I want to have my target function to be dependent on the amount of true decisions made by the model and instead of applying a certain factor (i.e. * 3) to it, want to assign certain values for differing amount of true decisions.
For example instead of
for week in tf do
(sum booleanDecision.[All,week,All]) * 3.0
to something like this
for week in tf do
match sum booleanDecision.[All,week,All] with
| 1.0<'u> -> 1.0
| 2.0<'u> -> 2.5
...
| _ -> 100.0
Or is there a principality of LP I'm missing that doesn't allow something akin to this
I want to have my target function to be dependent on the amount of true decisions made by the model and instead of applying a certain factor (i.e. * 3) to it, want to assign certain values for differing amount of true decisions.
For example instead of
to something like this
Or is there a principality of LP I'm missing that doesn't allow something akin to this