Re: [Scikit-learn-general] Binary wheel packages for Linux are coming

AMAZING! On 04/13/2016 05:52 PM, Olivier Grisel wrote: > OpenBLAS 0.2.18 has been released yesterday with many fixes several > bugs found via testing the scipy stack on: > > https://build.openblas.net/ > > Matthew Brett did a full-rebuild of BLAS dependent projects (including > numpy, scipy and sc

Re: [Scikit-learn-general] deep belief network contribution to scikit-learn

://github.com/scikit-learn-contrib/scikit-learn-contrib that will probably give your code some visibility and will provide you with some feedback. Cheers, Andy On 03/10/2016 11:05 AM, Albert Busqué wrote: Hi Andreas, My name is Albert and I'd like to contribute to scikit-learn project

Re: [Scikit-learn-general] Implementation of Bag-of-Features

On 02/23/2016 04:32 PM, Guillaume Lemaitre wrote: Since that I was working on a cluster I did not realize but loading all the image in memory will be problematic with a laptop-desktop configuration. Or we can learn the PCA projection on a subset and to apply the dimension reduction right afte

Re: [Scikit-learn-general] scikit-learn 0.17.1 is out

Thank you so much :) I'll be around for the next one again ;) On 02/18/2016 01:17 PM, Olivier Grisel wrote: > Hi all, > > The 0.17.1 bugfix release is out on PyPI and anaconda.org: > > pip install --upgrade scikit-learn > > or if you prefer conda: > > conda update scikit-learn > > I will update th

Re: [Scikit-learn-general] Pairwise Distances for n closest matches

Hi Deb. We don't really expose low-level functions like these, and only include them if there is a particular use-case. Why not create a pull request for scipy? Cheers, Andy On 02/01/2016 01:14 AM, Debanjan Bhattacharyya wrote: Hi I have written a method pairwise_distances_argmin_min

Re: [Scikit-learn-general] Dropping Python 2.6 compatibility

This seems like a good motivation. I think we heard a lot of people not having any issues, and more people going this way. So let's go for it? [that is change the readme, change the build matrix, grep the source for backports] On 01/08/2016 08:55 AM, Olivier Grisel wrote: > Yet another perspect

Re: [Scikit-learn-general] Contributing to Scikit-Learn(GSOC)

contributing to the project now, and see if you find a project and a mentor that are suitable. I'm a bit confused how you could talk about neural networks and deep learning without talking about regression or classification... Best, Andy On 01/10/2016 01:14 AM, Imaculate Mosha wrote: H

Re: [Scikit-learn-general] Fine tuning parameters of Multi label classification

Hi Sanant. Please provide the full traceback. Best, Andy On 12/28/2015 05:38 AM, Startup Hire wrote: Hi all, Hope you are doing well. I am working on fine tuning the following parameters in SGD Classifier which I am using inside OneVsRest Classifier. I am using GridSearch to use the same

Re: [Scikit-learn-general] Replicability in GridSearchCV without passing a CV generator/iterable

On 12/20/2015 09:31 AM, Alexandre Gramfort wrote: >> Is there any way to specify the random_state of the StratifiedKFold CV >> generator that is used when one uses sklearn.grid_search.GridSearchCV(cv=int >> or none)? > By default, there is no shuffling, so cv=int or None is deterministic.

Re: [Scikit-learn-general] scikit-learn-0.17, atlas/unittest issues...

On 12/27/2015 06:58 PM, Joe Cammisa wrote: > > greetings. i'm trying to build scikit-learn-0.17 into a homemade > python stack on ubuntu-14.04 and having a terrible time of it. no > matter what i try, the unittests invariably fail with such messages as > these: > > ERROR: Failure: ImportError

Re: [Scikit-learn-general] Dynamic Time Warping Contribution

On 12/07/2015 04:33 AM, Alexandre Gramfort wrote: >> How do you plan to represent variable-length time series? Lists of 1d numpy >> arrays work but would be slow I guess. The ideal representation needs to be >> compatible with grid search and fast. > good point. I was thinking of forcing all time s

Re: [Scikit-learn-general] sklearn.cross_decomposition.PLSRegression: how to re-scale my prediction?

