this may be a naive question but is it possible to specify my own objective function using one of the existing methods?
For example, based on the example given in the vignette, let's say i want to use my own objective but instead of my own algorithm, i want to use the method 'snmf/l'. If i run this:
n <- 50
r <- 3
p <- 20
V <-syntheticNMF(n, r, p)
V_model <- nmf(V, 3, method = 'snmf/l',
objective = function(model, target, ...){
( sum( (target-fitted(model))^4 ) )^{1/4}
}
)
it fails as this method already has its own objective baked in.
Can i somehow edit an existing method so i can use my own objective based on the model and fitted objects?
this may be a naive question but is it possible to specify my own objective function using one of the existing methods?
For example, based on the example given in the vignette, let's say i want to use my own objective but instead of my own algorithm, i want to use the method 'snmf/l'. If i run this:
it fails as this method already has its own objective baked in.
Can i somehow edit an existing method so i can use my own objective based on the model and fitted objects?