mlpy.planners.IPlanner.choose_action¶

IPlanner.choose_action(state, use_policy=False)[source]¶

Choose the optimal action for a state according to the current policy.

Parameters:

state : MDPState

The state for which to choose the next action for.

use_policy : bool, optional

When using a policy the next action is chosen according to the current policy, otherwise the best action is selected. Default is False.

Returns:

MDPAction :

The next action.