beast-dev / tracer

Posterior summarisation in Bayesian phylogenetics
https://beast.community/tracer
GNU Lesser General Public License v2.1
55 stars 17 forks source link

Categorical trace 'frequency of mode' should be divided by the number of states #45

Closed rambaut closed 7 years ago

rambaut commented 7 years ago

A proportion would be more useful.

walterxie commented 7 years ago

@rambaut Do you mean the number of samples? I added "mode probability" which equals to "mode frequency" divided by "sample size".