Hi Ola. Can you please raise an issue on the issue tracker, preferably with code and data to reproduce? If you can't share your data, see if you can reproduce the issue with synthetic data. Best, Andy On 12/02/2015 10:34 AM, Ola Pawluczyk wrote: Hello all, I'm doing a PLS reg

Re: [Scikit-learn-general] Jeff Levesque: '.predict_proba()' me tho for smaller datasets

On 12/01/2015 11:28 PM, Jeff Levesque wrote: > Is there a way to determine if the data used with the SVC class, used to > generate an SVM model, would generate a poor model, or confidence percentage > (or 'decision_function', if that's preferred)? > > I don't understand the question. ---

Re: [Scikit-learn-general] speceficity metric for Crossvalidation

There is no specificity metric, but it is easy to implement. What have you tried and what exactly was the error you got when using ``make_scorer``? You can either write a callable that takes "trained estimator, X_test, y_test" or you can write a function that takes y_test, y_pred and call mak

Re: [Scikit-learn-general] Import error for Robust scaler

On 12/02/2015 05:19 AM, Sumedh Arani wrote: > > Greetings!! > > Yet still the problem still arises and it still shows import error for > RobustScaler > > And I also am reading the documentation of 0.16 for reference. When I > tried to run one of the examples given in examples folder and it has >

Re: [Scikit-learn-general] Predicting on new data after OneVsRest Classifier (multi label)

You should also store you OneHotTransfomer. On 11/17/2015 01:19 AM, Startup Hire wrote: Hi Pypers, Hope you are doing well. I am doing multi label classification in which my X and Y are sparse matrices with Y properly binarized. I am able to get done with multi label classification with 123

[Scikit-learn-general] Sprint in SF

Hey all. We have a pretty busy and active sprint in SF. If any devs have any spare time, some help with reviews would be awesome! Thanks, Andy -- Presto, an open source distributed SQL query engine for big data

Re: [Scikit-learn-general] Interested in contributing to sikit learn

Oh and for "how long can you take to fix a bug": Usually you can take a couple of weeks if you like. If you stop responding to comments for like a week or two, someone else might take up the issue where you left off. But there is really no requirement on speed. On 11/04/2015 07:41 AM, Amandee

Re: [Scikit-learn-general] Interested in contributing to sikit learn

the solution in detail before taking up an issue. Cheers, Andy On 11/04/2015 07:41 AM, Amandeep Gautam wrote: Hi Andreas, I am Amandeep Gautam, a master's student at University of Minnesota. I want to start contributing to scikit-learn. I hope this direct email does not violate the gene

Re: [Scikit-learn-general] branch 0.17.X reset

On 10/31/2015 01:41 PM, Gael Varoquaux wrote: > > Why didn't you use "git revert" rather than rewriting history? > > Thanks for working on the release, by the way! > I didn't think it mattered for a branch no-one was using, and it is already a bit hard to keep track of what is happening in the rel

[Scikit-learn-general] branch 0.17.X reset

Hey. Just in case anyone was working on 0.17.X for some reason: I needed to remove a commit that was accidentally cherrypicked. So if you have a 0.17.X branch you need to update it to the rebased master: git fetch uptream && git checkout 0.17.X && git reset --hard upstream/0.17

Re: [Scikit-learn-general] MICE Imputation for SciKit Learn

. It would likely to be a good addition, but getting it merged may take a lot of effort and patience. You might want to try tackling an easy issue first to become familiar with our development practices. Cheers, Andy On 10/21/2015 05:13 PM, Ouwen Huang wrote: Hello all, MICE is a recent imput

Re: [Scikit-learn-general] Jeff Levesque: Prototype Interface (API) to sklearn

Hi Jeff. Do you have a hosted version? That is more likely to get feedback. Andy -- ___ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https

Re: [Scikit-learn-general] ROC for one-class-SVM classifier

You should use "decision_function" not "predict" -- ___ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/

Re: [Scikit-learn-general] PyCon 2016 scikit-learn tutorial

some of the notebooks that I have been using for teaching. Andy On 09/29/2015 06:48 PM, Jacob Vanderplas wrote: Hi All, PyCon 2016 call for proposals <https://us.pycon.org/2016/speaking/tutorials/> just opened. For the last several years Olivier and I have been teaching a two-part s

Re: [Scikit-learn-general] graphical lasso "non SPD result" error

, Andy On 09/17/2015 11:55 AM, conahorse wrote: Hi everyone, I am trying to apply |glasso| on a very simple as well as sparse dataset made by 60+ features and 30k+ observations. Here(http://www.mediafire.com/download/ek8kk0pg3jpc6ll/weight_comp_simple_prop.df.train.csv) <ht

Re: [Scikit-learn-general] Learning with counts feature transform.

