Missing parameters in EvolutionaryParameterOptimization

chris_ml

Hi,

for some learners the EvolutionaryParameterOptimization operator does not allow
to evaluate all possible parameters. For example, for BestRuleInduction the parameter
"utility_function" cannot be evaluated. Are there any reasons for that?

Cheers,
Chris

land

Hi Chris,
is is due to the fact, that the evolutionary parameter optimization only supports numerical values.
But you simply could wrap the evolutionary optimization within a grid optimization switching this two values. This will result in an equivalent behavior.

Greetings,
  Sebastian