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)