-fledged example? Cheers, Andy On 09/17/2015 07:41 PM, Christos F. Papadopoulos wrote: Hi all! Are there any plans to implement something like a countfeaturizer? https://msdn.microsoft.com/en-us/library/azure/dn913056.aspx It is a generic method for feature extraction that encodes multi-valued

Re: [Scikit-learn-general] Discrepancy in SkLearn Stratified Cross Validation

train_test_split is not stratified. In master, you can use "stratify=y" to make it stratified. Also: randomness. On 09/15/2015 10:55 AM, Mamun Rashid wrote: I am seeing a discrepancy between classification performance between two cross validation technique using the same data. I was wondering

Re: [Scikit-learn-general] GridSearchCV over OneVsRest wrapping SVC

On 09/12/2015 01:25 PM, Michael Eickenberg wrote: On Saturday, September 12, 2015, Andy <mailto:t3k...@gmail.com>> wrote: On 09/12/2015 02:54 AM, Daan Wynen wrote: Hi Andy, according to [1] "The multiclass support is handled according to a one-vs-one scheme.&q

Re: [Scikit-learn-general] GridSearchCV over OneVsRest wrapping SVC

On 09/12/2015 02:54 AM, Daan Wynen wrote: Hi Andy, according to [1] "The multiclass support is handled according to a one-vs-one scheme." That's why I was using the wrapper. SVC has one-vs-rest built-in. What should the docs say? The multiclass docs here are quite explicit

Re: [Scikit-learn-general] Common tests for functions vs deprecating functions

On 09/10/2015 09:22 AM, Gael Varoquaux wrote: > > These functions are important for reuse in an algorithmic setting: if I > am doing an algorithm that uses k-means or lars_path inside the > algorithm, it is much more natural to use the functions, and they have > less overhead. > > I think that the

Re: [Scikit-learn-general] Implementing the "Concordance correlation coefficient" in metrics

On 09/07/2015 06:03 AM, Stylianos Kampakis wrote: > > The interpretation of R^2 is less useful for machine learning models. > For example, Weka omits it all together for regression models. A > useful alternative is to simply use the correlation between the true > and the predicted values. Can yo

Re: [Scikit-learn-general] Implementing the "Concordance correlation coefficient" in metrics

odel instead of R^2? Why do you say it is robust? Best, Andy On 09/04/2015 08:15 AM, Stylianos Kampakis wrote: Hello everyone, I was thinking to add the concordance correlation coefficient as a metric for regression models and I wanted to ask first whether you think this is a good idea.

Re: [Scikit-learn-general] Turning on sample weights for linear_model.LogisticRegression

On 08/26/2015 09:29 PM, Joel Nothman wrote: > I agree. I suspect this was an unintentional omission, in fact. > > Apart from which, sample_weight support in liblinear could be merged > from https://github.com/scikit-learn/scikit-learn/pull/2784 which is > dormant, and merely needs some core contr

Re: [Scikit-learn-general] scikit-learn-0.16.1: Summary of errors from self-tests

Thanks for the report. This is due to changes in scipy 0.16 and fixed in master. On 08/10/2015 05:34 PM, Sergio Rojas wrote: > Summary of errors from self-tests of scikit-learn-0.16.1 > on ubuntu 14.04 under: > > > $ python3 > Python 3.4.3 (default, Jul 29 2015, 16:07:32) > [GCC 4.8.4] on linux

Re: [Scikit-learn-general] contributing

cikit-learn.org/dev/developers/contributing.html#contributing-code Code reviews of existing pull requests are also always welcome of course. After doing some "easy" fixes I can point you to more challenging ones. Cheers, Andy On 08/04/2015 10:18 PM, Jaret Flores wrote: Since I've come int

Re: [Scikit-learn-general] Added sample_weight to RFECV.fit but not sure how to test the change

(dropping points will clearly lose information otherwise). Have a look here: https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/utils/tests/test_class_weight.py#L41 Cheers, Andy On 07/24/2015 12:57 PM, Dale Smith wrote: Andy, I’ve thought a bit about your suggestion for testing. I’m

Re: [Scikit-learn-general] Added sample_weight to RFECV.fit but not sure how to test the change

I think my reply for this got swallowed by the sourceforge outage: The main thing that you should test is whether the added behavior is correct. For that you should confirm that changing sample weights is equivalent to duplicating / dropping a sample. On 07/22/2015 01:34 PM, Dale Smith wrote

Re: [Scikit-learn-general] [scikit-learn-general] Possible bug in RFECV.fit?

I don't think it is feasible to try to fail fast for all possible cases. I think users should try their code with something that runs through quickly before doing a multi-day run. I do that for all my experiments. On 07/22/2015 02:38 PM, Joel Nothman wrote: This isn't directly a problem with RF

Re: [Scikit-learn-general] Contribute to Scikit-learn

compatibility, testing and documentation. Have a look at the FAQ: http://scikit-learn.org/dev/faq.html And the developers guide: http://scikit-learn.org/dev/developers/index.html Cheers, Andy On 07/22/2015 10:48 PM, vaibhav sharma wrote: I am working on a project of machine learning in which i have

Re: [Scikit-learn-general] Moving mailing list to google groups

On 07/23/2015 01:46 AM, Stefan van der Walt wrote: > On 2015-07-22 01:36:47, Gael Varoquaux > wrote: >> Follow up (this message has been in the loop for a little while >> as sourceforge was down): we have decided that we would rather >> move to a mailing list hosted by python.org. > Good choice--I

[Scikit-learn-general] Moving mailing list to google groups

to the sourceforge list with a message that people should move. What do people think? Cheers, Andy -- Don't Limit Your Business. Reach for the Cloud. GigeNET's Cloud Solutions provide you with the tools and su

Re: [Scikit-learn-general] Scikit-learn with gcc 4.2 on FreeBSD.

Hi Nastran. This looks like a scipy problem. I guess it depends on which blas you link against. I'm not sure if it runs on BSD, but you could try running anaconda http://continuum.io/downloads Otherwise maybe check on the scipy list. Andy On 07/02/2015 10:04 PM, Nastaran Baradaran wrote

Re: [Scikit-learn-general] Sample weighting in RandomizedSearchCV

Hi Jose. That should work. Can you provide the full traceback? Also can you provide weights.shape? Andy On 06/08/2015 08:49 PM, José Guilherme Camargo de Souza wrote: > Hi all, > > I am having a different issue when trying to use sample_weights with > RandomizedSearchCV: > >

[Scikit-learn-general] Estimator Overview / Summary

, but not for classification or regression. Having time and memory complexity, whether they support sparse data or not, and whether they support class_weights / sample_weights and partial fit on one page would be useful, I think. I'd be happy to pick it up. Cheers,

Re: [Scikit-learn-general] [ an error in feature_selection/base.py]

Hi Yao. Please provide your code and what type and shape of X are and your version of scikit-learn. For questions like these, it might be that stackoverflow will give you a faster answer than the mailing list. Cheers, Andy On 05/09/2015 09:34 AM, Yao wrote: Hi, all, I came across an

Re: [Scikit-learn-general] RFC (also by users) on interpreting 1d X

On 05/04/2015 07:47 AM, Gael Varoquaux wrote: > On Mon, May 04, 2015 at 01:32:02PM +0200, federico vaggi wrote: >> I think Gael makes a very strong argument, but I think the error should be as >> explicit and informative as possible (for new users). > +1. Including suggesting the syntax X[:, np.new

[Scikit-learn-general] Congratulations GSOC Students!

improvements to the cross-validation and model selection modules, and Wei Xu will be working on the mixture models. There was a slight hickup with the melange assignment (which should be fixed soon). The final mentor assignment is: Raghav: Vlad, Andy Wei Xu: Olivier, Loic Artem: Kyle, Michael

Re: [Scikit-learn-general] bias in svm.LinearSVC classification accuracy in very small data sample? (Andreas Mueller)

Your expectation is right, if you randomly assign labels, you shouldn't get more than 50% correct with a large enough dataset. I imagine there is some issue in how you shuffled the labels. Without the code, it is hard to tell. Are you sure the classes have the same size? On 04/26/2015 11:22 AM,

Re: [Scikit-learn-general] sample weights for RandomForestClassifier to compute cross_val_score with roc_auc metric

On 04/25/2015 09:18 AM, Luca Puggini wrote: I think it depends by the role of sample weight during the construction of the forest. If I set sample_weight = 2 for one of my samples is this equivalent to duplicate the row in the data? During fitting, yes, during evaluation currently not. On Fr

