;0.020085;0.034935;0.94498;0.357251;0.112677;0.162845;0.11449;0.02295;0.32538
0.036144;0.021545;0.348864;0.030603;0.034935;0.119203;0;0.357251;0.112677;0.530072;0.11489;0.14152;0.743585;0.024195;0.162845;0.812965;0.349695;0.205096;0.32538;0.095495;0.01456;0.18781
0.402061;0.154955;0.200411;0.383701;0.146995;0.047426;0.335815;0.644193;0.058533;0.297273;0.413315;0.108035;0.478645;0.64462;0.004885;0.350495;0.647131;0.056199;0.005445;0.40603;0.673135;0.008125
0.383701;0.16909;0.644193;0.058533;0.004885;0.047426;0.165481;0.647131;0.056199;0.296669;0.416905;0.100475;0.48262;0.649845;0.005445;0.34471;0.65127;0.039951;0.008125;0.39438;0.667785;0.00753
0.058533;0.64462;0.647131;0.056199;0.005445;0.119203;0;0.65127;0.039951;0.308776;0.40603;0.04905;0.54492;0.673135;0.008125;0.31873;0.64146;0.035531;0.00753;0.376435;0.666245;0.00642
0.056199;0.649845;0.65127;0.039951;0.008125;0.731059;0.635758;0.64146;0.035531;0.323009;0.39438;0.03017;0.57545;0.667785;0.00753;0.324685;0.373052;0.032856;0.00642;0.34128;0.66615;0.030435
0;0;0.150847;0.094682;0.099565;0.006693;0.414353;0.849131;0.000118;0.150753;0.880315;0.000095;0.11959;0.882355;0.00012;0.11753;0.922273;0.000065;0.00003;0.992165;0.976785;0.00005
0.094682;0.02561;0.849131;0.000118;0.00012;0.731059;0.388648;0.922273;0.000065;0.077658;0.98756;0.00006;0.012375;0.956645;0.00003;0.04332;0.924201;0.00007;0.00005;0.99962;0.99817;0
0.000118;0.882355;0.922273;0.000065;0.00003;0.119203;0.513747;0.924201;0.00007;0.075727;0.992165;0.00007;0.007765;0.976785;0.00005;0.02316;0.957982;0.000043;0;0.99969;0.99949;0
0.000065;0.956645;0.924201;0.00007;0.00005;0.268941;0.348418;0.957982;0.000043;0.041973;0.99962;0.000055;0.00032;0.99817;0;0.00183;0.974686;0.000038;0;0.99974;0.999935;0
0.00007;0.976785;0.957982;0.000043;0;0.006693;0.649991;0.974686;0.000038;0.025273;0.99969;0.00005;0.00025;0.99949;0;0.00051;0.98024;0.000037;0;0.9996;0.99978;0
0.000043;0.99817;0.974686;0.000038;0;0.000911;0.53096;0.98024;0.000037;0.019725;0.99974;0.000055;0.00021;0.999935;0;0.000065;0.983794;0.000048;0;0.9994;0.99998;0
0.000038;0.99949;0.98024;0.000037;0;0.5;0.349327;0.983794;0.000048;0.016156;0.9996;0.000095;0.0003;0.99978;0;0.00022;0.990775;0.000049;0;0.99923;0.999985;0
0.000037;0.999935;0.983794;0.000048;0;0.047426;0.27768;0.990775;0.000049;0.009177;0.9994;0.0001;0.000505;0.99998;0;0.00002;0.99067;0.000062;0;0.99604;0.9999;0
0.000048;0.99978;0.990775;0.000049;0;0.119203;0.380365;0.99067;0.000062;0.009267;0.99923;0.00014;0.00063;0.999985;0;0.000015;0.986525;0.00013;0;0.986525;0.99898;0
0.000049;0.99998;0.99067;0.000062;0;0.880797;0.668853;0.986525;0.00013;0.013345;0.99604;0.000345;0.003615;0.9999;0;0.0001;0.983631;0.000132;0;0.96961;0.99821;0
0.000062;0.999985;0.986525;0.00013;0;0.002473;0.100833;0.983631;0.000132;0.016239;0.986525;0.00035;0.01313;0.99898;0;0.00102;0.971793;0.000128;0;0.950005;0.994645;0.000005
0.00013;0.9999;0.983631;0.000132;0;0.017986;0.557988;0.971793;0.000128;0.028079;0.96961;0.00034;0.03005;0.99821;0;0.00179;0.939341;0.001325;0.000005;0.7612;0.908295;0.000745
0.000132;0.99898;0.971793;0.000128;0;0.5;0.391741;0.939341;0.001325;0.059334;0.950005;0.00187;0.048125;0.994645;0.000005;0.00535;0.832202;0.010909;0.000745;0.20003;0.294375;0.000355
0.000128;0.99821;0.939341;0.001325;0.000005;0.268941;0.276678;0.832202;0.010909;0.156889;0.7612;0.00267;0.23613;0.908295;0.000745;0.09096;0.24415;0.012425;0.000355;0.12832;0.04852;0.00063
0.700558;0.00128;0.033528;0.51778;0.69314;0.119203;0.467048;0.957029;0.018677;0.024294;0.988125;0.00001;0.01186;0.9871;0.000125;0.01278;0.961403;0.011465;0;0.999395;0.99769;0.000005
0.51778;0.001365;0.957029;0.018677;0.000125;0.047426;0.316695;0.961403;0.011465;0.027132;0.99916;0.000005;0.000835;0.997235;0;0.002765;0.966611;0.011268;0.000005;0.999775;0.99896;0
0.018677;0.9871;0.961403;0.011465;0;0.017986;0.108515;0.966611;0.011268;0.022121;0.999395;0;0.000605;0.99769;0.000005;0.002305;0.97066;0.011267;0;0.99999;0.99999;0
0.011465;0.997235;0.966611;0.011268;0.000005;0.5;0.099302;0.97066;0.011267;0.018073;0.999775;0;0.000225;0.99896;0;0.00104;0.967128;0.010435;0;0.999975;0.999955;0
0.011268;0.99769;0.97066;0.011267;0;0.268941;0.567093;0.967128;0.010435;0.022437;0.99999;0;0.00001;0.99999;0;0.00001;0.96963;0.011402;0;0.99997;0.99994;0
0.011267;0.99896;0.967128;0.010435;0;0.119203;0.653169;0.96963;0.011402;0.018968;0.999975;0;0.000025;0.999955;0;0.000045;0.968817;0.008672;0;0.9999;0.999595;0
0.010435;0.99999;0.96963;0.011402;0;0.268941;0.797057;0.968817;0.008672;0.022511;0.99997;0;0.00003;0.99994;0;0.00006;0.952639;0.000037;0;0.99978;0.999795;0
0.011402;0.999955;0.968817;0.008672;0;0.731059;0.563653;0.952639;0.000037;0.047324;0.9999;0;0.0001;0.999595;0;0.000405;0.941544;0.000053;0;0.999345;0.99894;0
0.008672;0.99994;0.952639;0.000037;0;0.047426;0.613963;0.941544;0.000053;0.058401;0.99978;0;0.000215;0.999795;0;0.000205;0.953784;0.033575;0;0.99645;0.99539;0.00001
0.000037;0.999595;0.941544;0.000053;0;0.119203;0.529715;0.953784;0.033575;0.012639;0.999345;0.00004;0.00061;0.99894;0;0.00106;0.950305;0.037169;0.00001;0.992425;0.988895;0.00003
0.000053;0.999795;0.953784;0.033575;0;0.731059;0.488502;0.950305;0.037169;0.012526;0.99645;0.00008;0.003465;0.99539;0.00001;0.004605;0.953128;0.015558;0.00003;0.96765;0.981705;0.00012
0.033575;0.99894;0.950305;0.037169;0.00001;0.5;0.426536;0.953128;0.015558;0.031314;0.992425;0.000085;0.00749;0.988895;0.00003;0.011075;0.933835;0.016748;0.00012;0.82843;0.899765;0.000305
0.037169;0.99539;0.953128;0.015558;0.00003;0.119203;0.554285;0.933835;0.016748;0.049415;0.96765;0.00009;0.032255;0.981705;0.00012;0.018175;0.835794;0.015184;0.000305;0.151705;0.462245;0.000715
0.015558;0.988895;0.933835;0.016748;0.00012;0.5;0.60444;0.835794;0.015184;0.149021;0.82843;0.000105;0.171465;0.899765;0.000305;0.099925;0.312008;0.008636;0.000715;0.07423;0.025605;0.00077
0.016792;0.083015;0.117144;0.078285;0.04578;0.119203;0.546614;0.248611;0.003692;0.7477;0.21343;0.00002;0.786555;0.33779;0.00833;0.653885;0.393686;0.005201;0.00746;0.550185;0.461935;0.003405
0.078285;0.12086;0.248611;0.003692;0.00833;0.119203;0.455617;0.393686;0.005201;0.601113;0.38517;0;0.61483;0.4509;0.00746;0.54164;0.432389;0.001162;0.003405;0.888935;0.87477;0.00035
0.003692;0.33779;0.393686;0.005201;0.00746;0.119203;0.231831;0.432389;0.001162;0.566446;0.550185;0.00001;0.4498;0.461935;0.003405;0.534655;0.841113;0.000142;0.00035;0.97981;0.92331;0.00007
0.005201;0.4509;0.432389;0.001162;0.003405;0.5;0.280497;0.841113;0.000142;0.158745;0.888935;0.00001;0.111055;0.87477;0.00035;0.12488;0.900356;0.000046;0.00007;0.990275;0.97047;0.00004
0.001162;0.461935;0.841113;0.000142;0.00035;0.119203;0.523732;0.900356;0.000046;0.099599;0.97981;0.000005;0.020185;0.92331;0.00007;0.076625;0.968005;0.001144;0.00004;0.99949;0.99538;0.000115
0.000142;0.87477;0.900356;0.000046;0.00007;0.268941;0.439301;0.968005;0.001144;0.030853;0.990275;0.000005;0.009725;0.97047;0.00004;0.02949;0.993907;0.001241;0.000115;0.99965;0.996815;0.00004
0.000046;0.92331;0.968005;0.001144;0.00004;0.880797;0.350692;0.993907;0.001241;0.004852;0.99949;0;0.00051;0.99538;0.000115;0.004505;0.988958;0.001281;0.00004;0.99971;0.995905;0.000015
0.001144;0.97047;0.993907;0.001241;0.000115;0.5;0.329599;0.988958;0.001281;0.009761;0.99965;0;0.00035;0.996815;0.00004;0.003145;0.985293;0.000038;0.000015;0.999985;0.99454;0.00001
0.001241;0.99538;0.988958;0.001281;0.00004;0.119203;0.452642;0.985293;0.000038;0.014667;0.99971;0;0.000285;0.995905;0.000015;0.00408;0.979348;0.005637;0.00001;0.99995;0.995545;0.00001
0.001281;0.996815;0.985293;0.000038;0.000015;0.5;0.606828;0.979348;0.005637;0.015015;0.999985;0;0.000015;0.99454;0.00001;0.00545;0.994323;0.003886;0.00001;0.99995;0.99819;0
0.000038;0.995905;0.979348;0.005637;0.00001;0.5;0.719302;0.994323;0.003886;0.001791;0.99995;0;0.00005;0.995545;0.00001;0.004445;0.993189;0.000233;0;0.999355;0.999255;0
0.005637;0.99454;0.994323;0.003886;0.00001;0.268941;0.179167;0.993189;0.000233;0.006578;0.99995;0;0.00005;0.99819;0;0.00181;0.999115;0.000211;0;0.99927;0.99949;0
0.003886;0.995545;0.993189;0.000233;0;0.119203;0.67787;0.999115;0.000211;0.000674;0.999355;0;0.000645;0.999255;0;0.000745;0.998113;0.000171;0;0.99831;0.99966;0
0.000233;0.99819;0.999115;0.000211;0;0.731059;0;0.998113;0.000171;0.001716;0.99927;0.000005;0.000725;0.99949;0;0.00051;0.998331;0.000116;0;0.991755;0.99985;0
0.000211;0.999255;0.998113;0.000171;0;0.047426;0.286182;0.998331;0.000116;0.001553;0.99831;0.000005;0.001685;0.99966;0;0.00034;0.996304;0.000106;0;0.9818;0.99917;0
0.000171;0.99949;0.998331;0.000116;0;0.268941;0.441519;0.996304;0.000106;0.003589;0.991755;0.00001;0.008235;0.99985;0;0.00015;0.984398;0.000236;0;0.8794;0.99648;0.000005
0.000116;0.99966;0.996304;0.000106;0;0.119203;0.457354;0.984398;0.000236;0.015367;0.9818;0.000575;0.01763;0.99917;0;0.00083;0.94823;0.00062;0.000005;0.83004;0.981515;0.000025
0.000236;0.99917;0.94823;0.00062;0.000005;0.5;0.315831;0.916663;0.000691;0.082648;0.83004;0.002015;0.167945;0.981515;0.000025;0.018465;0.849844;0.000841;0.000045;0.60209;0.93436;0.009335
0.00062;0.99648;0.916663;0.000691;0.000025;0.119203;0.512497;0.849844;0.000841;0.149313;0.69651;0.002445;0.301045;0.943615;0.000045;0.056335;0.784729;0.00706;0.009335;0.550905;0.788395;0.01519
0.000691;0.981515;0.849844;0.000841;0.000045;0.119203;0.607305;0.784729;0.00706;0.208215;0.60209;0.01181;0.386105;0.93436;0.009335;0.05631;0.697286;0.014475;0.01519;0.00835;0.10455;0.00141
0.672462;0.005635;0.048684;0.372434;0.79551;0.5;0.023823;0.941118;0.006192;0.052691;0.9675;0.000015;0.032485;0.983405;0.00004;0.01656;0.955736;0.008144;0.000005;0.999815;0.999065;0
