For a relatively small datasets we are trying to find y=f(x)
We realized the results quality are highly dependent on the number of programs (population size).
The optimal number of programs seems to be a function of number of data points and overall correlation between x and y.
Any hint to define the optimal number of programs?
For a relatively small datasets we are trying to find y=f(x) We realized the results quality are highly dependent on the number of programs (population size).
The optimal number of programs seems to be a function of number of data points and overall correlation between x and y. Any hint to define the optimal number of programs?
Thanks