refine_plan.models.option
Class for options, which are temporally extended actions.
Author: Charlie Street Owner: Charlie Street
Module Contents
Classes
An option is a temporally extended behaviour. |
- class refine_plan.models.option.Option(name, transition_list, reward_list)
Bases:
objectAn option is a temporally extended behaviour.
Options are traditionally defined in terms of an initiation set I, a termination condition beta, and a policy pi.
Here, we just define the option in terms of its transition probabilities (assuming no preemption occurs) and reward model.
- _name
The option’s name
- _transition_list
The option’s transition model
- _reward_list
The option’s reward model
- get_name()
Return the option’s name.
- Returns:
The option’s name
- get_transition_prob(state, next_state)
Return the transition probability for a (s,s’) pair.
Assumes there is only one precondition which holds for each state.
- Parameters:
state – The first state
next_state – The next state
- Returns:
The transition probability
- get_reward(state)
Return the reward for executing an option in a given state.
- Parameters:
state – The state we want to check
- Returns:
The reward for the state
- get_transition_prism_string()
Return a PRISM string which captures all transitions for this option.
- Returns:
The transition PRISM string
- get_reward_prism_string()
Return a PRISM string which captures all rewards for this option.
- Returns:
The reward PRISM string