0.372434;0.006595;0.941118;0.006192;0.00004;0.047426;0.444974;0.955736;0.008144;0.036119;0.99976;0;0.000235;0.99842;0.000005;0.001575;0.976851;0.000012;0;0.99996;0.9999;0
0.006192;0.983405;0.955736;0.008144;0.000005;0.017986;0.199248;0.976851;0.000012;0.023137;0.999815;0.000005;0.00018;0.999065;0;0.000935;0.983451;0.000012;0;0.999985;0.99999;0
0.008144;0.99842;0.976851;0.000012;0;0.119203;0.057487;0.983451;0.000012;0.016538;0.99996;0.000005;0.000035;0.9999;0;0.0001;0.982751;0.000012;0;0.999985;0.99985;0
0.000012;0.999065;0.983451;0.000012;0;0.952574;0.597967;0.982751;0.000012;0.017236;0.999985;0.000005;0.000005;0.99999;0;0.00001;0.975726;0.000529;0;0.999965;0.999845;0
0.000012;0.9999;0.982751;0.000012;0;0.047426;0.574443;0.975726;0.000529;0.023745;0.999985;0.000005;0.00001;0.99985;0;0.00015;0.972878;0.000503;0;0.999435;0.999335;0
0.000012;0.99999;0.975726;0.000529;0;0.5;0.597967;0.972878;0.000503;0.026617;0.999965;0.000005;0.000025;0.999845;0;0.000155;0.967218;0.002922;0;0.99919;0.998885;0
0.000529;0.99985;0.972878;0.000503;0;0.880797;0.389361;0.967218;0.002922;0.029857;0.999435;0.000005;0.00055;0.999335;0;0.000665;0.946015;0.009153;0;0.998625;0.988535;0.00001
0.000503;0.999845;0.967218;0.002922;0;0.268941;0.563899;0.946015;0.009153;0.044832;0.99919;0.000005;0.000805;0.998885;0;0.001115;0.926613;0.003178;0.00001;0.99058;0.968105;0.00004
0.002922;0.999335;0.946015;0.009153;0;0.047426;0.58686;0.926613;0.003178;0.070208;0.998625;0.00002;0.001355;0.988535;0.00001;0.01145;0.853032;0.000042;0.00004;0.381295;0.802785;0.00148
0.009153;0.998885;0.926613;0.003178;0.00001;0.880797;0.279086;0.853032;0.000042;0.146931;0.99058;0.000025;0.0094;0.968105;0.00004;0.031865;0.558403;0.000559;0.00148;0.23099;0.6761;0.000085
0.002453;0.6761;0.256732;0.000116;0.00013;0.017986;0.44695;0.407903;0.000456;0.591639;0.19305;0.00006;0.806885;0.54496;0.00067;0.45437;0.658868;0.001987;0.00152;0.9086;0.75571;0.002275
0.000116;0.42864;0.407903;0.000456;0.00067;0.047426;0.516494;0.658868;0.001987;0.339146;0.579265;0.00031;0.420425;0.581085;0.00152;0.417395;0.787854;0.002991;0.002275;0.96037;0.92668;0.00197
0.000456;0.54496;0.658868;0.001987;0.00152;0.268941;0.347057;0.787854;0.002991;0.209154;0.9086;0.00226;0.089145;0.75571;0.002275;0.24201;0.881451;0.001828;0.00197;0.986125;0.96213;0.00024
0.002991;0.75571;0.881451;0.001828;0.00197;0.017986;0.350464;0.819756;0.00041;0.179832;0.986125;0.000085;0.013785;0.96213;0.00024;0.03763;0.807821;0.000242;0.000225;0.99317;0.98375;0.00066
0.001828;0.92668;0.819756;0.00041;0.00024;0.731059;0.45289;0.807821;0.000242;0.191934;0.9916;0.000265;0.008125;0.96689;0.000225;0.032885;0.833145;0.000286;0.00066;0.99834;0.985635;0.00004
0.00041;0.96213;0.807821;0.000242;0.000225;0.880797;0.282925;0.833145;0.000286;0.166569;0.99317;0.000075;0.006755;0.98375;0.00066;0.01559;0.933461;0.000049;0.00004;0.998295;0.99341;0.000005
0.000242;0.96689;0.833145;0.000286;0.00066;0.268941;0.233438;0.933461;0.000049;0.066492;0.99834;0.00004;0.001625;0.985635;0.00004;0.014325;0.948341;0.000038;0.000005;0.99838;0.997545;0
0.000038;0.99341;0.973207;0.000044;0;0.731059;0.676558;0.97968;0.00275;0.017572;0.998635;0.00019;0.001175;0.9994;0.000005;0.0006;0.97611;0.002973;0;0.998365;0.999885;0
0.000044;0.997545;0.97968;0.00275;0.000005;0.047426;0.498;0.97611;0.002973;0.020915;0.998615;0.000385;0.000995;0.99987;0;0.00013;0.980911;0.00034;0;0.996925;0.99964;0
0.00275;0.9994;0.97611;0.002973;0;0.5;0.634599;0.980911;0.00034;0.018749;0.998365;0.00099;0.000645;0.999885;0;0.000115;0.983061;0.000367;0;0.99072;0.997965;0
0.002973;0.99987;0.980911;0.00034;0;0.952574;0.843169;0.983061;0.000367;0.01657;0.996925;0.00107;0.001995;0.99964;0;0.00036;0.979397;0.000427;0;0.981765;0.996365;0.00001
0.00034;0.999885;0.983061;0.000367;0;0.5;0.093638;0.979397;0.000427;0.020175;0.99072;0.00125;0.008025;0.997965;0;0.002035;0.974694;0.00077;0.00001;0.94478;0.98166;0.00056
0.000367;0.99964;0.979397;0.000427;0;0.047426;0.39556;0.974694;0.00077;0.024536;0.981765;0.00227;0.015965;0.996365;0.00001;0.003625;0.907345;0.002525;0.00056;0.81602;0.90559;0.000965
0.000427;0.997965;0.974694;0.00077;0.00001;0.952574;0.42678;0.907345;0.002525;0.090128;0.94478;0.002605;0.052615;0.98166;0.00056;0.017775;0.806528;0.002708;0.000965;0.37423;0.168285;0.003255
0.00077;0.996365;0.907345;0.002525;0.00056;0.952574;0.149822;0.806528;0.002708;0.190764;0.81602;0.00275;0.18123;0.90559;0.000965;0.093445;0.25032;0.005192;0.003255;0.194985;0.056635;0.00331
0.509818;0.000275;0.05059;0.37672;0.54454;0.731059;0.282114;0.951709;0.024367;0.023924;0.980945;0.00005;0.019005;0.997815;0.00003;0.002155;0.95633;0.001666;0.000005;0.995795;0.999305;0
0.37672;0.00048;0.951709;0.024367;0.00003;0.047426;0.459837;0.95633;0.001666;0.042002;0.995415;0;0.004585;0.999065;0.000005;0.000925;0.976598;0.001666;0;0.99991;0.999895;0
0.024367;0.997815;0.95633;0.001666;0.000005;0.002473;0.037652;0.976598;0.001666;0.021734;0.995795;0.000005;0.004195;0.999305;0;0.000695;0.982954;0.003358;0;0.999945;0.999985;0
0.001666;0.999065;0.976598;0.001666;0;0.268941;0.284144;0.982954;0.003358;0.013688;0.99991;0.00002;0.00007;0.999895;0;0.000105;0.983102;0.002203;0;0.999865;0.999905;0
0.001666;0.999305;0.982954;0.003358;0;0.5;0.378246;0.983102;0.002203;0.01469;0.999945;0.00002;0.00003;0.999985;0;0.000005;0.975464;0.002224;0;0.999865;0.999855;0
0.003358;0.999895;0.983102;0.002203;0;0.268941;0.476268;0.975464;0.002224;0.02231;0.999865;0.00002;0.000115;0.999905;0;0.00009;0.969589;0.000018;0;0.998255;0.998425;0.00001
0.002203;0.999985;0.975464;0.002224;0;0.880797;0.693174;0.969589;0.000018;0.030394;0.999865;0.000025;0.000115;0.999855;0;0.000145;0.94434;0.002803;0.00001;0.99444;0.993315;0.000145
0.002224;0.999905;0.969589;0.000018;0;0.5;0.529217;0.94434;0.002803;0.052856;0.998255;0.000025;0.001715;0.998425;0.00001;0.001565;0.933727;0.006593;0.000145;0.99246;0.961665;0.0002
0.000018;0.999855;0.94434;0.002803;0.00001;0.047426;0.41096;0.933727;0.006593;0.059682;0.99444;0.000025;0.005535;0.993315;0.000145;0.006545;0.911364;0.002948;0.0002;0.495955;0.625965;0.00081
0.002803;0.998425;0.933727;0.006593;0.000145;0.880797;0.443246;0.911364;0.002948;0.085688;0.99246;0.00002;0.00752;0.961665;0.0002;0.038135;0.53995;0.000964;0.00081;0.000715;0.021055;0.000915
0.000021;0.977485;0.955565;0.000027;0.00002;0.731059;0.643136;0.981244;0.000015;0.018738;0.996045;0.00001;0.00394;0.998985;0;0.00101;0.980455;0.000014;0;0.997085;0.999905;0
0.000027;0.98628;0.981244;0.000015;0;0.268941;0.287614;0.980455;0.000014;0.019531;0.997015;0.00001;0.002975;0.999835;0;0.000165;0.977148;0.000026;0;0.9991;0.99988;0
0.000015;0.998985;0.980455;0.000014;0;0.268941;0.61822;0.977148;0.000026;0.022826;0.997085;0.000045;0.002875;0.999905;0;0.00009;0.985975;0.000036;0;0.998265;0.99845;0
0.000014;0.999835;0.977148;0.000026;0;0.880797;0.532952;0.985975;0.000036;0.013989;0.9991;0.000075;0.000825;0.99988;0;0.00012;0.979153;0.000048;0;0.99201;0.995045;0
0.000026;0.999905;0.985975;0.000036;0;0.119203;0.195761;0.979153;0.000048;0.020799;0.998265;0.00011;0.001625;0.99845;0;0.00155;0.965073;0.003846;0;0.962435;0.974465;0.000075
0.000036;0.99988;0.979153;0.000048;0;0.047426;0.537927;0.965073;0.003846;0.031079;0.99201;0.000105;0.00788;0.995045;0;0.004955;0.928092;0.000133;0.000075;0.90279;0.723185;0.01667
0.000048;0.99845;0.965073;0.003846;0;0.268941;0.536435;0.928092;0.000133;0.071776;0.962435;0.00029;0.03727;0.974465;0.000075;0.025465;0.789833;0.006586;0.01667;0.37088;0.27723;0.001685
0.003846;0.995045;0.928092;0.000133;0.000075;0.880797;0.675902;0.789833;0.006586;0.203578;0.90279;0.003055;0.09415;0.723185;0.01667;0.26014;0.251092;0.006111;0.001685;0.16336;0.06408;0.00334
0.655616;0.0014;0.028594;0.4833;0.96643;0.731059;0.123467;0.975345;0.001873;0.02278;0.989215;0.00001;0.010775;0.999055;0.00001;0.00093;0.974067;0.00476;0;0.99997;0.999835;0
0.4833;0.001525;0.975345;0.001873;0.00001;0.006693;0.393649;0.974067;0.00476;0.021172;0.999815;0;0.000185;0.999715;0;0.000285;0.976068;0.007189;0;0.999995;0.999965;0
0.001873;0.999055;0.974067;0.00476;0;0.006693;0.255213;0.976068;0.007189;0.016743;0.99997;0;0.00003;0.999835;0;0.000165;0.973135;0.001278;0;0.99999;0.999985;0
0.00476;0.999715;0.976068;0.007189;0;0.5;0.163693;0.973135;0.001278;0.025587;0.999995;0;0.000005;0.999965;0;0.000035;0.97794;0.002564;0;0.99996;0.99998;0
0.007189;0.999835;0.973135;0.001278;0;0.982014;0.511248;0.97794;0.002564;0.019495;0.99999;0;0.000005;0.999985;0;0.000015;0.978694;0.000011;0;0.999955;0.999965;0
0.001278;0.999965;0.97794;0.002564;0;0.268941;0.742691;0.978694;0.000011;0.021295;0.99996;0;0.00004;0.99998;0;0.00002;0.97844;0.000012;0;0.99834;0.99947;0
0.002564;0.999985;0.978694;0.000011;0;0.047426;0.529964;0.97844;0.000012;0.021546;0.999955;0;0.00004;0.999965;0;0.000035;0.949995;0.00235;0;0.9983;0.999175;0
0.000011;0.99998;0.97844;0.000012;0;0.880797;0.321475;0.949995;0.00235;0.047653;0.99834;0.000015;0.00164;0.99947;0;0.00053;0.937838;0.008506;0;0.99788;0.964215;0.000005
0.000012;0.999965;0.949995;0.00235;0;0.731059;0.579812;0.937838;0.008506;0.053654;0.9983;0.000015;0.00168;0.999175;0;0.000825;0.917462;0.003603;0.000005;0.995875;0.941665;0.00006
0.00235;0.99947;0.937838;0.008506;0;0.047426;0.528968;0.917462;0.003603;0.078933;0.99788;0.000095;0.00202;0.964215;0.000005;0.03578;0.88201;0.00159;0.00006;0.78624;0.905205;0.000245
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CF-tree data structure.jpg
Type: image/jpeg
Size: 54025 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170823/01071e01/attachment-0001.jpg>

