I'm using Randomise and I'm curious as to the difference of using binary vs weighted masks. To date I have only used binary, I then threshold my corrp images at .95.
Is there a statistical benefit to using the weighted mask?
If so I would I threshold the output?
|