Re: [Scikit-learn-general] recursive feature elimination

On 04/22/2015 01:53 PM, Pagliari, Roberto wrote: is it possible to pass a gridsearchCV object to RFE, as opposed to a simple estimator? See here: http://stackoverflow.com/questions/29538292/doing-hyperparameter-estimation-for-the-estimator-in-each-fold-of-recursive-feat I would advise against i

Re: [Scikit-learn-general] Degree parameter in Nu-Support Vector Classification

On 04/22/2015 06:06 PM, Artem wrote: Looks like a typo, indeed. Libsvm only uses `degree` for polynomial kernels. Yeah, that was fixed for SVC but not nuSVC it seems. -- BPM Camp - Free Virtual Workshop May 6th at 10am P

[Scikit-learn-general] [ANN] Scikit-learn 0.16.1 is out!

if you are using Anaconda! Cheers, Andy -- BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT Develop your own process in accordance with the BPMN 2 standard Learn Process modeling best practices with Bonita BPM th

Re: [Scikit-learn-general] Convolutive NMF

ce when working with convolutional (shouldn't it be that?) NMF. Why no use an autoencoder approach? Andy On 04/09/2015 08:20 AM, Dan Stowell wrote: > Hi all, > > Does anyone here have any experience/tips for _convolutive_ NMF in > scikit-learn (or in numpy more generally)? scikit-learn

Re: [Scikit-learn-general] 8.23.6. sklearn.neighbors.BallTree

Hi Nafiseh. Please direct questions like this to the scikit-learn mailing list or stackoverflow with the scikit-learn tag. However, you problem is unrelated to scikit-learn: you didn't import BallTree. Andy On 04/06/2015 03:26 AM, nafise mehdipoor wrote: Dear Andy, I have windows 7

Re: [Scikit-learn-general] Adaline (adaptive linear neuron) classifier

rd of this algorithm, and it is not mentioned in any of the prominent textbooks (ESL, Bishop, Murphy). So while it might be foundational, I don't think it is necessary for "completeness". Andy On 04/04/2015 12:16 AM, Sebastian Raschka wrote: > Hi, > e > maybe I overlooked

Re: [Scikit-learn-general] For Devs/Web Site Admins Of Sklearn

ly have the right legal infrastructure to actually handle any income, though. Cheers, Andy On 04/04/2015 04:38 AM, Jason Wolosonovich wrote: > Hello All, > > Have any of you (the developers/web site admins) considered placing the links > to lectures and videos in a more prominent pla

Re: [Scikit-learn-general] SGDClassifier implementation and shuffling

data is sorted, say, by labels. We benched shuffling once vs shuffling every time, and the runtime is basically the same, opted for shuffling every time. Maybe mblondel or larsmans can give better answers, though ;) Hth, Andy --

Re: [Scikit-learn-general] [ANN] scikit-learn 0.16.0 is out!

Thank you for your work in making the release happen Olivier! I've been dead to the world for two days, but I'll be back next week ;) Now let's go into GSoC mode! On 03/27/2015 12:26 PM, Olivier Grisel wrote: > Release highlights and full changelog available at: > > http://scikit-learn.org/0.16

Re: [Scikit-learn-general] [ANN] scikit-learn 0.16.0 is out! (Yogesh Karpate)

On 03/27/2015 01:46 PM, Wei Xue wrote: I have seen the updated images both in 0.16 and 0.15, in which 0.16 algorithms has less running time than in 0.15. You shouldn't interpret too much into the absolute numbers as they are not done on the same system. It might be my laptop (that generated the

Re: [Scikit-learn-general] GSoC2015 Hyperparameter Optimization topic

blish your code and we base our implementation on it. If your code is public and BSD / MIT licensed, it would probably be much safer. Why don't you just push your code under a permissive license? Thank you for providing your benchmarks, they might be quite helpful. Cheers, Andy On 03/26/201

Re: [Scikit-learn-general] GSoC2015 Hyperparameter Optimization topic

One thing that might also be interesting is "Bootstrapping" (in the compiler sense, not the statistics sense) the optimizer. The latest Jasper Snoek paper http://arxiv.org/abs/1502.05700 they used a hyper-parameter optimizer to optimize the parameter of a hyper-parameter optimizer on a set of opt

[Scikit-learn-general] [GSOC] Global optimization based Hyper parameter optimization Hamzeh Alsalhi

if you could mention relevant coursework etc. Thanks, Andy -- Dive into the World of Parallel Programming The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for

