tatk.policy.rule.crosswoz package¶
Submodules¶
tatk.policy.rule.crosswoz.evaluate module¶
-
tatk.policy.rule.crosswoz.evaluate.
begin_active_tuple_num
(data)¶
-
tatk.policy.rule.crosswoz.evaluate.
begin_da_type
(data)¶
-
tatk.policy.rule.crosswoz.evaluate.
calculateF1
(predict_golden)¶
-
tatk.policy.rule.crosswoz.evaluate.
calculateJointState
(predict_golden)¶
-
tatk.policy.rule.crosswoz.evaluate.
calculateSlotState
(predict_golden)¶
-
tatk.policy.rule.crosswoz.evaluate.
end_usr_da_type
(data)¶
-
tatk.policy.rule.crosswoz.evaluate.
eval_begin_da_predict
(data)¶
-
tatk.policy.rule.crosswoz.evaluate.
eval_simulator_performance
(data, goal_type=None)¶
-
tatk.policy.rule.crosswoz.evaluate.
eval_state_predict
(data)¶
-
tatk.policy.rule.crosswoz.evaluate.
read_zipped_json
(filepath, filename)¶
tatk.policy.rule.crosswoz.rule_simulator module¶
-
class
tatk.policy.rule.crosswoz.rule_simulator.
Simulator
¶ Bases:
tatk.policy.policy.Policy
-
__init__
()¶ Initialize self. See help(type(self)) for accurate signature.
-
begin_da
()¶
-
get_goal
()¶
-
get_reward
()¶
-
infer_goal_type
(goal)¶
-
init_session
(goal=None, state=None, turn_num=0, da_seq=[])¶ Init the class variables for a new session.
-
is_terminated
()¶
-
predict
(sys_act)¶ Predict the next agent action given dialog state. update state[‘system_action’] with predict system action
- Args:
- state (tuple or dict):
when the DST and Policy module are separated, the type of state is tuple. else when they are aggregated together, the type of state is dict (dialog act).
- Returns:
- action (list of list):
The next dialog action.
-
state_predict
()¶
-
state_update
(prev_user_da, prev_sys_da)¶
-