From msuzen at gmail.com  Wed Aug 23 06:44:58 2017
From: msuzen at gmail.com (Suzen, Mehmet)
Date: Wed, 23 Aug 2017 12:44:58 +0200
Subject: [scikit-learn] Accessing Clustering Feature Tree in Birch
In-Reply-To: <CAAir+Cq+Nu6k9=B1YrNt9YbN8cTNQfV3m94SMsVbb16Fac3m1Q@mail.gmail.com>
References: <CAAir+Cq+Nu6k9=B1YrNt9YbN8cTNQfV3m94SMsVbb16Fac3m1Q@mail.gmail.com>
Message-ID: <CAPtbhHw07BWz4KsyQmzF3_xxJzh8mQ1zx7umEiesuxfitRthKw@mail.gmail.com>

Hi Sema,

You can access CFNode from the fit output, assign fit output, so you
can have the object.

brc_fit = brc.fit(X)
brc_fit_cfnode = brc_fit.root_
<sklearn.cluster.birch._CFNode object at 0x7ff31acbf668>

Then you can access CFNode, see here
https://kite.com/docs/python/sklearn.cluster.birch._CFNode

Also, this example comparing mini batch kmeans.
http://scikit-learn.org/stable/auto_examples/cluster/plot_birch_vs_minibatchkmeans.html

Hope this was what you are after.

Best,
Mehmet

On 23 August 2017 at 10:55, Sema Atasever <s.atasever at gmail.com> wrote:
> Dear scikit-learn members,
>
> Considering the "CF-tree" data structure :
>
> - How can i access Clustering Feature Tree in Birch?
>
> - For example, how many clusters are there in the hierarchy under the root
> node and what are the data samples in this cluster?
>
> - Can I get them separately for 3 trees?
>
> Best.
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>

From rth.yurchak at gmail.com  Wed Aug 23 07:28:16 2017
From: rth.yurchak at gmail.com (Roman Yurchak)
Date: Wed, 23 Aug 2017 14:28:16 +0300
Subject: [scikit-learn] Accessing Clustering Feature Tree in Birch
In-Reply-To: <CAPtbhHw07BWz4KsyQmzF3_xxJzh8mQ1zx7umEiesuxfitRthKw@mail.gmail.com>
References: <CAAir+Cq+Nu6k9=B1YrNt9YbN8cTNQfV3m94SMsVbb16Fac3m1Q@mail.gmail.com>
 <CAPtbhHw07BWz4KsyQmzF3_xxJzh8mQ1zx7umEiesuxfitRthKw@mail.gmail.com>
Message-ID: <af01fe8a-f6f4-a00b-9fc8-eb69a9462c5b@gmail.com>

 > what are the data samples in this cluster

Mehmet's response below works for exploring the hierarchical tree. 
However, Birch currently doesn't store the data samples that belong to a 
given subcluster. If you need that, as far as I know, a reasonable 
approximation can be obtained by computing the data samples that are 
closest to the centroid of the considered subcluster (accessible via 
_CFNode.centroids_) as compared to all other subcluster centroids at 
this hierarchical tree depth.

Alternatively, the modifications in PR 
https://github.com/scikit-learn/scikit-learn/pull/8808 aimed to make 
this process easier..
-- 
Roman

On 23/08/17 13:44, Suzen, Mehmet wrote:
> Hi Sema,
>
> You can access CFNode from the fit output, assign fit output, so you
> can have the object.
>
> brc_fit = brc.fit(X)
> brc_fit_cfnode = brc_fit.root_
> <sklearn.cluster.birch._CFNode object at 0x7ff31acbf668>
>
> Then you can access CFNode, see here
> https://kite.com/docs/python/sklearn.cluster.birch._CFNode
>
> Also, this example comparing mini batch kmeans.
> http://scikit-learn.org/stable/auto_examples/cluster/plot_birch_vs_minibatchkmeans.html
>
> Hope this was what you are after.
>
> Best,
> Mehmet
>
> On 23 August 2017 at 10:55, Sema Atasever <s.atasever at gmail.com> wrote:
>> Dear scikit-learn members,
>>
>> Considering the "CF-tree" data structure :
>>
>> - How can i access Clustering Feature Tree in Birch?
>>
>> - For example, how many clusters are there in the hierarchy under the root
>> node and what are the data samples in this cluster?
>>
>> - Can I get them separately for 3 trees?
>>
>> Best.
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>


From g.lemaitre58 at gmail.com  Thu Aug 24 20:14:08 2017
From: g.lemaitre58 at gmail.com (=?UTF-8?Q?Guillaume_Lema=C3=AEtre?=)
Date: Fri, 25 Aug 2017 02:14:08 +0200
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn 0.19.0
Message-ID: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>

We are excited to announce the new release of the scikit-learn-contrib
imbalanced-learn, already available through conda and pip (cf. the
installation page https://tinyurl.com/y92flbab for more info)

Notable add-ons are:

* Support of sparse matrices
* Support of multi-class resampling for all methods
* A new BalancedBaggingClassifier using random under-sampling chained with
the scikit-learn BaggingClassifier
* Creation of a didactic user guide
* New API of the ratio parameter to fit the needs of multi-class resampling
* Migration from nosetests to pytest

You can check the full changelog at:
http://contrib.scikit-learn.org/imbalanced-learn/stable/whats_new.html#version-0-3

A big thank you to contributors to use, raise issues, and submit PRs to
imblearn.
-- 
Guillaume Lemaitre
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170825/7e236707/attachment.html>

From francois.dion at gmail.com  Thu Aug 24 20:31:08 2017
From: francois.dion at gmail.com (Francois Dion)
Date: Thu, 24 Aug 2017 20:31:08 -0400
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
Message-ID: <20170825003108.5587025.54506.170474@gmail.com>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170824/2f8261a7/attachment.html>

From ashimb9 at gmail.com  Thu Aug 24 21:48:07 2017
From: ashimb9 at gmail.com (Ashim Bhattarai)
Date: Thu, 24 Aug 2017 20:48:07 -0500
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <20170825003108.5587025.54506.170474@gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
 <20170825003108.5587025.54506.170474@gmail.com>
Message-ID: <CAJhGTCUaj0OwFFM_Ys+CoM04zRfk1_NZ7WttjdEf495L1jJAbQ@mail.gmail.com>

Thanks a lot!

On Thu, Aug 24, 2017 at 7:31 PM, Francois Dion <francois.dion at gmail.com>
wrote:

> Merci beaucoup.? Super utile. J' ai d'ailleurs introduit ton module a la
> conference data intelligence a Capital One il y a moins de 2 mois ( en
> banlieue de Washington DC).
>
> Sent from my BlackBerry 10 Darkphone
> *From: *Guillaume Lema?tre
> *Sent: *Thursday, August 24, 2017 20:15
> *To: *Scikit-learn user and developer mailing list
> *Reply To: *Scikit-learn mailing list
> *Subject: *[scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
> 0.19.0
>
> We are excited to announce the new release of the scikit-learn-contrib
> imbalanced-learn, already available through conda and pip (cf. the
> installation page https://tinyurl.com/y92flbab for more info)
>
> Notable add-ons are:
>
> * Support of sparse matrices
> * Support of multi-class resampling for all methods
> * A new BalancedBaggingClassifier using random under-sampling chained with
> the scikit-learn BaggingClassifier
> * Creation of a didactic user guide
> * New API of the ratio parameter to fit the needs of multi-class resampling
> * Migration from nosetests to pytest
>
> You can check the full changelog at:
> http://contrib.scikit-learn.org/imbalanced-learn/stable/
> whats_new.html#version-0-3
>
> A big thank you to contributors to use, raise issues, and submit PRs to
> imblearn.
> --
> Guillaume Lemaitre
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170824/ef6bb0ad/attachment.html>

From joel.nothman at gmail.com  Fri Aug 25 00:13:26 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Fri, 25 Aug 2017 14:13:26 +1000
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
Message-ID: <CAAkaFLVpdpY6vVxfHL-TOwAOUTdfPuD1qAN3DG2k6Yf4=OK7pg@mail.gmail.com>

Congratulations Guillaume and the imblearn team!

On 25 August 2017 at 10:14, Guillaume Lema?tre <g.lemaitre58 at gmail.com>
wrote:

> We are excited to announce the new release of the scikit-learn-contrib
> imbalanced-learn, already available through conda and pip (cf. the
> installation page https://tinyurl.com/y92flbab for more info)
>
> Notable add-ons are:
>
> * Support of sparse matrices
> * Support of multi-class resampling for all methods
> * A new BalancedBaggingClassifier using random under-sampling chained with
> the scikit-learn BaggingClassifier
> * Creation of a didactic user guide
> * New API of the ratio parameter to fit the needs of multi-class resampling
> * Migration from nosetests to pytest
>
> You can check the full changelog at:
> http://contrib.scikit-learn.org/imbalanced-learn/stable/
> whats_new.html#version-0-3
>
> A big thank you to contributors to use, raise issues, and submit PRs to
> imblearn.
> --
> Guillaume Lemaitre
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170825/c82bc04c/attachment-0001.html>

From gael.varoquaux at normalesup.org  Fri Aug 25 01:52:01 2017
From: gael.varoquaux at normalesup.org (Gael Varoquaux)
Date: Fri, 25 Aug 2017 07:52:01 +0200
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <CAAkaFLVpdpY6vVxfHL-TOwAOUTdfPuD1qAN3DG2k6Yf4=OK7pg@mail.gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
 <CAAkaFLVpdpY6vVxfHL-TOwAOUTdfPuD1qAN3DG2k6Yf4=OK7pg@mail.gmail.com>
Message-ID: <20170825055201.GQ153900@phare.normalesup.org>

Indeed, congratulations for the release!

Ga?l

On Fri, Aug 25, 2017 at 02:13:26PM +1000, Joel Nothman wrote:
> Congratulations Guillaume and the imblearn team!

