I think what I'm going to do for this is: Create the o.a.m.math.stats package and put it under there.
On Dec 30, 2009, at 12:26 PM, Grant Ingersoll wrote: > For MAHOUT-163, which I'm about to commit (after some refactoring), there is > a log-likelihood calculation in the ClusterLabels section (see > https://issues.apache.org/jira/secure/attachment/12429141/MAHOUT-163.patch). > Seems like it should be generalized for our Math library. > > If/how should this be made generally applicable and where should it live? > > -Grant
