🌐 Mailman 3 February 2019 - scikit-learn - python.org

Using sklearn-crfsuite on Production Systems
by Astha Agarwal Aug. 5, 2019

Aug. 5, 2019

Hi, I'm wondering if anyone is using sklearn-crfsuite on production systems? Is this library suitable for usage in industry on production systems (and not academia) for non-big data problems? Thanks, Astha

3 2

Consultation: eligibility of inclusion of speed-up?
by Gael Varoquaux Feb. 27, 2019

Feb. 27, 2019

I need core devs opinion (please, only core devs, I am sending this on the public ML for transparency): The following PR adds a speed up for expansion of polynomial kernels: https://github.com/scikit-learn/scikit-learn/pull/13003 According to the author, the speed up is significant (needs to be verified during a code review). The paper is a bit below citation level for inclusion of a new method, however this can be seen as a speed up of Nystrom. Strictly speaking, it is not just a speed-up, as it introduces a new estimator. The discussion on the PR is short and quickly reviews the relevant literature. My question: should we consider this as acceptable for inclusion (provided that it does show significant speedups with good prediction accuracy)? I am asking to know if we start the review and inclusion process or not. Cheers, Gaël

3 2

Sprint discussion points?
by Andreas Mueller Feb. 26, 2019

Feb. 26, 2019

Hey all. Should we collect some discussion points for the sprint? There's an unusual amount of core-devs present and I think we should seize the opportunity. Maybe we should create a page in the wiki or add it to the sprint page? Things that are high on my list of priorities are: * slicing pipelines * add get_feature_names to pipelines * freezing estimator * faster multi-metric scoring * fit_transform doing something other than fit.transform * imbalance-learn interface / subsampling in pipelines * Specifying search spaces and valid hyper parameters (https://github.com/scikit-learn/scikit-learn/issues/13031). * allowing EstimatorCV-style speed-up in GridSearches * storing pandas column names and using them as feature names Trying to discuss all of these might be too much, but maybe we can figure out a subset and make sure we have sleps to discuss? Most of these issues are on the roadmap, issue 13031 is reladed to #18 but not directly on the roadmap. Thanks, Andy

10 31

What's are the advantages and disadvantages of incremental learning?
by lampahome Feb. 26, 2019

Feb. 26, 2019

Generally speaking, we all know it's to save spaces with incremental learning. According to the ques in stackoverflow <https://cs.stackexchange.com/questions/51260/what-are-the-advantages-of-onl…> , it also said that. But what's the disadvantages? What I know from my experiments is two points below: 1. Train with subsets of data *but shouldn't be too small*. I prepared very small datasets and the predict result is very worse. 2. When training for a very long time, some elder behavors will be forgotten due to the multiple training epochs. That's all from my experience when training with *xgboost* incrementally. Or anything else?

1 0

New Governance document accepted!
by Andreas Mueller Feb. 25, 2019

Feb. 25, 2019

Hey y'all. It's my pleasure to announce that the new governance document has been accepted by a core-dev vote. Out of the 49 eligible core-devs, 22 voted "yes" on the mailing list and 4 voted "yes" on the issue tracker. The remaining core devs did not vote. You can find the document on github and soon also on the dev website: https://github.com/scikit-learn/scikit-learn/blob/master/doc/governance.rst This also means we now have a Technical Committee to resolve technical issues, with is (for now) Alexandre Gramfort, Olivier Grisel, Joel Nothman, Hanmin Qin, Gaël Varoquaux, Roman Yurchak and myself. As described in the governance document, we will also be reaching out to any developers that haven't been actively involved in the project in the past 12 month, asking if they want to transition to emeritus status, recanting their commit and voting rights until they become active again. If you're a current core dev and think you'd be more appropriately an emeritus member, feel free to comment here: https://github.com/scikit-learn/scikit-learn/issues/13257 Thanks, and hope to see you all soon at the sprint (online or in person). Andy

1 0

Fit and predict method
by Venkataraman B Feb. 24, 2019

Feb. 24, 2019

Hi, I had a question on the predict and fit methods The fit method is used to build the model ie classifier.fit(X,y). But when the predict method is called the model that is built is never passed. You only pass the test set. So what model does the predict function use to predict the output I am picking python after working on R and the predict function in R made more sense because the model that was built is passed along with the test set that has to be predicted Any response would be greatly appreciated -- Regards, Venkataraman B

