Thank you for the good news! [😊]
Best
Ludovico
________________________________
Da: scikit-learn <scikit-learn-bounces+ludo25_90=hotmail....@python.org> per
conto di scikit-learn-requ...@python.org <scikit-learn-requ...@python.org>
Inviato: mercoledì 7 dicembre 2016 22.06
A: scikit-learn@python.org
Oggetto: scikit-learn Digest, Vol 9, Issue 24
Send scikit-learn mailing list submissions to
scikit-learn@python.org
To subscribe or unsubscribe via the World Wide Web, visit
https://mail.python.org/mailman/listinfo/scikit-learn
scikit-learn Info Page -
Python<https://mail.python.org/mailman/listinfo/scikit-learn>
mail.python.org
To see the collection of prior postings to the list, visit the scikit-learn
Archives. Using scikit-learn: To post a message to all the list members ...
or, via email, send a message with subject or body 'help' to
scikit-learn-requ...@python.org
You can reach the person managing the list at
scikit-learn-ow...@python.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of scikit-learn digest..."
Today's Topics:
1. Re: Nested Leave One Subject Out (LOSO) cross validation with
scikit (Andreas Mueller)
----------------------------------------------------------------------
Message: 1
Date: Wed, 7 Dec 2016 16:06:29 -0500
From: Andreas Mueller <t3k...@gmail.com>
To: Scikit-learn user and developer mailing list
<scikit-learn@python.org>
Subject: Re: [scikit-learn] Nested Leave One Subject Out (LOSO) cross
validation with scikit
Message-ID: <e3971a8a-a4eb-2536-4e78-95edd2fec...@gmail.com>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"
PredefinedSplit allows to define your split but I think it's better if
you look at this pr:
https://github.com/scikit-learn/scikit-learn/pull/7990
[https://avatars2.githubusercontent.com/u/7454015?v=3&s=400]<https://github.com/scikit-learn/scikit-learn/pull/7990>
[WIP] Cache pipeline by glemaitre · Pull Request #7990 ·
scikit-learn/scikit-learn<https://github.com/scikit-learn/scikit-learn/pull/7990>
github.com
Reference Issue Address the discussions in #3951 Other related issues and PR:
#2086 #5082 #5080 What does this implement/fix? Explain your changes. It
implements a version of Pipeline which allows ...
That allows you to cache steps in a pipeline so you don't have to
recompute the feature selection for each parameter.
On 12/07/2016 03:13 PM, Ludovico Coletta wrote:
>
> Thank you for the answer.
>
>
> I also thought about ShuffleSplit (n_splits=1), but I need to control
> which indices are used for training and which for testing in the
> nested folds. The problem is that I did feature selection before
> hyperparameters optimization (with a nested Leave One Out schema) and
> now I need the same partitioning for hyperparameters optimization. The
> reason why I did this is that the feature selection step is incredibly
> slow, I hope I can get rid of that step in the permutation test. Is
> not clear to me if I have to include feature selection in the
> permutation test as well.
>
>
> Maybe LeavePOut is what I need.
>
>
> Best
>
> Ludovico
>
>
>
> ------------------------------------------------------------------------
> *Da:* scikit-learn
> <scikit-learn-bounces+ludo25_90=hotmail....@python.org> per conto di
> scikit-learn-requ...@python.org <scikit-learn-requ...@python.org>
> *Inviato:* mercoled? 7 dicembre 2016 17.48
> *A:* scikit-learn@python.org
> *Oggetto:* scikit-learn Digest, Vol 9, Issue 22
> Send scikit-learn mailing list submissions to
> scikit-learn@python.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://mail.python.org/mailman/listinfo/scikit-learn
scikit-learn Info Page -
Python<https://mail.python.org/mailman/listinfo/scikit-learn>
mail.python.org
To see the collection of prior postings to the list, visit the scikit-learn
Archives. Using scikit-learn: To post a message to all the list members ...
> <https://mail.python.org/mailman/listinfo/scikit-learn>
scikit-learn Info Page -
Python<https://mail.python.org/mailman/listinfo/scikit-learn>
mail.python.org
To see the collection of prior postings to the list, visit the scikit-learn
Archives. Using scikit-learn: To post a message to all the list members ...
> scikit-learn Info Page - Python
> <https://mail.python.org/mailman/listinfo/scikit-learn>
scikit-learn Info Page -
Python<https://mail.python.org/mailman/listinfo/scikit-learn>
mail.python.org
To see the collection of prior postings to the list, visit the scikit-learn
Archives. Using scikit-learn: To post a message to all the list members ...
> mail.python.org
> To see the collection of prior postings to the list, visit the
> scikit-learn Archives. Using scikit-learn: To post a message to all
> the list members ...
>
>
>
> or, via email, send a message with subject or body 'help' to
> scikit-learn-requ...@python.org
>
> You can reach the person managing the list at
> scikit-learn-ow...@python.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of scikit-learn digest..."
>
>
> Today's Topics:
>
> 1. Re: New to scikit (Andreas Mueller)
> 2. Re: Nested Leave One Subject Out (LOSO) cross validation with
> scikit (Andreas Mueller)
> 3. return type of StandardScaler (Nilay Shrivastava)
> 4. Re: return type of StandardScaler (Bharat Didwania .)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 7 Dec 2016 11:33:38 -0500
> From: Andreas Mueller <t3k...@gmail.com>
> To: Scikit-learn user and developer mailing list
> <scikit-learn@python.org>
> Subject: Re: [scikit-learn] New to scikit
> Message-ID: <f1d5d579-0646-d05d-4bde-9d44c34ec...@gmail.com>
> Content-Type: text/plain; charset="windows-1252"; Format="flowed"
>
> http://scikit-learn.org/dev/developers/contributing.html#deprecation
>
> On 12/07/2016 09:42 AM, Chinmay Talegaonkar wrote:
> > Yeah, I found an easy bug. Looking for some help in writing
> > deprecation cycles for a bug.
