The learned range test method for the inverse inclusion problem