Re: [Scikit-learn-general] gridSearchCV best_estimator_ best_score_

Hi Andy, Maybe it's just me, but the "left out data" threw me off. Perhaps, I would integrate with your previous comments: best_estimator_ estimator Estimator that was chosen by the search, i.e. estimator which gave highest average score (or smallest loss if specified) over the cross-validatio

Re: [Scikit-learn-general] GridSearchVC with SVM

Hi Andy, Thanks for the help, but I really just had issues with linux. I reinstall everything and it is working now. Cheers, From: Andy [mailto:[email protected]] Sent: Friday, July 25, 2014 3:58 PM To: [email protected] Subject: Re: [Scikit-learn-general] GridSearchVC wi

Re: [Scikit-learn-general] SGDClassifier with class_weight=auto fails on linux, but not on osx

This should really not happen. Can you please open an issue with sample code? Thanks, Andy On 07/25/2014 03:12 AM, Rose Perrone wrote: When I train an scikit-learn |SGDClassifier| with these options:

Re: [Scikit-learn-general] GridSearchVC with SVM

On 07/23/2014 06:21 PM, Pagliari, Roberto wrote: Hi Michael, Thanks for the clarifications. Is there a way to make prediction, once grid search is done? Right now I'm getting the error 'GridSearchCV' object has no attribute 'best_estimator_' And I've seen other people reporting the same e

Re: [Scikit-learn-general] ElasticNet for classification

On 07/23/2014 03:21 AM, Mathieu Blondel wrote: from sklearn.multiclass import OneVsRestClassifier clf = OneVsRestClassifier(ElasticNet()) But that would be trained using rmse loss. Why would you do that if we have logistic loss and hinge loss in SGDClassifier? should work. This is tested her

Re: [Scikit-learn-general] gridSearchCV best_estimator_ best_score_

The best_score_ is the mean score of the best parameter averaged over folds. best_estimator_ is the estimator with these parameters retrained on the whole training set. If this is not clear from the docstring, what alternative would you propose? Best, Andy On 07/25/2014 09:32 PM, Pagliari, Rob

[Scikit-learn-general] gridSearchCV best_estimator_ best_score_

I'm not sure if I understand the documentation of gridSearchCV: best_estimator_ estimator Estimator that was chosen by the search, i.e. estimator which gave highest score (or smallest loss if specified) on the left out data. best_score_ float Score of best_estimator on the left out data.

Re: [Scikit-learn-general] SGDClassifier with class_weight=auto fails on linux, but not on osx

The workaround succeeded. Thanks, Danny! I posted a github issue. On Fri, Jul 25, 2014 at 2:36 AM, Danny Sullivan wrote: > Hi Rose, > > That's a good question. Is your y in this case string labels? I believe > this is bug that occurs because label encoding is happening both in the > SGDClassif

[Scikit-learn-general] Windows CI is back online

Yesterday I fixed a couple of bugs in the setup of the windows CI and re-enabled it for the master and 0.15.X branches and all incoming pull requests to those branch. https://ci.appveyor.com/project/ogrisel/scikit-learn/history It's working well now. There is very few failures caused by the AppVe

Re: [Scikit-learn-general] SGDClassifier with class_weight=auto fails on linux, but not on osx

Hi Rose, That's a good question. Is your y in this case string labels? I believe this is bug that occurs because label encoding is happening both in the SGDClassifier and the compute_class_weight function. I've posted a work-around on your stackoverflow question so you can go ahead and give that