Well actually I've made X and y to both be binary, as in they are just the
presence or absence or a feature/class in the vector(s) so they should be fine
to pass into mutual_info_score.
--------------------------------
Matti Lyra
DPhil Student
Text Analytics Group
Chichester 1, R203
School of Engineering and Informatics
University of Sussex
Brighton, UK
[email protected]
Tel: +441273 872956
On 28 Aug 2013, at 15:30, Olivier Grisel wrote:
> Those two functions are apparently not computing the same thing. The
> mutual_info_score function is a clustering quality evaluation tool
> used to compute the mutual information between 2 sets of integer
> cluster label assignment. At least some of the integer label values
> must match for the score to be non zero. The goal is to assess how
> much 2 clusterings approximately agree on how to split the data.
>
> As the data in X is a floating point representation of features (e.g.
> TF-IDF features for text) it does not make sense to pass it as an
> argument to mutual_info_score.
>
> --
> Olivier
>
> ------------------------------------------------------------------------------
> Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
> Discover the easy way to master current and previous Microsoft technologies
> and advance your career. Get an incredible 1,500+ hours of step-by-step
> tutorial videos with LearnDevNow. Subscribe today and save!
> http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
------------------------------------------------------------------------------
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incredible 1,500+ hours of step-by-step
tutorial videos with LearnDevNow. Subscribe today and save!
http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
