convlab2.policy.rule.camrest package¶
Submodules¶
convlab2.policy.rule.camrest.policy_agenda_camrest module¶
-
class
convlab2.policy.rule.camrest.policy_agenda_camrest.
Agenda
(goal: convlab2.policy.rule.camrest.policy_agenda_camrest.Goal)¶ Bases:
object
-
close_session
()¶ Clear up all actions
-
get_action
(initiative=1)¶ get multiple acts based on initiative Args:
initiative (int): number of slots , just for ‘inform’
- Returns:
action (dict): user diaact
-
is_empty
()¶ Is the agenda already empty Returns:
(boolean): True for empty, False for not.
-
update
(sys_action, goal: convlab2.policy.rule.camrest.policy_agenda_camrest.Goal)¶ update Goal by current agent action and current goal. { A’ + G” + sys_action -> A” } Args:
sys_action (tuple): Preorder system action.s goal (Goal): User Goal
-
update_domain
(diaact, slot_vals, goal: convlab2.policy.rule.camrest.policy_agenda_camrest.Goal)¶ Handel Domain-XXX :param diaact: Dial-Act :param slot_vals: slot value pairs :param goal: Goal :return: True:user want to close the session. False:session is continue
-
-
class
convlab2.policy.rule.camrest.policy_agenda_camrest.
Goal
(goal_generator: convlab2.task.camrest.goal_generator.GoalGenerator)¶ Bases:
object
User Goal Model Class.
-
next_domain_incomplete
()¶
-
task_complete
()¶ Check that all requests have been met Returns:
(boolean): True to accomplish.
-
-
class
convlab2.policy.rule.camrest.policy_agenda_camrest.
UserPolicyAgendaCamrest
¶ Bases:
convlab2.policy.policy.Policy
The rule-based user policy model by agenda. Derived from the UserPolicy class
-
get_goal
()¶
-
get_reward
()¶
-
init_session
()¶ Build new Goal and Agenda for next session
-
is_terminated
()¶
-
predict
(state)¶ Predict an user act based on state and preorder system action. Args:
state (tuple): Dialog state.
- Returns:
action (tuple): User act. session_over (boolean): True to terminate session, otherwise session continues. reward (float): Reward given by user.
-
-
convlab2.policy.rule.camrest.policy_agenda_camrest.
test
()¶
-
convlab2.policy.rule.camrest.policy_agenda_camrest.
test_turn
(user_simulator, sys_action)¶
-
convlab2.policy.rule.camrest.policy_agenda_camrest.
test_with_system
()¶
convlab2.policy.rule.camrest.rule module¶
-
class
convlab2.policy.rule.camrest.rule.
Rule
(is_train=False, character='sys')¶ Bases:
convlab2.policy.policy.Policy
-
get_goal
()¶
-
get_reward
()¶
-
init_session
()¶ Restore after one session
-
is_terminated
()¶
-
predict
(state)¶ Predict an system action given state. Args:
state (dict): Dialog state. Please refer to util/state.py
- Returns:
action : System act, with the form of (act_type, {slot_name_1: value_1, slot_name_2, value_2, …})
-
convlab2.policy.rule.camrest.rule_based_camrest_bot module¶
-
class
convlab2.policy.rule.camrest.rule_based_camrest_bot.
RuleBasedCamrestBot
¶ Bases:
convlab2.policy.policy.Policy
Rule-based bot. Implemented for Camrest dataset.
-
choice
= ''¶
-
init_session
()¶ Init the class variables for a new session.
-
predict
(state)¶ - Args:
State, please refer to util/state.py
- Output:
DA(Dialog Act), in the form of {act_type1: [[slot_name_1, value_1], [slot_name_2, value_2], …], …}
-
-
convlab2.policy.rule.camrest.rule_based_camrest_bot.
check_diff
(last_state, state)¶
-
convlab2.policy.rule.camrest.rule_based_camrest_bot.
deduplicate
(lst)¶
-
convlab2.policy.rule.camrest.rule_based_camrest_bot.
fake_state
()¶
-
convlab2.policy.rule.camrest.rule_based_camrest_bot.
generate_ref_num
(length)¶ Generate a ref num for booking.
-
convlab2.policy.rule.camrest.rule_based_camrest_bot.
test_init_state
()¶
-
convlab2.policy.rule.camrest.rule_based_camrest_bot.
test_run
()¶