> On 25 August 2017 at 10:14, Guillaume Lema?tre <g.lemaitre58 at gmail.com> wrote:

>     We are excited to announce the new release of the scikit-learn-contrib
>     imbalanced-learn, already available through conda and pip (cf. the
>     installation page https://tinyurl.com/y92flbab for more info)

>     Notable add-ons are:

>     * Support of sparse matrices
>     * Support of multi-class resampling for all methods
>     * A new BalancedBaggingClassifier using random under-sampling chained with
>     the scikit-learn BaggingClassifier
>     * Creation of a didactic user guide
>     * New API of the ratio parameter to fit the needs of multi-class resampling
>     * Migration from nosetests to pytest

>     You can check the full changelog at:
>     http://contrib.scikit-learn.org/imbalanced-learn/stable/whats_new.html#
>     version-0-3

>     A big thank you to contributors to use, raise issues, and submit PRs to
>     imblearn.
-- 
    Gael Varoquaux
    Researcher, INRIA Parietal
    NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
    Phone:  ++ 33-1-69-08-79-68
    http://gael-varoquaux.info            http://twitter.com/GaelVaroquaux

From bertrand.thirion at inria.fr  Fri Aug 25 01:56:28 2017
From: bertrand.thirion at inria.fr (bthirion)
Date: Fri, 25 Aug 2017 07:56:28 +0200
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <20170825055201.GQ153900@phare.normalesup.org>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
 <CAAkaFLVpdpY6vVxfHL-TOwAOUTdfPuD1qAN3DG2k6Yf4=OK7pg@mail.gmail.com>
 <20170825055201.GQ153900@phare.normalesup.org>
Message-ID: <71818780-5165-7545-3eea-1e398780be9a@inria.fr>

+1 B

On 25/08/2017 07:52, Gael Varoquaux wrote:
> Indeed, congratulations for the release!
>
> Ga?l
>
> On Fri, Aug 25, 2017 at 02:13:26PM +1000, Joel Nothman wrote:
>> Congratulations Guillaume and the imblearn team!
>> On 25 August 2017 at 10:14, Guillaume Lema?tre <g.lemaitre58 at gmail.com> wrote:
>>      We are excited to announce the new release of the scikit-learn-contrib
>>      imbalanced-learn, already available through conda and pip (cf. the
>>      installation page https://tinyurl.com/y92flbab for more info)
>>      Notable add-ons are:
>>      * Support of sparse matrices
>>      * Support of multi-class resampling for all methods
>>      * A new BalancedBaggingClassifier using random under-sampling chained with
>>      the scikit-learn BaggingClassifier
>>      * Creation of a didactic user guide
>>      * New API of the ratio parameter to fit the needs of multi-class resampling
>>      * Migration from nosetests to pytest
>>      You can check the full changelog at:
>>      http://contrib.scikit-learn.org/imbalanced-learn/stable/whats_new.html#
>>      version-0-3
>>      A big thank you to contributors to use, raise issues, and submit PRs to
>>      imblearn.



From se.raschka at gmail.com  Fri Aug 25 02:18:49 2017
From: se.raschka at gmail.com (Sebastian Raschka)
Date: Fri, 25 Aug 2017 02:18:49 -0400
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
Message-ID: <07BF3B06-5970-42B6-9148-0D604FE5921F@gmail.com>

Just read through the summary of the new features and browsed through the user guide. The guide is really well structured and easy to navigate, thanks for putting all the work into it. Overall, thanks for this great contribution and new version :)

Best,
Sebastian

> On Aug 24, 2017, at 8:14 PM, Guillaume Lema?tre <g.lemaitre58 at gmail.com> wrote:
> 
> We are excited to announce the new release of the scikit-learn-contrib imbalanced-learn, already available through conda and pip (cf. the installation page https://tinyurl.com/y92flbab for more info)
> 
> Notable add-ons are:
> 
> * Support of sparse matrices
> * Support of multi-class resampling for all methods
> * A new BalancedBaggingClassifier using random under-sampling chained with the scikit-learn BaggingClassifier
> * Creation of a didactic user guide
> * New API of the ratio parameter to fit the needs of multi-class resampling
> * Migration from nosetests to pytest
> 
> You can check the full changelog at:
> http://contrib.scikit-learn.org/imbalanced-learn/stable/whats_new.html#version-0-3
> 
> A big thank you to contributors to use, raise issues, and submit PRs to imblearn.
> -- 
> Guillaume Lemaitre
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn


From jaquesgrobler at gmail.com  Fri Aug 25 04:53:33 2017
From: jaquesgrobler at gmail.com (Jaques Grobler)
Date: Fri, 25 Aug 2017 10:53:33 +0200
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <07BF3B06-5970-42B6-9148-0D604FE5921F@gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
 <07BF3B06-5970-42B6-9148-0D604FE5921F@gmail.com>
Message-ID: <CAHcSOR=qRfwpMot_oQBH34HqKTC4fojTte-eom2sGdiFqjXjmg@mail.gmail.com>

Congrats guys!

2017-08-25 8:18 GMT+02:00 Sebastian Raschka <se.raschka at gmail.com>:

> Just read through the summary of the new features and browsed through the
> user guide. The guide is really well structured and easy to navigate,
> thanks for putting all the work into it. Overall, thanks for this great
> contribution and new version :)
>
> Best,
> Sebastian
>
> > On Aug 24, 2017, at 8:14 PM, Guillaume Lema?tre <g.lemaitre58 at gmail.com>
> wrote:
> >
> > We are excited to announce the new release of the scikit-learn-contrib
> imbalanced-learn, already available through conda and pip (cf. the
> installation page https://tinyurl.com/y92flbab for more info)
> >
> > Notable add-ons are:
> >
> > * Support of sparse matrices
> > * Support of multi-class resampling for all methods
> > * A new BalancedBaggingClassifier using random under-sampling chained
> with the scikit-learn BaggingClassifier
> > * Creation of a didactic user guide
> > * New API of the ratio parameter to fit the needs of multi-class
> resampling
> > * Migration from nosetests to pytest
> >
> > You can check the full changelog at:
> > http://contrib.scikit-learn.org/imbalanced-learn/stable/
> whats_new.html#version-0-3
> >
> > A big thank you to contributors to use, raise issues, and submit PRs to
> imblearn.
> > --
> > Guillaume Lemaitre
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn at python.org
> > https://mail.python.org/mailman/listinfo/scikit-learn
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170825/4f85210c/attachment.html>

From jbbrown at kuhp.kyoto-u.ac.jp  Fri Aug 25 05:09:37 2017
From: jbbrown at kuhp.kyoto-u.ac.jp (Brown J.B.)
Date: Fri, 25 Aug 2017 11:09:37 +0200
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <CAHcSOR=qRfwpMot_oQBH34HqKTC4fojTte-eom2sGdiFqjXjmg@mail.gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
 <07BF3B06-5970-42B6-9148-0D604FE5921F@gmail.com>
 <CAHcSOR=qRfwpMot_oQBH34HqKTC4fojTte-eom2sGdiFqjXjmg@mail.gmail.com>
Message-ID: <CAJe_vxAQXBKVVgR81kFY25xggv7NK-QEdQv8o9fTMsyvO03Hcg@mail.gmail.com>

In drug discovery, if you are lucky you might get hit compounds 10% of the
time.
So if you do ML-based drug discovery, your datasets are strongly imbalanced.
It seems the imbalanced package would be perfect for this area.

J.B.

2017-08-25 10:53 GMT+02:00 Jaques Grobler <jaquesgrobler at gmail.com>:

> Congrats guys!
>
> 2017-08-25 8:18 GMT+02:00 Sebastian Raschka <se.raschka at gmail.com>:
>
>> Just read through the summary of the new features and browsed through the
>> user guide. The guide is really well structured and easy to navigate,
>> thanks for putting all the work into it. Overall, thanks for this great
>> contribution and new version :)
>>
>> Best,
>> Sebastian
>>
>> > On Aug 24, 2017, at 8:14 PM, Guillaume Lema?tre <g.lemaitre58 at gmail.com>
>> wrote:
>> >
>> > We are excited to announce the new release of the scikit-learn-contrib
>> imbalanced-learn, already available through conda and pip (cf. the
>> installation page https://tinyurl.com/y92flbab for more info)
>> >
>> > Notable add-ons are:
>> >
>> > * Support of sparse matrices
>> > * Support of multi-class resampling for all methods
>> > * A new BalancedBaggingClassifier using random under-sampling chained
>> with the scikit-learn BaggingClassifier
>> > * Creation of a didactic user guide
>> > * New API of the ratio parameter to fit the needs of multi-class
>> resampling
>> > * Migration from nosetests to pytest
>> >
>> > You can check the full changelog at:
>> > http://contrib.scikit-learn.org/imbalanced-learn/stable/what
>> s_new.html#version-0-3
>> >
>> > A big thank you to contributors to use, raise issues, and submit PRs to
>> imblearn.
>> > --
>> > Guillaume Lemaitre
>> > _______________________________________________
>> > scikit-learn mailing list
>> > scikit-learn at python.org
>> > https://mail.python.org/mailman/listinfo/scikit-learn
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170825/8372f520/attachment-0001.html>

From mario.michael.krell at gmail.com  Fri Aug 25 11:40:35 2017
From: mario.michael.krell at gmail.com (Dr. Mario Michael Krell)
Date: Fri, 25 Aug 2017 08:40:35 -0700
Subject: [scikit-learn] L-BFGS in MLPClassifier
Message-ID: <05E76DEC-1286-4A87-BA27-B48F786BD89C@gmail.com>

To whoever programmed the MLPClassifier (with the L-BFGS solver),

I just wanted to personally thank you and if I get your name(s), I would mention it/them in my paper additionally to the mandatory sklearn citation.

I hope that sklearn will be keeping this algorithm forever in their library despite the increasing amount of established deep learning libraries that seem to make this code obsolete. For my small scale, more theoretic analysis, it worked much better than any other algorithm and I would not have gotten such surprising results. Due to the high quality implementation, the integration of a much better solver than SGD, and the respective good documentation, I could show empirically how the VC dimension and another property of MLPs (MacKay dimension) actually scale linear with the number of edges in the respective graph which helped us to provide a new much more strict upper bound (https://arxiv.org/abs/1708.06019 <https://arxiv.org/abs/1708.06019>). This would have not been possible with other implementations. If there is an interest by the developers, I could try to contribute a tutorial documentation for sklearn. Just let me know.

Thank you a lot!!!

Best,

Mario
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170825/5a7515e6/attachment.html>

From soneill5045 at gmail.com  Fri Aug 25 19:30:52 2017
From: soneill5045 at gmail.com (Stephen O'Neill)
Date: Fri, 25 Aug 2017 19:30:52 -0400
Subject: [scikit-learn] different sized inputs in call to custom metric in
 KNN
Message-ID: <CAEws48zX7yZ9ijq2xQWZaaUgqxawq_T19joNBDmjaXSKTipz4w@mail.gmail.com>

Hey Gang,

I was wondering if anyone might be able to answer a question about the
sklearn,neighbors.NearestNeighbors class.  For reference, I'm on:
anaconda distribution python 2.7.11
sklearn version 0.17.1

I'm subclassing the NearestNeighbors class and using a custom distance
metric, something like the following:

class MyModel(NearestNeighbors):
    def __init__(self, some_info):
        def custom_dist(x, y, info=some_info):
            return numpy.sum(numpy.abs(x - y)/some_info)

            return scalar_value

        NearestNeighbors.__init__(self, metric=custom_dist)


So I build a dummy dataset based on some gaussians of shape (5000,3), then
later when I call

MyModel.fit()

I get the error:
"ValueError: operands could not be broadcast together with shapes (10,)
(3,)"


inside of my custom_dist function.  Naturally I checked with some simple
print x, print y statements inside of custom_dist, and sure enough the
shapes of x and y are both (10,), whereas I am expecting them to be of
shape (3,) since my dummy data has 3 columns. (Note the actual custom_dist
function written above is not what I'm truly using but it does reproduce
the same ValueError).

When I change my NearestNeighbors.__init__ call to (self,
algorithm='brute') instead of the default algorithm='auto', the x and y
values that get passed to my custom_dist are shape (3,) like I would
expect.  What is different in the distance metric between the
algorithm='auto' and algorithm='brute' that would transform a 3-dimensional
sample to a 10-dimensional sample?  Do the KDTree and/or BallTree classes
use the distance metric on tree nodes or something too?  I wasn't able to
figure out where the shape (10,) x and y samples could be coming from.

Thanks in advance!

Best,
Steve O'Neill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170825/f13984a1/attachment.html>

From mathieu at mblondel.org  Fri Aug 25 21:33:06 2017
From: mathieu at mblondel.org (Mathieu Blondel)
Date: Sat, 26 Aug 2017 10:33:06 +0900
Subject: [scikit-learn] L-BFGS in MLPClassifier
In-Reply-To: <05E76DEC-1286-4A87-BA27-B48F786BD89C@gmail.com>
References: <05E76DEC-1286-4A87-BA27-B48F786BD89C@gmail.com>
Message-ID: <CAOKSrLwswrb+_cswKWzGVvAWoQJNMHSC3Q4MyoVDOCVP4EAnPA@mail.gmail.com>

Thanks for this email. It is always nice to hear about success stories.

I assume the guilty party is Issam Laradji, as you can see from his Google
Summer of Code blog post:

http://issamlaradji.blogspot.jp/2014/06/week-3-gsoc-2014-extending-neural.html

L-BFGS is indeed usually a good default choice for medium-scale datasets.
It doesn't require any step size tuning and I found recently that it works
well for poorly conditioned problems.

You can also see a blog post by Nicolas Le Roux praising L-BFGS here:

http://labs.criteo.com/2014/09/poh-part-3-distributed-optimization/

Mathieu


On Sat, Aug 26, 2017 at 12:40 AM, Dr. Mario Michael Krell <
mario.michael.krell at gmail.com> wrote:

> To whoever programmed the MLPClassifier (with the L-BFGS solver),
>
> I just wanted to personally thank you and if I get your name(s), I would
> mention it/them in my paper additionally to the mandatory sklearn citation.
>
> I hope that sklearn will be keeping this algorithm forever in their
> library despite the increasing amount of established deep learning
> libraries that seem to make this code obsolete. For my small scale, more
> theoretic analysis, it worked much better than any other algorithm and I
> would not have gotten such surprising results. Due to the high quality
> implementation, the integration of a much better solver than SGD, and the
> respective good documentation, I could show empirically how the VC
> dimension and another property of MLPs (MacKay dimension) actually scale
> linear with the number of edges in the respective graph which helped us to
> provide a new much more strict upper bound (https://arxiv.org/abs/1708.
> 06019). This would have not been possible with other implementations. If
> there is an interest by the developers, I could try to contribute a
> tutorial documentation for sklearn. Just let me know.
>
> Thank you a lot!!!
>
> Best,
>
> Mario
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170826/6e9e886f/attachment.html>

From shabieiqbal at gmail.com  Sun Aug 27 14:48:42 2017
From: shabieiqbal at gmail.com (Shabie Iqbal)
Date: Sun, 27 Aug 2017 20:48:42 +0200
Subject: [scikit-learn] Clarification regarding SGDClassifier
Message-ID: <59a31409.c35c1c0a.5c153.9b50@mx.google.com>

Dear all,

could anyone of you kindly answer this question I posted on Stackoverflow:

https://stackoverflow.com/questions/45900330/sklearn-sgdclassifiers-decision-function-odd-behavior

Best,
Shabie
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170827/73d94a6d/attachment.html>

From rdslater at gmail.com  Sun Aug 27 18:21:38 2017
From: rdslater at gmail.com (Robert Slater)
Date: Sun, 27 Aug 2017 17:21:38 -0500
Subject: [scikit-learn] Clarification regarding SGDClassifier
In-Reply-To: <59a31409.c35c1c0a.5c153.9b50@mx.google.com>
References: <59a31409.c35c1c0a.5c153.9b50@mx.google.com>
Message-ID: <CAMt686YeczLz0B9DyWM8e=Hg0RmK+TtW47d7JMxcf9m3TKJ3MA@mail.gmail.com>

I think your use of "X" in X_mod should be "X_train" instead, as X_mod is
currently using unshuffled indices while X_check uses the shuffled indices.

<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon>
Virus-free.
www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Sun, Aug 27, 2017 at 1:48 PM, Shabie Iqbal <shabieiqbal at gmail.com> wrote:

> Dear all,
>
>
>
> could anyone of you kindly answer this question I posted on Stackoverflow:
>
>
>
> https://stackoverflow.com/questions/45900330/sklearn-
> sgdclassifiers-decision-function-odd-behavior
>
>
>
> Best,
>
> Shabie
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170827/8046e5cf/attachment.html>

From stuart at stuartreynolds.net  Sun Aug 27 18:23:45 2017
From: stuart at stuartreynolds.net (Stuart Reynolds)
Date: Sun, 27 Aug 2017 15:23:45 -0700
Subject: [scikit-learn] Decision stubs?
Message-ID: <CAAy-kdkCBh27fB_gY0cHfhVzddHcpaaT3+12Jq+N+d-ttbqoVw@mail.gmail.com>

Is it possible to efficiently get at the branch statistics that
decision tree algorithms iterate over in scikit?

For example if the root population has the class counts in the output vector:
   c0: 5000
   c1: 500

Then I'd like to iterate over:
# For a boolean (2 valued category)
   f1=True:      c0=3000,  c1=450
   f1=False:    c0=300,    c1=30
   f1=Null:       c0=1700,  c1=20  # ? Is considered?

# For a continuous value
   f2<10:         c0= ...  c1= ...
   f2>=10:         c0= ...  c1= ...

   f2<22:         c0= ...  c1= ...
   f2>=22:         c0= ...  c1= ...


I'd like to experiment with building models on-demand for each input
row in a predict.
To work efficiently, I'd like to reduce the training set to the 'most
significant' sub-space(s) using the population statistics.

I can do it in pandas, although its fairly inefficient to iterate over
each feature column many times.

Thanks,
- Stu

From shabieiqbal at gmail.com  Sun Aug 27 18:41:01 2017
From: shabieiqbal at gmail.com (Shabie Iqbal)
Date: Mon, 28 Aug 2017 00:41:01 +0200
Subject: [scikit-learn] Clarification regarding SGDClassifier
In-Reply-To: <CAMt686YeczLz0B9DyWM8e=Hg0RmK+TtW47d7JMxcf9m3TKJ3MA@mail.gmail.com>
References: <59a31409.c35c1c0a.5c153.9b50@mx.google.com>
 <CAMt686YeczLz0B9DyWM8e=Hg0RmK+TtW47d7JMxcf9m3TKJ3MA@mail.gmail.com>
Message-ID: <59a34a7d.53e81c0a.de71a.06b0@mx.google.com>

Got it? Thanks!

Sent from Mail for Windows 10

From: Robert Slater
Sent: Monday, August 28, 2017 12:24 AM
To: Scikit-learn mailing list
Subject: Re: [scikit-learn] Clarification regarding SGDClassifier

I think your use of "X" in X_mod should be "X_train" instead, as X_mod is currently using unshuffled indices while X_check uses the shuffled indices.


Virus-free. www.avast.com 
On Sun, Aug 27, 2017 at 1:48 PM, Shabie Iqbal <shabieiqbal at gmail.com> wrote:
Dear all,
?
could anyone of you kindly answer this question I posted on Stackoverflow:
?
https://stackoverflow.com/questions/45900330/sklearn-sgdclassifiers-decision-function-odd-behavior
?
Best,
Shabie

_______________________________________________
scikit-learn mailing list
scikit-learn at python.org
https://mail.python.org/mailman/listinfo/scikit-learn


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/00ac22ab/attachment-0001.html>

From raga.markely at gmail.com  Sun Aug 27 23:20:09 2017
From: raga.markely at gmail.com (Raga Markely)
Date: Sun, 27 Aug 2017 23:20:09 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic regression
 from a pipeline
Message-ID: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>

Hello,

I am wondering if it's possible to get the weight coefficients of logistic
regression from a pipeline?

For instance, I have the followings:

> clf_lr = LogisticRegression(penalty='l1', C=0.1)
> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
> pipe_lr.fit(X, y)


Does pipe_lr have an attribute that I can call to get the weight
coefficient?

Or do I have to get it from the classifier as follows?

> X_std = StandardScaler().fit_transform(X)
> clf_lr = LogisticRegression(penalty='l1', C=0.1)
> clf_lr.fit(X_std, y)
> clf_lr.coef_


Thank you,
Raga
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170827/a9fc90e6/attachment.html>

From joel.nothman at gmail.com  Mon Aug 28 00:01:40 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Mon, 28 Aug 2017 14:01:40 +1000
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
Message-ID: <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>

No, we do not have a way to get the coefficients with respect to the input
(pre-scaling) space.

On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com> wrote:

> Hello,
>
> I am wondering if it's possible to get the weight coefficients of logistic
> regression from a pipeline?
>
> For instance, I have the followings:
>
>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
>> pipe_lr.fit(X, y)
>
>
> Does pipe_lr have an attribute that I can call to get the weight
> coefficient?
>
> Or do I have to get it from the classifier as follows?
>
>> X_std = StandardScaler().fit_transform(X)
>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>> clf_lr.fit(X_std, y)
>> clf_lr.coef_
>
>
> Thank you,
> Raga
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/09d5a131/attachment.html>

From raga.markely at gmail.com  Mon Aug 28 00:08:43 2017
From: raga.markely at gmail.com (Raga Markely)
Date: Mon, 28 Aug 2017 00:08:43 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
Message-ID: <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>

No problem, thank you!

Best,
Raga

On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman <joel.nothman at gmail.com>
wrote:

> No, we do not have a way to get the coefficients with respect to the input
> (pre-scaling) space.
>
> On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com> wrote:
>
>> Hello,
>>
>> I am wondering if it's possible to get the weight coefficients of
>> logistic regression from a pipeline?
>>
>> For instance, I have the followings:
>>
>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
>>> pipe_lr.fit(X, y)
>>
>>
>> Does pipe_lr have an attribute that I can call to get the weight
>> coefficient?
>>
>> Or do I have to get it from the classifier as follows?
>>
>>> X_std = StandardScaler().fit_transform(X)
>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>> clf_lr.fit(X_std, y)
>>> clf_lr.coef_
>>
>>
>> Thank you,
>> Raga
>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/506025e7/attachment.html>

From ichkoar at gmail.com  Mon Aug 28 08:37:14 2017
From: ichkoar at gmail.com (Christos Aridas)
Date: Mon, 28 Aug 2017 15:37:14 +0300
Subject: [scikit-learn] imbalanced-learn 0.3.0 is chasing scikit-learn
 0.19.0
In-Reply-To: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
References: <CACDxx9gZekCTOFL_C=0pWsxJXkrZwCQqGepoWo5174ReKN9bsg@mail.gmail.com>
Message-ID: <CAJBJ-XHXc+HJP_9RFUwBM5d19e4c7Mu7nTjEOp2YhSaLHyuwiQ@mail.gmail.com>

Well done guys! Thanks a lot for this great release! I hope to be back soon.

Best,
Chris

On Fri, Aug 25, 2017 at 3:14 AM, Guillaume Lema?tre <g.lemaitre58 at gmail.com>
wrote:

> We are excited to announce the new release of the scikit-learn-contrib
> imbalanced-learn, already available through conda and pip (cf. the
> installation page https://tinyurl.com/y92flbab for more info)
>
> Notable add-ons are:
>
> * Support of sparse matrices
> * Support of multi-class resampling for all methods
> * A new BalancedBaggingClassifier using random under-sampling chained with
> the scikit-learn BaggingClassifier
> * Creation of a didactic user guide
> * New API of the ratio parameter to fit the needs of multi-class resampling
> * Migration from nosetests to pytest
>
> You can check the full changelog at:
> http://contrib.scikit-learn.org/imbalanced-learn/stable/
> whats_new.html#version-0-3
>
> A big thank you to contributors to use, raise issues, and submit PRs to
> imblearn.
> --
> Guillaume Lemaitre
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/7af6d43e/attachment-0001.html>

From t3kcit at gmail.com  Mon Aug 28 12:01:31 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Mon, 28 Aug 2017 12:01:31 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
Message-ID: <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>

Can can get the coefficients on the scaled data with
pipeline_lr.named_steps_['clf'].coef_
though


On 08/28/2017 12:08 AM, Raga Markely wrote:
> No problem, thank you!
>
> Best,
> Raga
>
> On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman <joel.nothman at gmail.com 
> <mailto:joel.nothman at gmail.com>> wrote:
>
>     No, we do not have a way to get the coefficients with respect to
>     the input (pre-scaling) space.
>
>     On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com
>     <mailto:raga.markely at gmail.com>> wrote:
>
>         Hello,
>
>         I am wondering if it's possible to get the weight coefficients
>         of logistic regression from a pipeline?
>
>         For instance, I have the followings:
>
>             clf_lr = LogisticRegression(penalty='l1', C=0.1)
>             pipe_lr = Pipeline([['sc', StandardScaler()], ['clf',
>             clf_lr]])
>             pipe_lr.fit(X, y)
>
>
>         Does pipe_lr have an attribute that I can call to get the
>         weight coefficient?
>
>         Or do I have to get it from the classifier as follows?
>
>             X_std = StandardScaler().fit_transform(X)
>             clf_lr = LogisticRegression(penalty='l1', C=0.1)
>             clf_lr.fit(X_std, y)
>             clf_lr.coef_
>
>
>         Thank you,
>         Raga
>
>
>         _______________________________________________
>         scikit-learn mailing list
>         scikit-learn at python.org <mailto:scikit-learn at python.org>
>         https://mail.python.org/mailman/listinfo/scikit-learn
>         <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
>
>     _______________________________________________
>     scikit-learn mailing list
>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>     https://mail.python.org/mailman/listinfo/scikit-learn
>     <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/1ba5a634/attachment.html>

From t3kcit at gmail.com  Mon Aug 28 13:16:19 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Mon, 28 Aug 2017 13:16:19 -0400
Subject: [scikit-learn] scikit-learn-commits mailing list defunct?
Message-ID: <465fa212-0fb6-cfe2-2281-416c672429ca@gmail.com>

Hey all.

