Re: [Scikit-learn-general] Issues with performance metrics

On Tue, Nov 12, 2013 at 04:44:49PM +0100, Juan Chacon wrote: > I just updated to V0.14 and when im trying to make cross validation using > mean_squared_error, i got negative values... It is the negated mean squared error, as it should be 'bigger is better'. Team: should we just call that thing 'n

Re: [Scikit-learn-general] Trouble with sklearn LDA

Gabriel, as I already said you're using a classifier for a regression problem. Please read the doc. Maybe : http://scikit-learn.org/stable/tutorial/statistical_inference/supervised_learning.html this help. hint : replace LDA by sklearn.linear_model.Ridge for example. HTH Alex On Sat, Nov 9,

[Scikit-learn-general] Issues with performance metrics

Hey guys I just updated to V0.14 and when im trying to make cross validation using mean_squared_error, i got negative values... I checked the metrics.py file, and the file seems to be correct, but somehow, a negative value jumps into the final result when using sklearn.cross_validation.cross_val