Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Hi, >Thanks! Could you also run a quick grid search to fine tune gamma and >C for all models independently? I have updated the code on the gist to match your requirement. I made a 5 fold splits where on each the parameters are finely estimated for both KPLS and SVC using gridsearchCV. In conclusi

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

2013/12/5 abdalrahman eweiwi : > Hi, > > I have a problem of getting GridSearchCV working with kpls for fine tunning > its parameters. I have reviewed the documentation for how to get a new > estimator work with GridSearchCV but I did not find enough information > there. I pursued myself on this bu

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Hi, I have a problem of getting GridSearchCV working with kpls for fine tunning its parameters. I have reviewed the documentation for how to get a new estimator work with GridSearchCV but I did not find enough information there. I pursued myself on this but finally I got stuck with a problem relat

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Thanks! Could you also run a quick grid search to fine tune gamma and C for all models independently? Another question: what is the time complexity for KPLS w.r.t. n_samples? quadatratic, cubic? Could you please extend your gist to randomly subsample the digits dataset to only keep 50% and re-run

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Hi, Here is a public gist of KPLS, https://gist.github.com/abdhk383/7788156 Regards, Eweiwi On Tue, Dec 3, 2013 at 12:51 PM, abdalrahman eweiwi < abdalrahman.ewe...@gmail.com> wrote: > > Hi, > > Here is a preliminary results on classification performance of KPLS using > a 20 fold cross validat

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Hi, Here is a preliminary results on classification performance of KPLS using a 20 fold cross validation with random splits of 0.5 train and 0.5 test for the digits dataset using SVC, linearSVC and KPLS. I used the same kernel parameters (rbf, gamma=0.001) of this example for SVC and KPLS: http://

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

2013/12/2 abdalrahman eweiwi : > Hi, > > You are right, infact I spent almost 1 month reviewing the code base of PLS > and CCA implementation in sklearn. I should say that the (old) code base in > my opinion should be somehow refactored to get into a simpler shape. I > remember I had some difficult

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Hi, You are right, infact I spent almost 1 month reviewing the code base of PLS and CCA implementation in sklearn. I should say that the (old) code base in my opinion should be somehow refactored to get into a simpler shape. I remember I had some difficulties in analyzing that code. Also the CCA r

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

> First I would like to thank you all for this great and handy machine > learning toolkit. Second, would it be of an interest to add an > implementation for Kernel Partial Least Squares for the library? Nelle Varoquaux (my sister, @NelleV on github) wants to do a Kernel CCA, which is heavily relat

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Hi, Ok then, I will evaluate the classification result on the digits dataset and compare it with the linearSVC and SVC with gaussian kernel. For regression I will consider the Boston dataset, one remark here is what error criteria should I use? square loss or do you suggest something else. Regard

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

2013/11/29 abdalrahman eweiwi : > Well, Kernel partial least squares can be used for regression as well as for > classification. It has been shown to work as good as Gaussian processes for > regression on head pose estimation applications. See for example this CVPR > paper > > http://iselab.cvc.ua

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

Well, Kernel partial least squares can be used for regression as well as for classification. It has been shown to work as good as Gaussian processes for regression on head pose estimation applications. See for example this CVPR paper http://iselab.cvc.uab.es/files/Publications/2012/PDF/AGD2012.pd

Re: [Scikit-learn-general] Contributing with Kernel partial least squares

2013/11/29 abdalrahman eweiwi : > Hi, > > First I would like to thank you all for this great and handy machine > learning toolkit. Second, would it be of an interest to add an > implementation for Kernel Partial Least Squares for the library? Hi Eweiwi, I am not familiar with Kernel PLS myself. Wh

[Scikit-learn-general] Contributing with Kernel partial least squares

Hi, First I would like to thank you all for this great and handy machine learning toolkit. Second, would it be of an interest to add an implementation for Kernel Partial Least Squares for the library? Thanks Eweiwi -- Rap