Component
data class ComponentCountRow(val numComponents: Int, val logLikelihood: Double?, val marginalGain: Double?, val penaltyIncrement: Double?, val criterionValues: Map<String, Double>, val isFeasible: Boolean)(source)
What is known about one candidate number of components.
Parameters
num
the candidate count
log
the observed-data log-likelihood of the best mixture at this count
marginal
the improvement in log-likelihood over the next count down. May be negative, and that is not a fault — see ComponentCountEvidence
penalty
what the ranking criterion charges for the extra component
criterion
each reported criterion's value at this count
is
whether the count could be fitted at all