Re: [Scikit-learn-general] GSoC2015 Hyperparameter Optimization topic

On 03/24/2015 04:38 PM, Christof Angermueller wrote: > Thanks Andy! I replied to your comments: > https://docs.google.com/document/d/1bAWdiu6hZ6-FhSOlhgH-7x3weTluxRfouw9op9bHBxs/edit?usp=sharing. > > I summary, > * I will not mentioned parallelization as an extended features, >

Re: [Scikit-learn-general] My personal suggestion regarding topics for GSoC (and my official application :-) )

dules/generated/sklearn.linear_model.MultiTaskLasso.html#sklearn.linear_model.MultiTaskLasso The project sounds great, the hardest part might be finding the right mentor (Gael?) Cheers, Andy On 03/06/2015 07:57 PM, Luca Puggini wrote: Thanks a lot for the material provided on randomized pca and random forest it

Re: [Scikit-learn-general] Subject: Hyperparameters in scikit-learn

intuition about the problem. The parameter spaces that you want to search are probably different between GridSearchCV and a model-based approach, too. Do you have any examples or benchmarks available online? Cheers, Andy On 03/24/2015 03:50 PM, Matthias Feurer wrote: Dear scikit-learn team, After

Re: [Scikit-learn-general] GSoC2015 Improve GMM

rocesses. So far I think your proposal looks solid. It would be great if you could work on some pull requests to support your application. Best, Andy On 03/16/2015 04:23 PM, Wei Xue wrote: Hi groups, I am a PhD student in Florida International University, US

Re: [Scikit-learn-general] [GSoC 2015] Cross-validation and Meta-Estimators for semi-supervised learning

-learn/pull/3758 I was a bit surprised to see that the algorithms currently are iterative, when I think there is a closed form solution. Maybe you could look into that, too? If you have other related ideas, feel free to include them as well. Best, Andy On 03/22/2015 05:47 PM, Boyuan Deng wrote

Re: [Scikit-learn-general] 8.23.6. sklearn.neighbors.BallTree

f you have it installed, using help(BallTree). Andy On 03/24/2015 10:52 AM, nafise mehdipoor wrote: Hi, Would you please let me know about some documents of "Ball Tree" of version 0.14 or 0.15. When I search for this, I just find the link below which is for version 0.13.1 http://s

Re: [Scikit-learn-general] grid search random state

8:17 PM *To:* scikit-learn-general@lists.sourceforge.net *Subject:* Re: [Scikit-learn-general] grid search random state Thank you! *From:*Andy [mailto:t3k...@gmail.com] *Sent:* Wednesday, February 25, 2015 3:24 PM *To:* scikit-learn-general@lists.sourceforge.net <mailto:scikit-learn-gene

Re: [Scikit-learn-general] [ANN] scikit-learn 0.16b1 is out!

On 03/09/2015 10:44 PM, Joel Nothman wrote: Congratulations! This has been a long time coming, and if not only for the swathe of features it'll be great to see the documentation improvements appearing on stable soon! My thoughts on development priorities for the next release (and ideally to f

Re: [Scikit-learn-general] GSoC2015 Hyperparameter Optimization topic

On 03/09/2015 07:11 PM, Saket Choudhary wrote: > On 9 March 2015 at 15:42, Andreas Mueller wrote: >> We wanted a bot that tells us about violations on PRs. >> Not sure if landscape.io can provide that:\ >> https://github.com/scikit-learn/scikit-learn/issues/3888#issuecomment-76037183 >> >> ragv al

Re: [Scikit-learn-general] GSoC2015 Hyperparameter Optimization topic

old" Spearmint is GPL, "new" spearmint is under a non-commercial license. Best, Andy On 03/07/2015 07:39 AM, Christof Angermueller wrote: Hi Andreas (and others), I am a PhD student in Bioinformatics at the University of Cambridge, (EBI/EMBL), supervised by Oliver Stegle and

Re: [Scikit-learn-general] feature names after OneHotEncoder

Well, the columns after the OneHotEncoder correspond to feature values, not feature names, right? There is ``feature_indices_`` which maps each feature to a range of features in the encoded matrix. The features in the input matrix don't really have names in scikit-learn, as they are represented

