While diving into warpSet, I noticed the jittering. The comment claims adding %1 of sd aka a small amount of noise. Unfortunately, sd is not calculated and the doc of jitter says that the default leads to d/5 where d is about the smallest difference between x values (factor being 1). When there are 2 landmarks, jitter generates a uniform random number between -20% and +20% of the gap between landmarks, which quite huge IMO.
I know you are considering warpSet and gaussNorm as unsupported, but I wanted to let you know.
Hi,
While diving into warpSet, I noticed the jittering. The comment claims adding %1 of sd aka a small amount of noise. Unfortunately, sd is not calculated and the doc of jitter says that the default leads to
d/5 where d is about the smallest difference between x values(factor being 1). When there are 2 landmarks, jitter generates a uniform random number between -20% and +20% of the gap between landmarks, which quite huge IMO.I know you are considering warpSet and gaussNorm as unsupported, but I wanted to let you know.
I added
, factor = 0.1for testing.Best.
https://github.com/RGLab/flowStats/blob/fabb13191a2567e41a89691cab31507d4cbddc7a/R/warpSet.R#L766-L767