🌐 Mailman 3 February 2018 - scikit-learn - python.org

Applying clustering to cosine distance matrix
by prince gosavi Feb. 12, 2018

Feb. 12, 2018

I have generated a cosine distance matrix and would like to apply clustering algorithm to the given matrix. np.shape(distance_matrix)==(14000,14000) I would like to know which clustering suits better and is there any need to process the data further to get it in the form so that a model can be applied. Also any performance tip as the matrix takes around 3-4 hrs of processing. You can find my code here https://github.com/maxyodedara5/BE_Project/blob/master/main.ipynb Code for READ ONLY PURPOSE. -- Regards

2 1

PipeGraph examples: areas of interest
by Manuel Castejón Limas Feb. 10, 2018

Feb. 10, 2018

Hi all! The good news is that we made GridSearchCv work on PipeGraph! In order to create diverse examples, we welcome some feedback on which other libraries you use in order to acquire/process data before applying scikit learn. For example: 'I work in computer vision and I usually get image feature using the python bindings provided by OpenCV.' This will help us provide interesting examples. Best Manuel

1 0

(no subject)
by Erik Fransson Feb. 9, 2018

Feb. 9, 2018

Hi everyone, I have a simple questions in regards to how LassoCV works. In the documentation it states that the best model is selected via cross-validation, however I'm wondering how is this best model constructed. Is it simply running a normal lasso fit with all of the training data available using the optimal alpha? And are these the parameters stored in .coef_? Regards, Erik

2 1

Re: [scikit-learn] clustering on big dataset
by Manuel Castejón Limas Feb. 7, 2018

Feb. 7, 2018

Hope this helps! Manuel @Article{Ciampi2008, author="Ciampi, Antonio and Lechevallier, Yves and Limas, Manuel Castej{\'o}n and Marcos, Ana Gonz{\'a}lez", title="Hierarchical clustering of subpopulations with a dissimilarity based on the likelihood ratio statistic: application to clustering massive data sets", journal="Pattern Analysis and Applications", year="2008", month="Jun", day="01", volume="11", number="2", pages="199--220", abstract="The problem of clustering subpopulations on the basis of samples is considered within a statistical framework: a distribution for the variables is assumed for each subpopulation and the dissimilarity between any two populations is defined as the likelihood ratio statistic which compares the hypothesis that the two subpopulations differ in the parameter of their distributions to the hypothesis that they do not. A general algorithm for the construction of a hierarchical classification is described which has the important property of not having inversions in the dendrogram. The essential elements of the algorithm are specified for the case of well-known distributions (normal, multinomial and Poisson) and an outline of the general parametric case is also discussed. Several applications are discussed, the main one being a novel approach to dealing with massive data in the context of a two-step approach. After clustering the data in a reasonable number of `bins' by a fast algorithm such as k-Means, we apply a version of our algorithm to the resulting bins. Multivariate normality for the means calculated on each bin is assumed: this is justified by the central limit theorem and the assumption that each bin contains a large number of units, an assumption generally justified when dealing with truly massive data such as currently found in modern data analysis. However, no assumption is made about the data generating distribution.", issn="1433-755X", doi="10.1007/s10044-007-0088-4", url="https://doi.org/10.1007/s10044-007-0088-4" } 2018-01-04 12:55 GMT+01:00 Joel Nothman <joel.nothman(a)gmail.com>: > Can you use nearest neighbors with a KD tree to build a distance matrix > that is sparse, in that distances to all but the nearest neighbors of a > point are (near-)infinite? Yes, this again has an additional parameter > (neighborhood size), just as BIRCH has its threshold. I suspect you will > not be able to improve on having another, approximating, parameter. You do > not need to set n_clusters to a fixed value for BIRCH. You only need to > provide another clusterer, which has its own parameters, although you > should be able to experiment with different "global clusterers". > > On 4 January 2018 at 11:04, Shiheng Duan <shiduan(a)ucdavis.edu> wrote: > >> Yes, it is an efficient method, still, we need to specify the number of >> clusters or the threshold. Is there another way to run hierarchy clustering >> on the big dataset? The main problem is the distance matrix. >> Thanks. >> >> On Tue, Jan 2, 2018 at 6:02 AM, Olivier Grisel <olivier.grisel(a)ensta.org> >> wrote: >> >>> Have you had a look at BIRCH? >>> >>> http://scikit-learn.org/stable/modules/clustering.html#birch >>> >>> -- >>> Olivier >>> >>> >>> _______________________________________________ >>> 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 > >

1 0

Kernel Extreme Learning Machine in Python 3
by Carlos Perales Feb. 7, 2018

Feb. 7, 2018

Dear all, After working with Extreme Learning Machine classifiers for a while in MATLAB, I think it is a surprisingly good supervised classifier for multiclass problems. Although originally it was designed as a single hidden layer neural network with some randon weights, from 2012 it has a similar mathematical formulation than SVM (DOI: https://doi.org/10.1109/TSMCB.2011.2168604 , download link: http://www.neuromorphs.net/nm/raw-attachment/wiki/2015/scc15/ELM-Unified-Le… ). This allows to use ELM with kernels, just as SVM, but avoiding "one-by-one" and "one-against-rest" techniques. Pull request is 10602 ( https://github.com/scikit-learn/scikit-learn/pull/10602 ), and all contributions are welcome! Un saludo, Carlos Perales.

1 0

Jeff Levesque: custom json encoder
by Jeffrey Levesque Feb. 6, 2018

Feb. 6, 2018

Hi, Was wondering if anyone has written custom json encoder for sklearn classes, like SVC, or ensemble classes. Is it even possible to json encode, with a custom encoder for ensemble classes, like bagging techniques? Thank you, Jeff Levesque https://github.com/jeff1evesque

2 2

Need Help with Failing Travis/Appveyor Build
by Yacine MAZARI Feb. 5, 2018

Feb. 5, 2018

Hello, I added some additional unit tests to this PR <https://github.com/scikit-learn/scikit-learn/pull/10581>, which required me to do > from numpy.testing import assert_array_compare > This caused the build on both Travis <https://travis-ci.org/scikit-learn/scikit-learn/builds/337229008?utm_source…> and Appveyor <https://ci.appveyor.com/project/sklearn-ci/scikit-learn/build/1.0.19893> to fail with > ImportError: cannot import name assert_array_compare > I don't have this issue on my locale (Ubuntu 16.04, Python 2.7). Any hints? Thank you.

2 1

Weighted feature sampling in Random Forest
by RAPHAEL RODRIGUES CAMPOS Feb. 5, 2018

Feb. 5, 2018

Hello Scikit community, I want the Random Forest to be able to draw the candidate features for split using weighted sampling. Basically, I want to reproduce this paper's results https://pdfs.semanticscholar.org/9b2f/84d85e5b6979bf375a2d4b15f7526597fc70.… I wonder whether I only need to change the function on the following line: https://github.com/scikit-learn/scikit-learn/blob/6919a22c8023f4c2f30849c7c…, and permutate the weight of the features as well. Or it would be more complicated. Could someone shed a light on this? Thanks in advance.

1 0