> >
> > On Wed, Dec 7, 2016 at 8:05 PM, Siddharth Gupta
> > <siddharthgupta...@gmail.com <mailto:siddharthgupta...@gmail.com>>
> wrote:
> >
> > Great! Welcome to the community. I would suggest you to check out
> > the issues page on the github repo, raise hand to the issues you
> > feel like you can give a go to, check out the issues that are
> > tagged as require contributor. Issues are a good way to start,
> > they will direct you about the areas of the code base to explore.
> >
> > On Dec 7, 2016 6:02 PM, "Chinmay Talegaonkar"
> > <chinmay0...@gmail.com <mailto:chinmay0...@gmail.com>> wrote:
> >
> > Hi everyone,
> > I have a prior experience in python, and
> > have started learning machine learning recently. I wanted to
> > contribute to scikit, can anyone suggest a relatively easy
> > codebase to explore.
> > Thanks in advance!
> >
> >
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn@python.org <mailto:scikit-learn@python.org>
> > https://mail.python.org/mailman/listinfo/scikit-learn
> > <https://mail.python.org/mailman/listinfo/scikit-learn>
> >
> >
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn@python.org <mailto:scikit-learn@python.org>
> > https://mail.python.org/mailman/listinfo/scikit-learn
> > <https://mail.python.org/mailman/listinfo/scikit-learn>
> >
> >
> >
> >
> > --
> > --
> > *Chinmay Talegaonkar*
> > Cultural and Events Coordinator, Mood Indigo
> > ..............................................
> >
> >
> > +91-8879178724
> > chinmay0...@gmail.com <mailto:bajajkshiti...@gmail.com>
> > www.moodi.org<http://www.moodi.org> <http://www.moodi.org>
> > <http://www.moodi.org/>
> >
> >
> >
> >
> >
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn@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/20161207/4f86464e/attachment-0001.html>
>
> ------------------------------
>
> Message: 2
> Date: Wed, 7 Dec 2016 11:33:00 -0500
> From: Andreas Mueller <t3k...@gmail.com>
> To: Scikit-learn user and developer mailing list
> <scikit-learn@python.org>
> Subject: Re: [scikit-learn] Nested Leave One Subject Out (LOSO) cross
> validation with scikit
> Message-ID: <e9e342f5-d09b-92b7-e304-a0bfa37ae...@gmail.com>
> Content-Type: text/plain; charset="windows-1252"; Format="flowed"
>
>
>
> On 12/07/2016 07:41 AM, Ludovico Coletta wrote:
> >
> > Dear scikit experts,
> >
> >
> > I did as you suggested, but it is not exactly what I would like to do
> > ( I also read this:
> >
> http://stackoverflow.com/questions/40400351/nested-cross-validation-with-stratifiedshufflesplit-in-sklearn)
> >
> > Perhaps I should ask my question in another way: it is possible to
> > split the nested cv folds just once? It seems to me that this is not
> > possible, do you have any hints?
> >
> >
> Not sure I understand your question.
> You can do a single split by using ShuffleSplit(n_splits=1) for example.
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> <http://mail.python.org/pipermail/scikit-learn/attachments/20161207/063e6a44/attachment-0001.html>
>
> ------------------------------
>
> Message: 3
> Date: Wed, 7 Dec 2016 22:14:18 +0530
> From: Nilay Shrivastava <nilay.eule...@gmail.com>
> To: scikit-learn@python.org
> Subject: [scikit-learn] return type of StandardScaler
> Message-ID:
> <caklfaqra_qgdyprddtg58nhybqurwbct2_2nzkzriomww5r...@mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> StandardScaler returns numpy array even if the object passed is a pandas
> dataframe, shouldn't it return a dataframe?
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> <http://mail.python.org/pipermail/scikit-learn/attachments/20161207/feecd6be/attachment-0001.html>
>
> ------------------------------
>
> Message: 4
> Date: Wed, 7 Dec 2016 08:48:47 -0800
> From: "Bharat Didwania ." <bharat.didwania.ee...@itbhu.ac.in>
> To: Scikit-learn user and developer mailing list
> <scikit-learn@python.org>
> Subject: Re: [scikit-learn] return type of StandardScaler
> Message-ID:
> <caa3g_m_jqwalgkqa530vp5pshdry+mcwqiuts5qhdv-ejtc...@mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> you can use pandas.get_dummies()
> <http://pandas.pydata.org/pandas-docs/stable/generated/pandas.get_dummies.html>.
> It will perform one hot encoding on categorical columns, and produce a
> dataframe as the result. From there you can use
> pandas.concat([existing_df,
> new_df],axis=0) to add the new columns to your existing dataframe. This
> will avoid the use of a numpy array.
>
>
> On Wed, Dec 7, 2016 at 8:44 AM, Nilay Shrivastava
> <nilay.eule...@gmail.com>
> wrote:
>
> > StandardScaler returns numpy array even if the object passed is a pandas
> > dataframe, shouldn't it return a dataframe?
> >
> >
> > _______________________________________________
> > scikit-learn mailing list
> > scikit-learn@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/20161207/d6afb4d5/attachment.html>
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn@python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>
> ------------------------------
>
> End of scikit-learn Digest, Vol 9, Issue 22
> *******************************************
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn@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/20161207/8e3450c9/attachment.html>
------------------------------
Subject: Digest Footer
_______________________________________________
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn
------------------------------
End of scikit-learn Digest, Vol 9, Issue 24
*******************************************
_______________________________________________
scikit-learn mailing list
scikit-learn@python.org
https://mail.python.org/mailman/listinfo/scikit-learn