Is it just me or is the scikit-learn-commits mailing list no longer working?
Given that it's still on sourceforge, that seems somewhat likely.
I find the mailing list helpful in case I can't keep track of the issue 
tracker (i.e. for the last 3 years?).

It looks like it was set up here:
https://github.com/scikit-learn/scikit-learn/settings/hooks/28838

I propose we transition this to either a google group or another 
python.org mailing list (they already gave us 2 ;).

Cheers,
Andy

From raga.markely at gmail.com  Mon Aug 28 14:12:05 2017
From: raga.markely at gmail.com (Raga Markely)
Date: Mon, 28 Aug 2017 14:12:05 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
 <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
Message-ID: <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>

Thank you, Andreas.

When I try

> pipe_lr.named_steps['clf'].coef_


I get:

> AttributeError: 'LogisticRegression' object has no attribute 'coef_'


And when I try:

> pipe_lr.named_steps['clf']


I get:

> LogisticRegression(C=0.1, class_weight=None, dual=False,
> fit_intercept=True, intercept_scaling=1, max_iter=100, multi_class='ovr',
> n_jobs=1, penalty='l2', random_state=None, solver='liblinear', tol=0.0001,
> verbose=0, warm_start=False)


I wonder what I am missing?

Thanks,
Raga


On Mon, Aug 28, 2017 at 12:01 PM, Andreas Mueller <t3kcit at gmail.com> wrote:

> Can can get the coefficients on the scaled data with
> pipeline_lr.named_steps_['clf'].coef_
> though
>
>
> On 08/28/2017 12:08 AM, Raga Markely wrote:
>
> No problem, thank you!
>
> Best,
> Raga
>
> On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman <joel.nothman at gmail.com>
> wrote:
>
>> No, we do not have a way to get the coefficients with respect to the
>> input (pre-scaling) space.
>>
>> On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com> wrote:
>>
>>> Hello,
>>>
>>> I am wondering if it's possible to get the weight coefficients of
>>> logistic regression from a pipeline?
>>>
>>> For instance, I have the followings:
>>>
>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
>>>> pipe_lr.fit(X, y)
>>>
>>>
>>> Does pipe_lr have an attribute that I can call to get the weight
>>> coefficient?
>>>
>>> Or do I have to get it from the classifier as follows?
>>>
>>>> X_std = StandardScaler().fit_transform(X)
>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>> clf_lr.fit(X_std, y)
>>>> clf_lr.coef_
>>>
>>>
>>> Thank you,
>>> Raga
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
>
> _______________________________________________
> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/719eb738/attachment-0001.html>

From susan_liu at brown.edu  Mon Aug 28 14:17:04 2017
From: susan_liu at brown.edu (Liu, Susan)
Date: Mon, 28 Aug 2017 14:17:04 -0400
Subject: [scikit-learn] remoe from list
Message-ID: <CANB1q+Ha5n+FXq2G0C8uabD2jqSZ68WmM6Op9WM1Na1hhkiqoA@mail.gmail.com>

hi there,

just wanted to ask if i could be removed from list?


Thanks,
Susan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/3731034d/attachment.html>

From alekhka at gmail.com  Mon Aug 28 14:19:57 2017
From: alekhka at gmail.com (Alekh Karkada Ashok)
Date: Mon, 28 Aug 2017 23:49:57 +0530
Subject: [scikit-learn] remoe from list
In-Reply-To: <CANB1q+Ha5n+FXq2G0C8uabD2jqSZ68WmM6Op9WM1Na1hhkiqoA@mail.gmail.com>
References: <CANB1q+Ha5n+FXq2G0C8uabD2jqSZ68WmM6Op9WM1Na1hhkiqoA@mail.gmail.com>
Message-ID: <CAPTvW4McpFd5CCK8oajXzOwJ-0YahSNWe9kCp1NS34PYgaO0bg@mail.gmail.com>

Hi Susan,

You can visit https://mail.python.org/mailman/listinfo/scikit-learn and
unsubscribe from the list there.

Thanks,
Alekh

On Mon, Aug 28, 2017 at 11:47 PM, Liu, Susan <susan_liu at brown.edu> wrote:

> hi there,
>
> just wanted to ask if i could be removed from list?
>
>
> Thanks,
> Susan
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/f4e625f5/attachment.html>

From t3kcit at gmail.com  Mon Aug 28 14:55:08 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Mon, 28 Aug 2017 14:55:08 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
 <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
 <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>
Message-ID: <b4587533-b806-7bce-8fde-9aa19c0cc0d2@gmail.com>

Have you called "fit" on the pipeline?

On 08/28/2017 02:12 PM, Raga Markely wrote:
> Thank you, Andreas.
>
> When I try
>
>     pipe_lr.named_steps['clf'].coef_
>
>
> I get:
>
>     AttributeError: 'LogisticRegression' object has no attribute 'coef_'
>
>
> And when I try:
>
>     pipe_lr.named_steps['clf']
>
>
> I get:
>
>     LogisticRegression(C=0.1, class_weight=None, dual=False,
>     fit_intercept=True, intercept_scaling=1, max_iter=100,
>     multi_class='ovr', n_jobs=1, penalty='l2', random_state=None,
>     solver='liblinear', tol=0.0001, verbose=0, warm_start=False)
>
>
> I wonder what I am missing?
>
> Thanks,
> Raga
>
>
> On Mon, Aug 28, 2017 at 12:01 PM, Andreas Mueller <t3kcit at gmail.com 
> <mailto:t3kcit at gmail.com>> wrote:
>
>     Can can get the coefficients on the scaled data with
>     pipeline_lr.named_steps_['clf'].coef_
>     though
>
>
>     On 08/28/2017 12:08 AM, Raga Markely wrote:
>>     No problem, thank you!
>>
>>     Best,
>>     Raga
>>
>>     On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman
>>     <joel.nothman at gmail.com <mailto:joel.nothman at gmail.com>> wrote:
>>
>>         No, we do not have a way to get the coefficients with respect
>>         to the input (pre-scaling) space.
>>
>>         On 28 August 2017 at 13:20, Raga Markely
>>         <raga.markely at gmail.com <mailto:raga.markely at gmail.com>> wrote:
>>
>>             Hello,
>>
>>             I am wondering if it's possible to get the weight
>>             coefficients of logistic regression from a pipeline?
>>
>>             For instance, I have the followings:
>>
>>                 clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>                 pipe_lr = Pipeline([['sc', StandardScaler()], ['clf',
>>                 clf_lr]])
>>                 pipe_lr.fit(X, y)
>>
>>
>>             Does pipe_lr have an attribute that I can call to get the
>>             weight coefficient?
>>
>>             Or do I have to get it from the classifier as follows?
>>
>>                 X_std = StandardScaler().fit_transform(X)
>>                 clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>                 clf_lr.fit(X_std, y)
>>                 clf_lr.coef_
>>
>>
>>             Thank you,
>>             Raga
>>
>>
>>             _______________________________________________
>>             scikit-learn mailing list
>>             scikit-learn at python.org <mailto:scikit-learn at python.org>
>>             https://mail.python.org/mailman/listinfo/scikit-learn
>>             <https://mail.python.org/mailman/listinfo/scikit-learn>
>>
>>
>>
>>         _______________________________________________
>>         scikit-learn mailing list
>>         scikit-learn at python.org <mailto:scikit-learn at python.org>
>>         https://mail.python.org/mailman/listinfo/scikit-learn
>>         <https://mail.python.org/mailman/listinfo/scikit-learn>
>>
>>
>>
>>
>>     _______________________________________________
>>     scikit-learn mailing list
>>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>>     https://mail.python.org/mailman/listinfo/scikit-learn
>>     <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
>     _______________________________________________
>     scikit-learn mailing list
>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>     https://mail.python.org/mailman/listinfo/scikit-learn
>     <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/d1814c04/attachment-0001.html>

From raga.markely at gmail.com  Mon Aug 28 15:07:44 2017
From: raga.markely at gmail.com (Raga Markely)
Date: Mon, 28 Aug 2017 15:07:44 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <b4587533-b806-7bce-8fde-9aa19c0cc0d2@gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
 <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
 <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>
 <b4587533-b806-7bce-8fde-9aa19c0cc0d2@gmail.com>
Message-ID: <CAOLKFqtTCrQ3CSNqSR1vxibyZLB4tD51ZST27VQe4G6xtEQ1MA@mail.gmail.com>

Ah.. got it :D..

The pipeline was run in gridsearchcv..

It works now after calling fit..

Thanks!
Raga

On Mon, Aug 28, 2017 at 2:55 PM, Andreas Mueller <t3kcit at gmail.com> wrote:

> Have you called "fit" on the pipeline?
>
>
> On 08/28/2017 02:12 PM, Raga Markely wrote:
>
> Thank you, Andreas.
>
> When I try
>
>> pipe_lr.named_steps['clf'].coef_
>
>
> I get:
>
>> AttributeError: 'LogisticRegression' object has no attribute 'coef_'
>
>
> And when I try:
>
>> pipe_lr.named_steps['clf']
>
>
> I get:
>
>> LogisticRegression(C=0.1, class_weight=None, dual=False,
>> fit_intercept=True, intercept_scaling=1, max_iter=100, multi_class='ovr',
>> n_jobs=1, penalty='l2', random_state=None, solver='liblinear', tol=0.0001,
>> verbose=0, warm_start=False)
>
>
> I wonder what I am missing?
>
> Thanks,
> Raga
>
>
> On Mon, Aug 28, 2017 at 12:01 PM, Andreas Mueller <t3kcit at gmail.com>
> wrote:
>
>> Can can get the coefficients on the scaled data with
>> pipeline_lr.named_steps_['clf'].coef_
>> though
>>
>>
>> On 08/28/2017 12:08 AM, Raga Markely wrote:
>>
>> No problem, thank you!
>>
>> Best,
>> Raga
>>
>> On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman <joel.nothman at gmail.com>
>> wrote:
>>
>>> No, we do not have a way to get the coefficients with respect to the
>>> input (pre-scaling) space.
>>>
>>> On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com> wrote:
>>>
>>>> Hello,
>>>>
>>>> I am wondering if it's possible to get the weight coefficients of
>>>> logistic regression from a pipeline?
>>>>
>>>> For instance, I have the followings:
>>>>
>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
>>>>> pipe_lr.fit(X, y)
>>>>
>>>>
>>>> Does pipe_lr have an attribute that I can call to get the weight
>>>> coefficient?
>>>>
>>>> Or do I have to get it from the classifier as follows?
>>>>
>>>>> X_std = StandardScaler().fit_transform(X)
>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>> clf_lr.fit(X_std, y)
>>>>> clf_lr.coef_
>>>>
>>>>
>>>> Thank you,
>>>> Raga
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>>
>> _______________________________________________
>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
>
> _______________________________________________
> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/c4cee731/attachment.html>

From t3kcit at gmail.com  Mon Aug 28 15:20:27 2017
From: t3kcit at gmail.com (Andreas Mueller)
Date: Mon, 28 Aug 2017 15:20:27 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <CAOLKFqtTCrQ3CSNqSR1vxibyZLB4tD51ZST27VQe4G6xtEQ1MA@mail.gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
 <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
 <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>
 <b4587533-b806-7bce-8fde-9aa19c0cc0d2@gmail.com>
 <CAOLKFqtTCrQ3CSNqSR1vxibyZLB4tD51ZST27VQe4G6xtEQ1MA@mail.gmail.com>
Message-ID: <f71b0fb0-747e-f96c-9984-6b2735d457b4@gmail.com>

you can also use grid.best_estimator_ (and then all the rest)