Re: [Scikit-learn-general] ANN: scikit-image 0.11

Awesome, congratulations Stefan! On 03/05/2015 04:59 AM, Stefan van der Walt wrote: > Announcement: scikit-image 0.11.0 > = > > We're happy to announce the release of scikit-image v0.11.0! > > scikit-image is an image processing toolbox for SciPy that includes algo

Re: [Scikit-learn-general] SOLVED: Scaling a Subset of Features in SKLEARN

FYI in the context of transforming only certain fields, you might also be interested in this addition to FeatureUnion, which will handle parts of dicts or dataframes: https://github.com/scikit-learn/scikit-learn/pull/3886 It's still waiting for reviews, though... On 03/03/2015 05:46 PM, Jason Wo

Re: [Scikit-learn-general] SVM: Matlab vs skleanr

Matlab references the libsvm authors on working set selection for the svm implementation, so they seem to be either implementing the same, or also using libsvm. On 03/03/2015 03:53 PM, Pagliari, Roberto wrote: Has anybody ever compared Matlab SVM vs sklearn or libsvm? It’d be interested to

Re: [Scikit-learn-general] tolerance and cache_size

It is the libsvm default: -e epsilon : set tolerance of termination criterion (default 0.001) from http://www.csie.ntu.edu.tw/~cjlin/libsvm/ Larger cache size doesn't hurt, but I'd be surprised if you have runtime issues on many UCI datasets. On 03/03/2015 03:46 PM, Pagliari, Roberto wrote:

Re: [Scikit-learn-general] GSoC2015 topics

On 03/03/2015 11:31 AM, Artem wrote: There was a discussion on metric learning a while ago, and several people expressed interest to see (and contribute to) it in sklearn. But, it looks like that attempt di

Re: [Scikit-learn-general] Asking for tasks of summer code 2015

Hi Hank. We have not fully fleshed out all ideas, but there is a preliminary list here: https://github.com/scikit-learn/scikit-learn/wiki/Google-summer-of-code-%28GSOC%29-2015 If you are interested, you should also read the expectations carefully. Usually we expect students to start contributing

Re: [Scikit-learn-general] Scaling a Subset of Features in SKLEARN

Hi Jason. We don't have any support for groups or types of features currently, sorry. And you do need to convert all categorical features to one-hot encoded features for use with sklearn. The underlying issue is that we use numpy arrays as our main data structure, and they are not very easy to

Re: [Scikit-learn-general] SciPy 2015 Austin

few years.Just hit *submit* on my talk so we will see what > happens there. > > On Wed, Feb 25, 2015 at 6:57 PM, Jacob Vanderplas > wrote: >> Andy, Kyle – great to hear you're planning on going! >> Perhaps the two of you can team-up on a sklearn tutorial? I think i

Re: [Scikit-learn-general] Score function in Extra-Trees

PR welcome! On 02/26/2015 01:19 PM, Pierre-Luc Bacon wrote: Thanks ! Perhaps the documentation could be updated to make this clear ? Pierre-Luc On Tue, Feb 24, 2015 at 5:24 AM, Arnaud Joly > wrote: Hi Pierre-Luc, This is the same criterion, but with a differ

Re: [Scikit-learn-general] Self Organizing Maps

The thing is: I have rarely seen a compelling use. If you have a use-case where the algorithms you listed out-perform things that are already in scikit-learn, I think we'd be happy to adopt them. Until now, I haven't seen a great application. -

Re: [Scikit-learn-general] SciPy 2015 Austin

#x27;t more > seasoned hands, but am also ok with helping refine the material from > the past few years.Just hit *submit* on my talk so we will see what > happens there. > > On Wed, Feb 25, 2015 at 6:57 PM, Jacob Vanderplas > wrote: >> Andy, Kyle – great to hear you're

Re: [Scikit-learn-general] Confidence value for each sample: Random Forest

That looks fine. in line 125, can you try assert(np.all(np.argmax(y_score, axis=-1) == y_pred)) That should go through. On 02/25/2015 05:38 PM, shalu jhanwar wrote: Hi Andy, please find this version of the code in which I changed the refit issue. thanks! Shalu On Wed, Feb 25, 2015 at 11

Re: [Scikit-learn-general] Confidence value for each sample: Random Forest

Wed, Feb 25, 2015 at 9:32 PM, Andy mailto:t3k...@gmail.com>> wrote: You fit the data again before calling predict_proba. You did not fix the random seed, so the outcome of the fit will be different and you can't expect it to be consistent. Just remove

Re: [Scikit-learn-general] SVC.predict_proba result inconsistent with SVC.predict result

