Re: [Scikit-learn-general] ANOVA SVM pipeline and cross validation

gt; Outside the framework, everything is possible by calling the "fit", > "transform", and "predict" methods of the various objects. > > Gaël > > On Fri, Jan 15, 2016 at 07:55:45PM +0100, Fabrizio Fasano wrote: >> Thank a lot, Andreas, >

Re: [Scikit-learn-general] ANOVA SVM pipeline and cross validation

an a-priori? Or there are other (scikit learn supported) CV methods? Thank you very much again, Best, Fabrizio On Jan 15, 2016, at 7:44 PM, Andreas Mueller wrote: > > > On 01/15/2016 01:16 PM, Fabrizio Fasano wrote: >> Dear community, >> >> I would like to

[Scikit-learn-general] ANOVA SVM pipeline and cross validation

Dear community, I would like to use ANOVA + SVM pipeline to check 2 group classification performances of neuroimaging datasets, My questions are: 1) In pipeline approach implemented by Scikit-learn (http://scikit-learn.org/stable/auto_examples/svm/plot_svm_anova.html) is the cross validation

[Scikit-learn-general] permutation test score with LeaveOneOut

Dear community, I was wondering if it is possible to combine permutation test score with leaveOneOut cross validation My code: >>>loo = cross_validation.LeaveOneOut(len(age)) >>> score, permutation_scores, pvalue = >>> cross_validation.permutation_test_score(svr_rbf, ALL,age, >>> scoring='mea

Re: [Scikit-learn-general] Dramatic improvement by standardizing data?

mean > and sdev on the training set and standardize the test set using those > estimated values. If this method worsens your results, there may be an > unaccounted-for trend in your data. > > Michael > > > On Thu, Apr 30, 2015 at 10:32 AM, Fabrizio Fasano > mailto:fa

Re: [Scikit-learn-general] Dramatic improvement by standardizing data?

t valid as that includes > the test data. > > It is hard to say whether 100% is believable or not, but you should > probably only take scaling over training data. > > On Wed, Apr 29, 2015 at 11:13 AM, Fabrizio Fasano wrote: >> Dear experts, >> >> I’m experiencing a

Re: [Scikit-learn-general] Dramatic improvement by standardizing data?

lievable or not, but you should > probably only take scaling over training data. > > On Wed, Apr 29, 2015 at 11:13 AM, Fabrizio Fasano wrote: >> Dear experts, >> >> I’m experiencing a dramatic improvement in cross-validation when data are >> standardised >>

Re: [Scikit-learn-general] Dramatic improvement by standardizing data?

t > implementations. > > Best, > Sebastian > >> On Apr 29, 2015, at 11:13 AM, Fabrizio Fasano wrote: >> >> Dear experts, >> >> I’m experiencing a dramatic improvement in cross-validation when data are >> standardised >> >>

[Scikit-learn-general] Dramatic improvement by standardizing data?

Dear experts, I’m experiencing a dramatic improvement in cross-validation when data are standardised I mean accuracy increased from 48% to 100% when I shift from X to X_scaled = preprocessing.scale(X) Does it make sense in your opinion? Thank You a lot for any suggestion, Fabrizio my COD

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

that is where that comes from. > If you repeat over different assignments, you will get 50/50. > > On 04/27/2015 11:33 AM, Fabrizio Fasano wrote: >> Dear Andy, >> >> Yes, the classes have the same size, 8 and 8 >> >> this is one example of code I used to cr

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

tell. > Are you sure the classes have the same size? > > On 04/26/2015 11:22 AM, Fabrizio Fasano wrote: >> Dear Andreas, >> >> Thanks a lot for your help, >> >> about the random assignment of values to my labels y. What I mean is that >> being suspici

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

Hi Sebastian, Thank You for your answer, what I mean is that by using a cross validation test I get 100% accuracy (on the testing set, not on the training set). It seemed to me a too good result, thus I changed the y labels (I mean, I replaced the true labels with false ones) to check that, a

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

Dear Andreas, Thanks a lot for your help, about the random assignment of values to my labels y. What I mean is that being suspicious about the too good performances, I changed the labels manually, retaining the 50% 1,0 but in different orders, and the labels were always predicted very well, wi

[Scikit-learn-general] bias in svm.LinearSVC classification accuracy in very small data sample?

Dear community, I'm performing a binary classification on a very small data set: details: -binary classification (Y=0,1) -small dataset (16 samples) -large features set (112 features) -balanced labels (y=0 and y=1 occur 8 times each) -linear SVM classifier. accuracy was 100% when tested on the t