Re: [Scikit-learn-general] feature union

You can even just edit the file directly at https://github.com/scikit-learn/scikit-learn/blob/master/examples/feature_stacker.py On 8 October 2014 08:16, Lars Buitinck wrote: > 2014-10-07 23:03 GMT+02:00 Pagliari, Roberto : > > Do I just use the bug tracker? > > You can, but we'd much rather hav

Re: [Scikit-learn-general] feature union

2014-10-07 23:03 GMT+02:00 Pagliari, Roberto : > Do I just use the bug tracker? You can, but we'd much rather have a patch :) See https://github.com/scikit-learn/scikit-learn/blob/master/CONTRIBUTING.md on how to create one. (The change is tiny, so you don't need to read the whole thing.) ---

Re: [Scikit-learn-general] feature union

Do I just use the bug tracker? From: Joel Nothman [mailto:[email protected]] Sent: Tuesday, October 07, 2014 4:35 PM To: scikit-learn-general Subject: Re: [Scikit-learn-general] feature union I don't think it should be fit. You can create a PR to remove it, afaik. On 8 October 2014 04:48,

Re: [Scikit-learn-general] feature union

I don't think it should be fit. You can create a PR to remove it, afaik. On 8 October 2014 04:48, Pagliari, Roberto wrote: > I read this page on the documentation > > > > http://scikit-learn.org/stable/auto_examples/feature_stacker.html > > > > why is svm.fit needed before gridsearch? > > > > Th

Re: [Scikit-learn-general] cross_val_score crashes python every time

Gael - I tried separating out the lambda function into a standalone function. Unfortunately no luck - same result :(. Olivier - I'm using the default, which is n_jobs=1, so hopefully that shouldn't be a problem. The reason I'm using the threads is that I want to be able to train models using a v

[Scikit-learn-general] feature union

I read this page on the documentation http://scikit-learn.org/stable/auto_examples/feature_stacker.html why is svm.fit needed before gridsearch? Thanks, -- Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer

[Scikit-learn-general] Regarding ROC from multiple classifiers

Hi Scikit Users, I want to make a ROC curve to compare various classifiers. Can anybody tell me how to plot it? I am already familiar of plotting ROC using cross-validation. But I am unable to plot a single ROC having all the classifiers to compare the performance. Many thanks! Shalu ---

Re: [Scikit-learn-general] Scikit-learn-general Digest, Vol 57, Issue 5

Thanks Manoj and others - very useful Dominic Steinitz [email protected] http://idontgetoutmuch.wordpress.com On 7 Oct 2014, at 11:20, [email protected] wrote: > > Message: 3 > Date: Mon, 6 Oct 2014 22:51:32 +0200 > From: Manoj Kumar > Subject: Re: [Scikit-

Re: [Scikit-learn-general] Better results with R than with Scikit

Thanks for all the suggestions, I figure out was happening. In R we were training with all the data(train and test) and later evaluating on test :) Sorry. Now the results are very similar(0.73 vs 0.74). Andy, for the results I am using clf.predict_proba. Thanks. El 06/10/14 20:13, "Andy" escribi

Re: [Scikit-learn-general] Scikit-learn-general Digest, Vol 57, Issue 5

2014-10-07 11:05 GMT+02:00 Olivier Grisel : > I would be more precise and say Python's multiprocessing module can > only be used on picklable callables (e.g. functions or classes with a > __call__ method) applied to picklable arguments. Even more precisely, the problem is sending a lambda to a for

Re: [Scikit-learn-general] Scikit-learn-general Digest, Vol 57, Issue 5

2014-10-07 7:41 GMT+02:00 Gael Varoquaux : > On Mon, Oct 06, 2014 at 05:35:12PM -0400, Alan G Isaac wrote: >> On 10/6/2014 Gael Varoquaux wrote: >> > Parallel computing has problems with lambda functions. > >> Can you elaborate on that please? > > Lambdas don't pickle. Parallel computing needs pick