2011/10/4 Gael Varoquaux <[email protected]>: > By the way, why we are modifying this `permutation_test_score` function, > it seems to me that the `score_func` argument should be optional, and > that without it, the `permutation_test_score` should try to use the > estimator's score. In such a situation, as such a score should always be > `bigger is better`, people shouldn't fall in the trap in which Satra has > fallen. > > What do people think? Should I implement that (trivial) change?
It returns several stuff: the array of scores and the p-value. In the array of scores the "bigger is better" rule is respected. But yeah this is not a real "_score" function as it does not return a single scalar float. The same holds for cross_val_score which also returns an array. Maybe we should just rename then to "permutation_test_scores" and "cross_validation_scores" (plural) to make this more specific? -- Olivier http://twitter.com/ogrisel - http://github.com/ogrisel ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity and more. Splunk takes this data and makes sense of it. Business sense. IT sense. Common sense. http://p.sf.net/sfu/splunk-d2dcopy1 _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