I just told you that the repeated call to fit is the reason. -- Dive into the World of Parallel Programming The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for al

Re: [Scikit-learn-general] Confidence value for each sample: Random Forest

You fit the data again before calling predict_proba. You did not fix the random seed, so the outcome of the fit will be different and you can't expect it to be consistent. Just remove the second call to fit. On 02/25/2015 06:35 AM, shalu jhanwar wrote: Hey Guys, I am using Random forest clas

Re: [Scikit-learn-general] grid search random state

On 02/24/2015 08:26 PM, Pagliari, Roberto wrote: I have two questions about gridsearchcv 1.Is it possible to fix the random state of the underlying kfold, for testing purposes? 2.When passing parameters, such as C and gamma for svm, does grid search go through them in order? 1. Yes. By

Re: [Scikit-learn-general] SVC.predict_proba result inconsistent with SVC.predict result

Hi Shalu. Can you give your code. The prediction is just the argmax of predict_proba, so I'd be very surprised if they are not consistent. Cheers, Andy On 02/25/2015 08:33 AM, shalu jhanwar wrote: Hi all, I'm facing the same problem with predict_proba for Random_forest classifi

[Scikit-learn-general] SciPy 2015 Austin

Hey everybody. Is anyone going to / submitting talks to scipy? My institute (or rather Moore-Sloan) is a sponsor so they'll sent me :) Cheers, Andy -- Dive into the World of Parallel Programming The Go Parallel We

Re: [Scikit-learn-general] Re : Pull Request : Renyi entropy and Cauchy-Schwartz mutual information

ature selector would be welcome, I think. Well, by me, anyway. On 23 February 2015 at 09:37, Andy <mailto:t3k...@gmail.com>> wrote: Hi Cecilia. An MI estimate currently seems a bit out of scope of sklearn. What context would a user apply it in? Sklearn currently contains m

Re: [Scikit-learn-general] GSoC2015 topics

bly by ragv data-independent cross-validation : already a bunch of people working on that, I don't think we should make it GSOC. Feedback welcome. Andy -- Dive into the World of Parallel Programming The Go Par

Re: [Scikit-learn-general] Re : Pull Request : Renyi entropy and Cauchy-Schwartz mutual information

Hi Cecilia. An MI estimate currently seems a bit out of scope of sklearn. What context would a user apply it in? Sklearn currently contains more out-of-the-box methods, while an MI estimator seems more like a building block. Cheers, Andy On 02/23/2015 06:01 AM, cécilia wrote: Hi, May you

Re: [Scikit-learn-general] Perceptron implementation: Perceptron Rule or Stochastic Gradient Descent?

So indeed in the perceptron update yi_pred is {-1, 1}, not real, in sklearn, right? On 02/23/2015 08:35 AM, Mathieu Blondel wrote: Rosenblatt's Perceptron is a special case of SGD, see: https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/linear_model/tests/test_perceptron.py The

Re: [Scikit-learn-general] set_params and get_params (and 1.0 API)

arams. On 20 February 2015 at 09:17, Andy <mailto:t3k...@gmail.com>> wrote: On 02/19/2015 12:58 PM, Gael Varoquaux wrote: > The question is: can we do this without breaking our pipeline delegation > mechanism that we use to set parameters during cross-validation in a

Re: [Scikit-learn-general] GSoC2015 topics

..@normalesup.org>> wrote: On Tue, Feb 17, 2015 at 04:42:11PM -0800, Andy wrote: > On 02/13/2015 07:08 AM, Ronnie Ghose wrote: > > -1 we would have to build in support for more clustering methods > > ,sounds like a not-very-standalone proj > Why? We alre

Re: [Scikit-learn-general] Regarding viewing the decision boundaries of classifiers

I think it is. The last character of my answer was "_". That is the variable name to which you assign the thresholds in your code. On 02/22/2015 01:54 PM, shalu jhanwar wrote: Hi Andy, thanks for the reply. I guess your second question's ans. is not completely displayed in t

Re: [Scikit-learn-general] Regarding viewing the decision boundaries of classifiers

On 02/22/2015 01:21 PM, shalu jhanwar wrote: Hi guys, thanks a lot for all your interesting replies. i) How can I get threshold value which the classifier has decided to take the decision for a particular sample to be in 0 or 1 class in binary classification using scikit? The whole purpose o

  1   2   3   4   >