PoissonLoss {PeakSegOptimal}

PoissonLoss

Description

Compute the weighted Poisson loss function, which is seg.mean - count * log(seg.mean). The edge case is when the mean is zero, in which case the probability mass function takes a value of 1 when the data is 0 (and 0 otherwise). Thus the log-likelihood of a maximum likelihood segment with mean zero must be zero.

Usage

PoissonLoss(count, seg.mean, 
    weight = 1)

Arguments

count

count

seg.mean

seg.mean

weight

weight

Author(s)

Toby Dylan Hocking

Examples

PoissonLoss(1, 1)
PoissonLoss(0, 0)
PoissonLoss(1, 0)
PoissonLoss(0, 1)