convlab2.policy.vector package¶
Submodules¶
convlab2.policy.vector.dataset module¶
-
class
convlab2.policy.vector.dataset.ActDataset(s_s, a_s)¶ Bases:
torch.utils.data.dataset.Dataset
-
class
convlab2.policy.vector.dataset.ActStateDataset(s_s, a_s, next_s)¶ Bases:
torch.utils.data.dataset.Dataset
convlab2.policy.vector.vector_camrest module¶
-
class
convlab2.policy.vector.vector_camrest.CamrestVector(voc_file, voc_opp_file, character='sys', intent_file='/home/travis/build/thu-coai/ConvLab-2/data/camrest/trackable_intent.json')¶ Bases:
convlab2.policy.vec.Vector-
action_devectorize(action_vec)¶ recover an action
- Args:
- action_vec (np.array):
Dialog act vector
- Returns:
action (tuple): Dialog act
-
action_vectorize(action)¶
-
generate_dict()¶ init the dict for mapping state/action into vector
-
one_hot_vector(num)¶ Return number of available entities for particular domain.
-
pointer(turn)¶
-
state_vectorize(state)¶ vectorize a state
- Args:
- state (dict):
Dialog state
- action (tuple):
Dialog act
- Returns:
state_vec (np.array): Dialog state vector
-
convlab2.policy.vector.vector_crosswoz module¶
-
class
convlab2.policy.vector.vector_crosswoz.CrossWozVector(sys_da_voc_json, usr_da_voc_json)¶ Bases:
convlab2.policy.vec.Vector-
action_devectorize(action_vec)¶ must call state_vectorize func before :param action_vec: :return:
-
action_vectorize(da)¶
-
generate_dict()¶ init the dict for mapping state/action into vector
-
state_vectorize(state)¶ vectorize a state
- Args:
- state (tuple):
Dialog state
- Returns:
- state_vec (np.array):
Dialog state vector
-
convlab2.policy.vector.vector_multiwoz module¶
-
class
convlab2.policy.vector.vector_multiwoz.MultiWozVector(voc_file, voc_opp_file, character='sys', intent_file='/home/travis/build/thu-coai/ConvLab-2/data/multiwoz/trackable_intent.json', composite_actions=False, vocab_size=500)¶ Bases:
convlab2.policy.vec.Vector-
action_devectorize(action_vec)¶ recover an action Args:
- action_vec (np.array):
Dialog act vector
- Returns:
- action (tuple):
Dialog act
-
action_vectorize(action)¶
-
dbquery_domain(domain)¶ query entities of specified domain Args:
- domain string:
domain to query
- Returns:
- entities list:
list of entities of the specified domain
-
generate_dict()¶ init the dict for mapping state/action into vector
-
load_composite_actions()¶ load the composite actions to self.da_voc
-
one_hot_vector(num, domain, vector)¶ Return number of available entities for particular domain.
-
pointer(turn)¶
-
state_vectorize(state)¶ vectorize a state
- Args:
- state (dict):
Dialog state
- action (tuple):
Dialog act
- Returns:
- state_vec (np.array):
Dialog state vector
-