Jake, it looks like you mixed up precision and recall::
>>> print TP / float(TP + FN) # recall 0.948113562768 >>> print TP / float(TP + FP) # precision 0.142337086782 2012/2/28 Peter Prettenhofer <[email protected]>: > Hi Jake, > > the tutorial looks great! Unfortunately, I haven't got the time to go > through all of it yet, however, I spotted something that confused me: > in the last paragraph of Section 2.3.4 you state: "Of the points we > label quasars, only 14% of them are correctly labeled." By "we labeled > quasars" you mean those points who's true label is positive (=quasar) > and not who's predicted label is positive? Because the latter would > refer to precision rather than recall. Maybe you should change the > wording to avoid confusion. > > BTW: "we are correctly identifying 95% of all quasars." should > actually be "among the points identified as quasars 95% are correct" > because the former would be recall rather than precision. > > best, > Peter > > 2012/2/27 Jacob VanderPlas <[email protected]>: >> Thanks for all the feedback. >> I pushed an update this morning which addressing some of the easy fixes >> that were brought up, as well as adding the final two exercises. Thanks! >> http://jakevdp.github.com/tutorial/astronomy/exercises.html >> Jake >> >> Lars Buitinck wrote: >>> 2012/2/27 Jacob VanderPlas <[email protected]>: >>> >>>> If you have a few minutes to look it over, I'd really appreciate some >>>> feedback as I add the finishing touches this week. Also, as there's no >>>> way I'll get through all of it in a two hour tutorial, I'd like feedback >>>> on which parts you think I should focus on! >>>> >>> >>> Looks nice! >>> >>> Any reason not to use the sklearn.metrics module for evaluation? >>> >>> In "2.3.5.2. A Simple Method: Decision Tree Regression" you announce >>> an NN method, but then actually use a decision tree. >>> >>> >> >> ------------------------------------------------------------------------------ >> Try before you buy = See our experts in action! >> The most comprehensive online learning library for Microsoft developers >> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, >> Metro Style Apps, more. Free future releases when you subscribe now! >> http://p.sf.net/sfu/learndevnow-dev2 >> _______________________________________________ >> Scikit-learn-general mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general > > > > -- > Peter Prettenhofer -- Peter Prettenhofer ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