On 08/28/2017 03:07 PM, Raga Markely wrote:
> Ah.. got it :D..
>
> The pipeline was run in gridsearchcv..
>
> It works now after calling fit..
>
> Thanks!
> Raga
>
> On Mon, Aug 28, 2017 at 2:55 PM, Andreas Mueller <t3kcit at gmail.com 
> <mailto:t3kcit at gmail.com>> wrote:
>
>     Have you called "fit" on the pipeline?
>
>
>     On 08/28/2017 02:12 PM, Raga Markely wrote:
>>     Thank you, Andreas.
>>
>>     When I try
>>
>>         pipe_lr.named_steps['clf'].coef_
>>
>>
>>     I get:
>>
>>         AttributeError: 'LogisticRegression' object has no attribute
>>         'coef_'
>>
>>
>>     And when I try:
>>
>>         pipe_lr.named_steps['clf']
>>
>>
>>     I get:
>>
>>         LogisticRegression(C=0.1, class_weight=None, dual=False,
>>         fit_intercept=True, intercept_scaling=1, max_iter=100,
>>         multi_class='ovr', n_jobs=1, penalty='l2', random_state=None,
>>         solver='liblinear', tol=0.0001, verbose=0, warm_start=False)
>>
>>
>>     I wonder what I am missing?
>>
>>     Thanks,
>>     Raga
>>
>>
>>     On Mon, Aug 28, 2017 at 12:01 PM, Andreas Mueller
>>     <t3kcit at gmail.com <mailto:t3kcit at gmail.com>> wrote:
>>
>>         Can can get the coefficients on the scaled data with
>>         pipeline_lr.named_steps_['clf'].coef_
>>         though
>>
>>
>>         On 08/28/2017 12:08 AM, Raga Markely wrote:
>>>         No problem, thank you!
>>>
>>>         Best,
>>>         Raga
>>>
>>>         On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman
>>>         <joel.nothman at gmail.com <mailto:joel.nothman at gmail.com>> wrote:
>>>
>>>             No, we do not have a way to get the coefficients with
>>>             respect to the input (pre-scaling) space.
>>>
>>>             On 28 August 2017 at 13:20, Raga Markely
>>>             <raga.markely at gmail.com <mailto:raga.markely at gmail.com>>
>>>             wrote:
>>>
>>>                 Hello,
>>>
>>>                 I am wondering if it's possible to get the weight
>>>                 coefficients of logistic regression from a pipeline?
>>>
>>>                 For instance, I have the followings:
>>>
>>>                     clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>                     pipe_lr = Pipeline([['sc', StandardScaler()],
>>>                     ['clf', clf_lr]])
>>>                     pipe_lr.fit(X, y)
>>>
>>>
>>>                 Does pipe_lr have an attribute that I can call to
>>>                 get the weight coefficient?
>>>
>>>                 Or do I have to get it from the classifier as follows?
>>>
>>>                     X_std = StandardScaler().fit_transform(X)
>>>                     clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>                     clf_lr.fit(X_std, y)
>>>                     clf_lr.coef_
>>>
>>>
>>>                 Thank you,
>>>                 Raga
>>>
>>>
>>>                 _______________________________________________
>>>                 scikit-learn mailing list
>>>                 scikit-learn at python.org <mailto:scikit-learn at python.org>
>>>                 https://mail.python.org/mailman/listinfo/scikit-learn
>>>                 <https://mail.python.org/mailman/listinfo/scikit-learn>
>>>
>>>
>>>
>>>             _______________________________________________
>>>             scikit-learn mailing list
>>>             scikit-learn at python.org <mailto:scikit-learn at python.org>
>>>             https://mail.python.org/mailman/listinfo/scikit-learn
>>>             <https://mail.python.org/mailman/listinfo/scikit-learn>
>>>
>>>
>>>
>>>
>>>         _______________________________________________
>>>         scikit-learn mailing list
>>>         scikit-learn at python.org <mailto:scikit-learn at python.org>
>>>         https://mail.python.org/mailman/listinfo/scikit-learn
>>>         <https://mail.python.org/mailman/listinfo/scikit-learn>
>>
>>
>>         _______________________________________________
>>         scikit-learn mailing list
>>         scikit-learn at python.org <mailto:scikit-learn at python.org>
>>         https://mail.python.org/mailman/listinfo/scikit-learn
>>         <https://mail.python.org/mailman/listinfo/scikit-learn>
>>
>>
>>
>>
>>     _______________________________________________
>>     scikit-learn mailing list
>>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>>     https://mail.python.org/mailman/listinfo/scikit-learn
>>     <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
>     _______________________________________________
>     scikit-learn mailing list
>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>     https://mail.python.org/mailman/listinfo/scikit-learn
>     <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/54ab7a7d/attachment-0001.html>

From raga.markely at gmail.com  Mon Aug 28 16:32:48 2017
From: raga.markely at gmail.com (Raga Markely)
Date: Mon, 28 Aug 2017 16:32:48 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <f71b0fb0-747e-f96c-9984-6b2735d457b4@gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
 <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
 <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>
 <b4587533-b806-7bce-8fde-9aa19c0cc0d2@gmail.com>
 <CAOLKFqtTCrQ3CSNqSR1vxibyZLB4tD51ZST27VQe4G6xtEQ1MA@mail.gmail.com>
 <f71b0fb0-747e-f96c-9984-6b2735d457b4@gmail.com>
Message-ID: <CAOLKFqtiiquVjRfpya4+2exyRm2x9ookq9JC_Ha9DWnyNt_MNg@mail.gmail.com>

Sounds good.. tried it and works.. thank you!

On Mon, Aug 28, 2017 at 3:20 PM, Andreas Mueller <t3kcit at gmail.com> wrote:

> you can also use grid.best_estimator_ (and then all the rest)
>
> On 08/28/2017 03:07 PM, Raga Markely wrote:
>
> Ah.. got it :D..
>
> The pipeline was run in gridsearchcv..
>
> It works now after calling fit..
>
> Thanks!
> Raga
>
> On Mon, Aug 28, 2017 at 2:55 PM, Andreas Mueller <t3kcit at gmail.com> wrote:
>
>> Have you called "fit" on the pipeline?
>>
>>
>> On 08/28/2017 02:12 PM, Raga Markely wrote:
>>
>> Thank you, Andreas.
>>
>> When I try
>>
>>> pipe_lr.named_steps['clf'].coef_
>>
>>
>> I get:
>>
>>> AttributeError: 'LogisticRegression' object has no attribute 'coef_'
>>
>>
>> And when I try:
>>
>>> pipe_lr.named_steps['clf']
>>
>>
>> I get:
>>
>>> LogisticRegression(C=0.1, class_weight=None, dual=False,
>>> fit_intercept=True, intercept_scaling=1, max_iter=100, multi_class='ovr',
>>> n_jobs=1, penalty='l2', random_state=None, solver='liblinear', tol=0.0001,
>>> verbose=0, warm_start=False)
>>
>>
>> I wonder what I am missing?
>>
>> Thanks,
>> Raga
>>
>>
>> On Mon, Aug 28, 2017 at 12:01 PM, Andreas Mueller <t3kcit at gmail.com>
>> wrote:
>>
>>> Can can get the coefficients on the scaled data with
>>> pipeline_lr.named_steps_['clf'].coef_
>>> though
>>>
>>>
>>> On 08/28/2017 12:08 AM, Raga Markely wrote:
>>>
>>> No problem, thank you!
>>>
>>> Best,
>>> Raga
>>>
>>> On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman <joel.nothman at gmail.com>
>>> wrote:
>>>
>>>> No, we do not have a way to get the coefficients with respect to the
>>>> input (pre-scaling) space.
>>>>
>>>> On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com>
>>>> wrote:
>>>>
>>>>> Hello,
>>>>>
>>>>> I am wondering if it's possible to get the weight coefficients of
>>>>> logistic regression from a pipeline?
>>>>>
>>>>> For instance, I have the followings:
>>>>>
>>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>>> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
>>>>>> pipe_lr.fit(X, y)
>>>>>
>>>>>
>>>>> Does pipe_lr have an attribute that I can call to get the weight
>>>>> coefficient?
>>>>>
>>>>> Or do I have to get it from the classifier as follows?
>>>>>
>>>>>> X_std = StandardScaler().fit_transform(X)
>>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>>> clf_lr.fit(X_std, y)
>>>>>> clf_lr.coef_
>>>>>
>>>>>
>>>>> Thank you,
>>>>> Raga
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> scikit-learn mailing list
>>>>> scikit-learn at python.org
>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>>
>> _______________________________________________
>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
>
> _______________________________________________
> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/b383eff5/attachment.html>

From olivier.grisel at ensta.org  Mon Aug 28 16:42:27 2017
From: olivier.grisel at ensta.org (Olivier Grisel)
Date: Mon, 28 Aug 2017 22:42:27 +0200
Subject: [scikit-learn] scikit-learn-commits mailing list defunct?
In-Reply-To: <465fa212-0fb6-cfe2-2281-416c672429ca@gmail.com>
References: <465fa212-0fb6-cfe2-2281-416c672429ca@gmail.com>
Message-ID: <CAFvE7K6nrdFsOLc2KP_CTAcnKH_Nqpe0zmMjtZ0aVRgwOpnF5g@mail.gmail.com>

+1
?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/aebf0918/attachment.html>

From olivier.grisel at ensta.org  Mon Aug 28 16:42:51 2017
From: olivier.grisel at ensta.org (Olivier Grisel)
Date: Mon, 28 Aug 2017 22:42:51 +0200
Subject: [scikit-learn] scikit-learn-commits mailing list defunct?
In-Reply-To: <CAFvE7K6nrdFsOLc2KP_CTAcnKH_Nqpe0zmMjtZ0aVRgwOpnF5g@mail.gmail.com>
References: <465fa212-0fb6-cfe2-2281-416c672429ca@gmail.com>
 <CAFvE7K6nrdFsOLc2KP_CTAcnKH_Nqpe0zmMjtZ0aVRgwOpnF5g@mail.gmail.com>
Message-ID: <CAFvE7K5Twt1v=k1sz-1FOGCwRRHusA6z_xzeOWWYpJx5utcTHw@mail.gmail.com>

+1 for python.org if they accept this kind of mailing lists.
?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/2fd9e939/attachment-0001.html>

From joel.nothman at gmail.com  Mon Aug 28 20:23:03 2017
From: joel.nothman at gmail.com (Joel Nothman)
Date: Tue, 29 Aug 2017 10:23:03 +1000
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <CAOLKFqtiiquVjRfpya4+2exyRm2x9ookq9JC_Ha9DWnyNt_MNg@mail.gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
 <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
 <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>
 <b4587533-b806-7bce-8fde-9aa19c0cc0d2@gmail.com>
 <CAOLKFqtTCrQ3CSNqSR1vxibyZLB4tD51ZST27VQe4G6xtEQ1MA@mail.gmail.com>
 <f71b0fb0-747e-f96c-9984-6b2735d457b4@gmail.com>
 <CAOLKFqtiiquVjRfpya4+2exyRm2x9ookq9JC_Ha9DWnyNt_MNg@mail.gmail.com>
Message-ID: <CAAkaFLUWwRVUiQJSK=LfNCmkjNkEoGv6LKts+juwtGcs5a2JuA@mail.gmail.com>

Sorry if I misunderstood your question.

On 29 August 2017 at 06:32, Raga Markely <raga.markely at gmail.com> wrote:

> Sounds good.. tried it and works.. thank you!
>
> On Mon, Aug 28, 2017 at 3:20 PM, Andreas Mueller <t3kcit at gmail.com> wrote:
>
>> you can also use grid.best_estimator_ (and then all the rest)
>>
>> On 08/28/2017 03:07 PM, Raga Markely wrote:
>>
>> Ah.. got it :D..
>>
>> The pipeline was run in gridsearchcv..
>>
>> It works now after calling fit..
>>
>> Thanks!
>> Raga
>>
>> On Mon, Aug 28, 2017 at 2:55 PM, Andreas Mueller <t3kcit at gmail.com>
>> wrote:
>>
>>> Have you called "fit" on the pipeline?
>>>
>>>
>>> On 08/28/2017 02:12 PM, Raga Markely wrote:
>>>
>>> Thank you, Andreas.
>>>
>>> When I try
>>>
>>>> pipe_lr.named_steps['clf'].coef_
>>>
>>>
>>> I get:
>>>
>>>> AttributeError: 'LogisticRegression' object has no attribute 'coef_'
>>>
>>>
>>> And when I try:
>>>
>>>> pipe_lr.named_steps['clf']
>>>
>>>
>>> I get:
>>>
>>>> LogisticRegression(C=0.1, class_weight=None, dual=False,
>>>> fit_intercept=True, intercept_scaling=1, max_iter=100, multi_class='ovr',
>>>> n_jobs=1, penalty='l2', random_state=None, solver='liblinear', tol=0.0001,
>>>> verbose=0, warm_start=False)
>>>
>>>
>>> I wonder what I am missing?
>>>
>>> Thanks,
>>> Raga
>>>
>>>
>>> On Mon, Aug 28, 2017 at 12:01 PM, Andreas Mueller <t3kcit at gmail.com>
>>> wrote:
>>>
>>>> Can can get the coefficients on the scaled data with
>>>> pipeline_lr.named_steps_['clf'].coef_
>>>> though
>>>>
>>>>
>>>> On 08/28/2017 12:08 AM, Raga Markely wrote:
>>>>
>>>> No problem, thank you!
>>>>
>>>> Best,
>>>> Raga
>>>>
>>>> On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman <joel.nothman at gmail.com>
>>>> wrote:
>>>>
>>>>> No, we do not have a way to get the coefficients with respect to the
>>>>> input (pre-scaling) space.
>>>>>
>>>>> On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Hello,
>>>>>>
>>>>>> I am wondering if it's possible to get the weight coefficients of
>>>>>> logistic regression from a pipeline?
>>>>>>
>>>>>> For instance, I have the followings:
>>>>>>
>>>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>>>> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
>>>>>>> pipe_lr.fit(X, y)
>>>>>>
>>>>>>
>>>>>> Does pipe_lr have an attribute that I can call to get the weight
>>>>>> coefficient?
>>>>>>
>>>>>> Or do I have to get it from the classifier as follows?
>>>>>>
>>>>>>> X_std = StandardScaler().fit_transform(X)
>>>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>>>> clf_lr.fit(X_std, y)
>>>>>>> clf_lr.coef_
>>>>>>
>>>>>>
>>>>>> Thank you,
>>>>>> Raga
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> scikit-learn mailing list
>>>>>> scikit-learn at python.org
>>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> scikit-learn mailing list
>>>>> scikit-learn at python.org
>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>
>>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>>
>> _______________________________________________
>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170829/4e0aa6e3/attachment.html>

From raga.markely at gmail.com  Mon Aug 28 21:06:49 2017
From: raga.markely at gmail.com (Raga Markely)
Date: Mon, 28 Aug 2017 21:06:49 -0400
Subject: [scikit-learn] Getting weight coefficient of logistic
 regression from a pipeline
In-Reply-To: <CAAkaFLUWwRVUiQJSK=LfNCmkjNkEoGv6LKts+juwtGcs5a2JuA@mail.gmail.com>
References: <CAOLKFqvU=avHobSpf-CKiOX2yFvFNv8MF7=wz+e63-xL7k9jVQ@mail.gmail.com>
 <CAAkaFLUWAzAdDPR3usuXKz57mV4X04h_9F-CcoZ2hqf4RvZ-uQ@mail.gmail.com>
 <CAOLKFqsiSvhK0S0KGK_BYJZGHe5Wi8U7Tsf6zhh+8FJrQwkUTg@mail.gmail.com>
 <c694536f-b28d-d4d6-b8db-c572c8b125a8@gmail.com>
 <CAOLKFqv=J1it3f+AeR7jec-MOEECCMw9f=iOxLr5qxi7UXbL6Q@mail.gmail.com>
 <b4587533-b806-7bce-8fde-9aa19c0cc0d2@gmail.com>
 <CAOLKFqtTCrQ3CSNqSR1vxibyZLB4tD51ZST27VQe4G6xtEQ1MA@mail.gmail.com>
 <f71b0fb0-747e-f96c-9984-6b2735d457b4@gmail.com>
 <CAOLKFqtiiquVjRfpya4+2exyRm2x9ookq9JC_Ha9DWnyNt_MNg@mail.gmail.com>
 <CAAkaFLUWwRVUiQJSK=LfNCmkjNkEoGv6LKts+juwtGcs5a2JuA@mail.gmail.com>
Message-ID: <CAOLKFquJov44LRgowk9bAR26rLEMX-4dVb1vDp0eE5wSoxuTwA@mail.gmail.com>

No worries.. ur answer is helpful for me too.. I was actually exploring
different ways to get the coeff, what i can and can't get :)..

Thanks!

On Aug 28, 2017 8:24 PM, "Joel Nothman" <joel.nothman at gmail.com> wrote:

> Sorry if I misunderstood your question.
>
> On 29 August 2017 at 06:32, Raga Markely <raga.markely at gmail.com> wrote:
>
>> Sounds good.. tried it and works.. thank you!
>>
>> On Mon, Aug 28, 2017 at 3:20 PM, Andreas Mueller <t3kcit at gmail.com>
>> wrote:
>>
>>> you can also use grid.best_estimator_ (and then all the rest)
>>>
>>> On 08/28/2017 03:07 PM, Raga Markely wrote:
>>>
>>> Ah.. got it :D..
>>>
>>> The pipeline was run in gridsearchcv..
>>>
>>> It works now after calling fit..
>>>
>>> Thanks!
>>> Raga
>>>
>>> On Mon, Aug 28, 2017 at 2:55 PM, Andreas Mueller <t3kcit at gmail.com>
>>> wrote:
>>>
>>>> Have you called "fit" on the pipeline?
>>>>
>>>>
>>>> On 08/28/2017 02:12 PM, Raga Markely wrote:
>>>>
>>>> Thank you, Andreas.
>>>>
>>>> When I try
>>>>
>>>>> pipe_lr.named_steps['clf'].coef_
>>>>
>>>>
>>>> I get:
>>>>
>>>>> AttributeError: 'LogisticRegression' object has no attribute 'coef_'
>>>>
>>>>
>>>> And when I try:
>>>>
>>>>> pipe_lr.named_steps['clf']
>>>>
>>>>
>>>> I get:
>>>>
>>>>> LogisticRegression(C=0.1, class_weight=None, dual=False,
>>>>> fit_intercept=True, intercept_scaling=1, max_iter=100, multi_class='ovr',
>>>>> n_jobs=1, penalty='l2', random_state=None, solver='liblinear', tol=0.0001,
>>>>> verbose=0, warm_start=False)
>>>>
>>>>
>>>> I wonder what I am missing?
>>>>
>>>> Thanks,
>>>> Raga
>>>>
>>>>
>>>> On Mon, Aug 28, 2017 at 12:01 PM, Andreas Mueller <t3kcit at gmail.com>
>>>> wrote:
>>>>
>>>>> Can can get the coefficients on the scaled data with
>>>>> pipeline_lr.named_steps_['clf'].coef_
>>>>> though
>>>>>
>>>>>
>>>>> On 08/28/2017 12:08 AM, Raga Markely wrote:
>>>>>
>>>>> No problem, thank you!
>>>>>
>>>>> Best,
>>>>> Raga
>>>>>
>>>>> On Mon, Aug 28, 2017 at 12:01 AM, Joel Nothman <joel.nothman at gmail.com
>>>>> > wrote:
>>>>>
>>>>>> No, we do not have a way to get the coefficients with respect to the
>>>>>> input (pre-scaling) space.
>>>>>>
>>>>>> On 28 August 2017 at 13:20, Raga Markely <raga.markely at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Hello,
>>>>>>>
>>>>>>> I am wondering if it's possible to get the weight coefficients of
>>>>>>> logistic regression from a pipeline?
>>>>>>>
>>>>>>> For instance, I have the followings:
>>>>>>>
>>>>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>>>>> pipe_lr = Pipeline([['sc', StandardScaler()], ['clf', clf_lr]])
>>>>>>>> pipe_lr.fit(X, y)
>>>>>>>
>>>>>>>
>>>>>>> Does pipe_lr have an attribute that I can call to get the weight
>>>>>>> coefficient?
>>>>>>>
>>>>>>> Or do I have to get it from the classifier as follows?
>>>>>>>
>>>>>>>> X_std = StandardScaler().fit_transform(X)
>>>>>>>> clf_lr = LogisticRegression(penalty='l1', C=0.1)
>>>>>>>> clf_lr.fit(X_std, y)
>>>>>>>> clf_lr.coef_
>>>>>>>
>>>>>>>
>>>>>>> Thank you,
>>>>>>> Raga
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> scikit-learn mailing list
>>>>>>> scikit-learn at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> scikit-learn mailing list
>>>>>> scikit-learn at python.org
>>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> scikit-learn mailing list
>>>>> scikit-learn at python.org
>>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>>
>>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing listscikit-learn at python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170828/ae619303/attachment-0001.html>

From Jonny.Evans at soton.ac.uk  Wed Aug 30 05:50:47 2017
From: Jonny.Evans at soton.ac.uk (Evans J.R.A.)
Date: Wed, 30 Aug 2017 09:50:47 +0000
Subject: [scikit-learn] FW: Random Forest Regressor criterion
Message-ID: <EC7519B9C403B6419F6BC83AFD4F31B6757D9273@SRV00047.soton.ac.uk>

Hi there,

I would like to fully understand how the Random Forest Regressor chooses how to split the data at each node.

I understand that each tree considers a boostrap sample of the training data, and on each split a random subset of features (using max_features) are considered. But among these features, how does the algorithm work out which is the best split to make? I am using the default criterion 'mse', but don't understand the given explanation "equal to variance reduction as feature selection criterion". Does this mean that for each possible split that could be made, the sum of variances of data in the child nodes is calculated, then the algorithm would use the split with the least sum of variances?

Kind regards,

Jonny Evans
Doctoral Researcher
Transportation Research Group
Faculty of Engineering and the Environment
University of Southampton
Email: Jonny.Evans at soton.ac.uk<mailto:Jonny.Evans at soton.ac.uk>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170830/b0223dd5/attachment.html>

From se.raschka at gmail.com  Thu Aug 31 01:47:44 2017
From: se.raschka at gmail.com (Sebastian Raschka)
Date: Thu, 31 Aug 2017 01:47:44 -0400
Subject: [scikit-learn] Random Forest Regressor criterion
In-Reply-To: <EC7519B9C403B6419F6BC83AFD4F31B6757D9273@SRV00047.soton.ac.uk>
References: <EC7519B9C403B6419F6BC83AFD4F31B6757D9273@SRV00047.soton.ac.uk>
Message-ID: <F8644B06-F937-41E8-80C4-93E4B0CFBBA0@gmail.com>

Hi, 

regarding MSE minimization vs variance reduction; it's been a few years but I remember that we had a discussion about that, where Gilles Louppe explained that those two are identical when I was confused about the wikipedia equation at https://en.wikipedia.org/wiki/Decision_tree_learning#Variance_reduction (I didn't read carefully and somehow thought that x_i etc was referring to feature columns instead of x being the target variable :P). 

A better resource: I think Gilles also had a page about that in his thesis but I currently can't find the page. The thesis should be accessible from https://arxiv.org/abs/1407.7502 though, and I would recommend taking a look at "3.6.3 Finding the best binary split"  and page 108+ on how it's implemented (if this is still up to date with the current implementation!?). This would probably address all your questions :).

Best,
Sebastian



> On Aug 30, 2017, at 5:50 AM, Evans J.R.A. <Jonny.Evans at soton.ac.uk> wrote:
> 
> Hi there,
>  
> I would like to fully understand how the Random Forest Regressor chooses how to split the data at each node.
>  
> I understand that each tree considers a boostrap sample of the training data, and on each split a random subset of features (using max_features) are considered. But among these features, how does the algorithm work out which is the best split to make? I am using the default criterion ?mse?, but don?t understand the given explanation ?equal to variance reduction as feature selection criterion?. Does this mean that for each possible split that could be made, the sum of variances of data in the child nodes is calculated, then the algorithm would use the split with the least sum of variances?
>  
> Kind regards,
>  
> Jonny Evans
> Doctoral Researcher
> Transportation Research Group
> Faculty of Engineering and the Environment
> University of Southampton
> Email: Jonny.Evans at soton.ac.uk
>  
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn