Re: [scikit-learn] MLPClassifier as a feature selector

[Manoj Kumar]

Hi,

The weights and intercepts are available in the coefs_ and intercepts_
attribute respectively.

See
https://github.com/scikit-learn/scikit-learn/blob/a24c8b46/sklearn/neural_network/multilayer_perceptron.py#L835

On Wed, Dec 6, 2017 at 4:56 PM, Brown J.B. via scikit-learn <
scikit-learn@python.org> wrote:

> I am also very interested in knowing if there is a sklearn cookbook
> solution for getting the weights of a one-hidde-layer MLPClassifier.
> J.B.
>
> 2017-12-07 8:49 GMT+09:00 Thomas Evangelidis :
>
>> Greetings,
>>
>> I want to train a MLPClassifier with one hidden layer and use it as a
>> feature selector for an MLPRegressor.
>> Is it possible to get the values of the neurons from the last hidden
>> layer of the MLPClassifier to pass them as input to the MLPRegressor?
>>
>> If it is not possible with scikit-learn, is anyone aware of any
>> scikit-compatible NN library that offers this functionality? For example
>> this one:
>>
>> http://scikit-neuralnetwork.readthedocs.io/en/latest/index.html
>>
>> I wouldn't like to do this in Tensorflow because the MLP there is much
>> slower than scikit-learn's implementation.
>>
>>
>> Thomas
>>
>>
>> --
>>
>> ==
>>
>> Dr Thomas Evangelidis
>>
>> Post-doctoral Researcher
>> CEITEC - Central European Institute of Technology
>> Masaryk University
>> Kamenice 5/A35/2S049,
>> 62500 Brno, Czech Republic
>>
>> email: tev...@pharm.uoa.gr
>>
>>   teva...@gmail.com
>>
>>
>> website: https://sites.google.com/site/thomasevangelidishomepage/
>>
>>
>> ___
>> scikit-learn mailing list
>> scikit-learn@python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> ___
> scikit-learn mailing list
> scikit-learn@python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>


-- 
Manoj,
http://github.com/MechCoder
___
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn

Re: [scikit-learn] MLPClassifier as a feature selector

[Brown J.B. via scikit-learn]

I am also very interested in knowing if there is a sklearn cookbook
solution for getting the weights of a one-hidde-layer MLPClassifier.
J.B.

2017-12-07 8:49 GMT+09:00 Thomas Evangelidis :

> Greetings,
>
> I want to train a MLPClassifier with one hidden layer and use it as a
> feature selector for an MLPRegressor.
> Is it possible to get the values of the neurons from the last hidden layer
> of the MLPClassifier to pass them as input to the MLPRegressor?
>
> If it is not possible with scikit-learn, is anyone aware of any
> scikit-compatible NN library that offers this functionality? For example
> this one:
>
> http://scikit-neuralnetwork.readthedocs.io/en/latest/index.html
>
> I wouldn't like to do this in Tensorflow because the MLP there is much
> slower than scikit-learn's implementation.
>
>
> Thomas
>
>
> --
>
> ==
>
> Dr Thomas Evangelidis
>
> Post-doctoral Researcher
> CEITEC - Central European Institute of Technology
> Masaryk University
> Kamenice 5/A35/2S049,
> 62500 Brno, Czech Republic
>
> email: tev...@pharm.uoa.gr
>
>   teva...@gmail.com
>
>
> website: https://sites.google.com/site/thomasevangelidishomepage/
>
>
> ___
> scikit-learn mailing list
> scikit-learn@python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
___
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn

[scikit-learn] MLPClassifier as a feature selector

[Thomas Evangelidis]

Greetings,

I want to train a MLPClassifier with one hidden layer and use it as a
feature selector for an MLPRegressor.
Is it possible to get the values of the neurons from the last hidden layer
of the MLPClassifier to pass them as input to the MLPRegressor?

If it is not possible with scikit-learn, is anyone aware of any
scikit-compatible NN library that offers this functionality? For example
this one:

http://scikit-neuralnetwork.readthedocs.io/en/latest/index.html

I wouldn't like to do this in Tensorflow because the MLP there is much
slower than scikit-learn's implementation.


Thomas


-- 

==

Dr Thomas Evangelidis

Post-doctoral Researcher
CEITEC - Central European Institute of Technology
Masaryk University
Kamenice 5/A35/2S049,
62500 Brno, Czech Republic

email: tev...@pharm.uoa.gr

  teva...@gmail.com


website: https://sites.google.com/site/thomasevangelidishomepage/
___
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn

Re: [scikit-learn] Error while running 'python setup.py build_ext --inplace'

[Joel Nothman]

We're biased, but we reckon the skills to make a PR are (a) not
insurmountable with a bit of homework; and (b) very worthwhile to have. So
try pick it up by yourself, but give us a shout if you're struggling.
___
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn

Re: [scikit-learn] Error while running 'python setup.py build_ext --inplace'

[Aniket Meshram]

Alright, i'll make a pull request. But let me tell you guys, I'm totally
new to github. This is my first contribution. And until few days back, i
didn't even knew what a pull request was.
Anyways, what i mean is even though i make a request, it'll take time for
me to understand this whole changing something and reflecting it to the
master branch.
Meanwhile, I'm doing my homework on this, any suggestions would really be
appreciated.

Thanks,
Aniket

On Wed, Dec 6, 2017 at 5:42 AM, Joel Nothman  wrote:

> A PR is welcome if you can improve documentation. Thanks
>
> On 6 December 2017 at 04:01, Aniket Meshram 
> wrote:
>
>> Yeah. That did it. After updating Cython to latest 0.27.3, the issue is
>> resolved now.
>> Thanks all. I guess this should also be updated on the site / github as
>> well. What'd you say?
>>
>> Best,
>> Aniket
>>
>> On Tue, Dec 5, 2017 at 12:58 PM, Aniket Meshram <
>> aniket.g.mesh...@gmail.com> wrote:
>>
>>> I'm using Ubuntu 16.04 LTS Xenial, which shows 0.23.4-0ubuntu5 as the
>>> latest.
>>>
>>> 
>>> https://packages.ubuntu.com/search?keywords=cython
>>>
>>> 
>>> But yes, you are right, I checked on official Cython and I'll install
>>> the latest using PyPI. Thought Ubuntu gives the latest, but that isn't true
>>> anymore.
>>> Thanks Andreas.
>>>
>>> I'll let you guys know, once I update and rerun pip install ...
>>> Thanks
>>>
>>> On Tue, Dec 5, 2017 at 12:45 AM, Andreas Mueller 
>>> wrote:
>>>
 Please stay on the mailing list.
 That's not the current version. Please try updating as Olivier
 suggested.


 On 12/04/2017 01:52 PM, Aniket Meshram wrote:

 $ cython --version
 Cython version 0.23.4

 On Mon, Dec 4, 2017 at 10:28 PM, Andreas Mueller 
 wrote:

> What version of Cython are you using?
>
>
>
> On 12/04/2017 09:20 AM, Aniket Meshram wrote:
>
> I updated all the packages before running install.
>
> On Mon, Dec 4, 2017 at 6:07 PM, Olivier Grisel <
> olivier.gri...@ensta.org> wrote:
>
>> Maybe update your version of Cython?
>>
>> --
>> Olivier
>> ​
>>
>> ___
>> scikit-learn mailing list
>> scikit-learn@python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
>
> --
> Regards,
>
> Aniket G. Meshram
>
>
> ___
> scikit-learn mailing 
> listscikit-learn@python.orghttps://mail.python.org/mailman/listinfo/scikit-learn
>
>
>


 --
 Regards,

 Aniket G. Meshram



>>>
>>>
>>> --
>>> Regards,
>>>
>>> Aniket G. Meshram
>>>
>>
>>
>>
>> --
>> Regards,
>>
>> Aniket G. Meshram
>>
>> ___
>> scikit-learn mailing list
>> scikit-learn@python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> ___
> scikit-learn mailing list
> scikit-learn@python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>


-- 
Regards,

Aniket G. Meshram
___
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn

[scikit-learn] Add Grabit model to gradient boosting

[Fabio Sigrist]

Dear all,

I added the Tobit loss function to gradient boosting, see
https://github.com/scikit-learn/scikit-learn/pull/9961. Recently, I also
added a reference to a preprint of an article with documentation on the
methodology (https://arxiv.org/abs/1711.08695).

What are to next steps in order to decide whether this feature will be
added to sklearn?

Thanks a lot in advance.

Best regards,
Fabio Sigrist


*Lucerne University of Applied Sciences and Arts*

Institute of Financial Services Zug IFZ
Grafenauweg 10, CH-6300 Zug


*Fabio Sigrist, PhD *Lecturer

T +41 41 757 67 61
fabio.sigr...@hslu.ch
___
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn