Thank you David for phrasing well exactly my thoughts:

 * Bayesian inference != Bayes Net (Graphical models)

 * To solve Bayesian problems efficiently with sampling requires quite an
   infrastructure, PyMC does this very well, let us not replicate this.
   In specific cases of Bayesian inference without sampling, it
   perfectly fits in the scope of the scikit, and we already have some 
   code (mixture models, or Gaussian processes, for instance)

 * Graphical models: why not (I actually do some Gaussian graphical
   models research that, I hope, will benefit to the scikit one day), but
   as usual, ironing the code takes a while.

My 2 cents,

Gaël

------------------------------------------------------------------------------
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

Reply via email to