2 2

Re: [scikit-learn] Sprint discussion points?
by Hanmin Qin Feb. 24, 2019

Feb. 24, 2019

Thanks. I'll take part in the OPTICS discussion and I'd like to see it at 14:00, though 10:00 will also be acceptable. The core issue now is how to design the API (i.e., use multiple extraction methods without calculating RD/CD again), and how to deal with the mysterious additions in _extract_optics (See https://github.com/scikit-learn/scikit-learn/issues/12375. I'm unable to contact the original author so I tend to follow the original paper and remove these additions. This will get rid of some parameters and make the interface much more friendly IMO). For other issues, I don't think you need to consider my time. I'll comment on relevant issues if I have any thoughts. Hanmin Qin ----- Original Message ----- From: Joel Nothman <joel.nothman(a)gmail.com> To: Scikit-learn user and developer mailing list <scikit-learn(a)python.org>, Hanmin Qin <qinhanmin2005(a)sina.com> Subject: Re: [scikit-learn] Sprint discussion points? Date: 2019-02-21 15:40 @Hanmin are there particular conversations you are keen to take part in, and particular times that suit you? On Thu., 21 Feb. 2019, 9:13 am Andreas Mueller, <t3kcit(a)gmail.com> wrote: On 2/20/19 4:40 PM, Gael Varoquaux wrote: > On Tue, Feb 19, 2019 at 06:16:20PM -0500, Andreas Mueller wrote: >> I put a draft schedule here: >> https://github.com/scikit-learn/scikit-learn/wiki/Upcoming-events#technical… > I'd like to discuss sample_props. They are important to me. > > Should I add them somewhere on the schedule? Maybe in a place where > people who care about them (AFAIK Joel and Alex also do) are available? > Sure, sounds like a plan. If they are discussed I'd like to be part of the discussion if possible given the complexity involved (and because I tried to implement it twice). But feel free to have it Monday without me if that works better. _______________________________________________ scikit-learn mailing list scikit-learn(a)python.org https://mail.python.org/mailman/listinfo/scikit-learn

2 1

Incremental learning but predict the older data not well?
by lampahome Feb. 23, 2019

Feb. 23, 2019

I tried to use SGDRegressor to train data incrementally because I have newer data everyday. But I found when I train with data for 30 days, and then predict the result of 1st day. The result is very different. Then I predict from 1st to 14th day, I found the result are all the same value. But result of each day are all different and have multiple patterns. Is that correct? or just the size of training data too small? thx

1 0

Re: [scikit-learn] VOTE: scikit-learn governance document
by Hanmin Qin Feb. 20, 2019

Feb. 20, 2019

+1 (personally I still think it's better to keep the flow chart, it seems useful for beginners) Hanmin Qin ----- Original Message ----- From: Alexandre Gramfort <alexandre.gramfort(a)inria.fr> To: Scikit-learn mailing list <scikit-learn(a)python.org> Subject: Re: [scikit-learn] VOTE: scikit-learn governance document Date: 2019-02-11 01:29 +1 for me too Alex On Sat, Feb 9, 2019 at 10:06 PM Gilles Louppe <g.louppe(a)gmail.com> wrote: Hi Andy, I read through to document. Even though I have not been really active these past months/years, I think it summarizes well our governance model. +1. Gilles On Sat, 9 Feb 2019 at 12:01, Adrin <adrin.jalali(a)gmail.com> wrote: > > +1 > > Thanks for the work you've put in it! > > On Sat, Feb 9, 2019, 03:00 Andreas Mueller <t3kcit(a)gmail.com wrote: >> >> Hey all. >> >> I want to call a vote on the final version on the scikit-learn >> governance document, which can be found in this PR: >> >> https://github.com/scikit-learn/scikit-learn/pull/12878 >> >> It underwent some significant changes in the last couple of weeks. >> >> The two-sentence summary is: conflicts are resolved by vote among core >> devs, with a technical committee resolving anything that can not be >> decided by at least a 2/3 majority. The initial technical committee is >> Alexander Gramfort, Olivier Grisel, Joel Nothman, Hanmin Qin, Gaël >> Varoquaux and myself (Andreas Müller). >> >> I would ask all of the *core developers* to either vote +1 for the >> governance doc, -1 against it, or to explicitly abstain here on the >> public mailing list (which is the way any vote will be conducted >> according to the new governance document). >> >> I suggest we leave the vote open for two weeks, so that the decision is >> made before the sprint and we can take actions. >> >> Anyone can still comment on the PR or here, though I would rather not >> make more changes as this has already been discussed to some length. >> >> Thank you for participating, >> >> Andy >> >> _______________________________________________ >> scikit-learn mailing list >> scikit-learn(a)python.org >> https://mail.python.org/mailman/listinfo/scikit-learn > > _______________________________________________ > scikit-learn mailing list > scikit-learn(a)python.org > https://mail.python.org/mailman/listinfo/scikit-learn _______________________________________________ scikit-learn mailing list scikit-learn(a)python.org https://mail.python.org/mailman/listinfo/scikit-learn _______________________________________________ scikit-learn mailing list scikit-learn(a)python.org https://mail.python.org/mailman/listinfo/scikit-learn

20 20

Does random forest work if there are very few features?
by lampahome Feb. 19, 2019

Feb. 19, 2019

I read doc and know tree-based model is determined by entropy or gini impurity. When model try to create leaf nodes, it will determine based on the feature, right? Ex: I have 2 features A,B, and I divide it with A. So I have left and right nodes based on A. It should have the best shape if I create nodes based on A, right? Now if I have 100 estimators but I only have two features, do I have different trees which are all based on feature A? or the shape of trees based on A are all the same cuz they were created by feature A? thx

3 2