def init_session(self):
self.nlu.init_session()
self.policy.init_session()
self.nlg.init_session()
self.history = []
self.state = default_state()
pass
def main():
s = default_state()
s['history'] = [
['sys', ''],
[
'user',
'Actually I need a expensively priced restaurant . Are there any fitting that description ?'
],
['sys', 'yes , there are 57 in the centre. do you have a preference?'],
['user', 'Can you give me the address ?']
]
# s['belief_state']['attraction']['semi']['area'] = 'centre'
s['belief_state']['restaurant']['semi']['pricerange'] = 'expensive'
testPolicy = MDRGWordPolicy()
# print(s)
print(testPolicy.predict(s))
s['history'].append(
['sys', 'the address is 106 Regent Street City Centre.'])
s['history'].append([
'user',
'It will be for 4 people . Arriving at the restaurant by 18:15 . I \'ll be needing a table for friday .'
])
s['belief_state']['restaurant']['book']['people'] = '4'
s['belief_state']['restaurant']['book']['time'] = '18:15'
s['belief_state']['restaurant']['book']['day'] = 'friday'
# s = default_state()
# s['history'] = [['null', 'I want a chinese restaurant']]
# # s['belief_state']['attraction']['semi']['area'] = 'centre'
# s['belief_state']['restaurant']['semi']['area'] = 'south'
# testPolicy = MDRGWordPolicy()
# print(s)
print(testPolicy.predict(s))
def fake_state():
user_action = {
'Hotel-Request': [['Name', '?']],
'Train-Inform': [['Day', 'don\'t care']]
}
from convlab2.util.multiwoz.state import default_state
init_belief_state = default_state()['belief_state']
kb_results = [None, None]
kb_results[0] = {
'name': 'xxx_train',
'day': 'tuesday',
'dest': 'cam',
'phone': '123-3333',
'area': 'south'
}
kb_results[1] = {
'name': 'xxx_train',
'day': 'tuesday',
'dest': 'cam',
'phone': '123-3333',
'area': 'north'
}
state = {
'user_action': user_action,
'belief_state': init_belief_state,
'kb_results_dict': kb_results,
'hotel-request': [['phone']]
}
'''
state = {'user_action': dict(),
'belief_state: dict(),
'kb_results_dict': kb_results
}
'''
return state
def _build_data(self, root_dir, processed_dir):
self.data = {}
data_loader = ActPolicyDataloader(
dataset_dataloader=MultiWOZDataloader())
for part in ['train', 'val', 'test']:
self.data[part] = []
raw_data = data_loader.load_data(data_key=part, role='sys')[part]
for belief_state, context_dialog_act, terminated, dialog_act in \
zip(raw_data['belief_state'], raw_data['context_dialog_act'], raw_data['terminated'], raw_data['dialog_act']):
state = default_state()
state['belief_state'] = belief_state
state['user_action'] = context_dialog_act[-1]
state['system_action'] = context_dialog_act[-2] if len(
context_dialog_act) > 1 else {}
state['terminated'] = terminated
action = dialog_act
self.data[part].append([
self.vector.state_vectorize(state),
self.vector.action_vectorize(action)
])
os.makedirs(processed_dir)
for part in ['train', 'val', 'test']:
with open(os.path.join(processed_dir, '{}.pkl'.format(part)),
'wb') as f:
pickle.dump(self.data[part], f)
def main():
s = default_state()
s['history'] = [['null', 'I want a chinese restaurant']]
# s['belief_state']['attraction']['semi']['area'] = 'centre'
s['belief_state']['restaurant']['semi']['area'] = 'south'
s['belief_state']['restaurant']['semi']['food'] = 'mexican'
testPolicy = MDRGWordPolicy()
print(s)
print(testPolicy.predict(s))
def __init__(self):
DST.__init__(self)
self.state = default_state()
path = os.path.dirname(
os.path.dirname(
os.path.dirname(
os.path.dirname(os.path.dirname(
os.path.abspath(__file__))))))
path = os.path.join(path, 'data/multiwoz/value_dict.json')
self.value_dict = json.load(open(path))
def init_session(self):
self.state = default_state()
if not self.param_restored:
if os.path.isfile(os.path.join(DOWNLOAD_DIRECTORY, 'pytorch_model.bin')):
print('loading weights from downloaded model')
self.load_weights(model_path=os.path.join(DOWNLOAD_DIRECTORY, 'pytorch_model.bin'))
elif os.path.isfile(os.path.join(SUMBT_PATH, args.output_dir, 'pytorch_model.bin')):
print('loading weights from trained model')
self.load_weights(model_path=os.path.join(SUMBT_PATH, args.output_dir, 'pytorch_model.bin'))
else:
raise ValueError('no availabel weights found.')
self.param_restored = True
def __init__(self):
super(MultiWozSUMBT, self).__init__()
convert_to_glue_format()
self.belief_tracker = BeliefTracker()
self.batch = None # generated with dataloader
self.current_turn = 0
self.idx2slot = {}
self.idx2value = {} # slot value for each slot, use processor.get_labels()
if DEVICE == 'cuda':
if not torch.cuda.is_available():
raise ValueError('cuda not available')
n_gpu = torch.cuda.device_count()
if n_gpu < N_GPU:
raise ValueError('gpu not enough')
print("device: {} n_gpu: {}, distributed training: {}, 16-bits training: {}".format(DEVICE, n_gpu,
bool(N_GPU > 1), FP16))
# Get Processor
self.processor = Processor()
self.label_list = self.processor.get_labels()
self.num_labels = [len(labels) for labels in self.label_list] # number of slot-values in each slot-type
self.belief_tracker.init_session(self.num_labels)
if N_GPU > 1:
self.belief_tracker = torch.nn.DataParallel(self.belief_tracker)
# tokenizer
vocab_dir = os.path.join(BERT_DIR, 'vocab.txt')
if not os.path.exists(vocab_dir):
raise ValueError("Can't find %s " % vocab_dir)
self.tokenizer = BertTokenizer.from_pretrained(vocab_dir, do_lower_case=DO_LOWER_CASE)
self.num_train_steps = None
self.accumulation = False
logger.info('dataset processed')
if os.path.exists(OUTPUT_DIR) and os.listdir(OUTPUT_DIR):
print("output dir {} not empty".format(OUTPUT_DIR))
else:
os.mkdir(OUTPUT_DIR)
fileHandler = logging.FileHandler(os.path.join(OUTPUT_DIR, "log.txt"))
logger.addHandler(fileHandler)
random.seed(SEED)
np.random.seed(SEED)
torch.manual_seed(SEED)
if N_GPU > 0:
torch.cuda.manual_seed_all(SEED)
self.state = default_state()
def __init__(self, ontology_vectors, ontology, slots, data_dir):
DST.__init__(self)
# data profile
self.data_dir = data_dir
self.validation_url = os.path.join(self.data_dir, 'data/validate.json')
self.word_vectors_url = os.path.join(
self.data_dir, 'word-vectors/paragram_300_sl999.txt')
self.training_url = os.path.join(self.data_dir, 'data/train.json')
self.ontology_url = os.path.join(self.data_dir, 'data/ontology.json')
self.testing_url = os.path.join(self.data_dir, 'data/test.json')
self.model_url = os.path.join(self.data_dir, 'models/model-1')
self.graph_url = os.path.join(self.data_dir, 'graphs/graph-1')
self.results_url = os.path.join(self.data_dir, 'results/log-1.txt')
self.kb_url = os.path.join(self.data_dir, 'data/') # not used
self.train_model_url = os.path.join(self.data_dir,
'train_models/model-1')
self.train_graph_url = os.path.join(self.data_dir,
'train_graph/graph-1')
self.model_variables = model_definition(ontology_vectors,
len(ontology),
slots,
num_hidden=None,
bidir=True,
net_type=None,
test=True,
dev='cpu')
self.state = default_state()
_config = tf.ConfigProto()
_config.gpu_options.allow_growth = True
_config.allow_soft_placement = True
self.sess = tf.Session(config=_config)
self.param_restored = False
self.det_dic = {}
for domain, dic in REF_USR_DA.items():
for key, value in dic.items():
assert '-' not in key
self.det_dic[key.lower()] = key + '-' + domain
self.det_dic[value.lower()] = key + '-' + domain
def parent_dir(path, time=1):
for _ in range(time):
path = os.path.dirname(path)
return path
root_dir = parent_dir(os.path.abspath(__file__), 4)
self.value_dict = json.load(
open(os.path.join(root_dir, 'data/multiwoz/value_dict.json')))
def generate_dict(self):
"""
init the dict for mapping state/action into vector
"""
self.act2vec = dict((a, i) for i, a in enumerate(self.da_voc))
self.vec2act = dict((v, k) for k, v in self.act2vec.items())
self.da_dim = len(self.da_voc)
self.opp2vec = dict((a, i) for i, a in enumerate(self.da_voc_opp))
self.da_opp_dim = len(self.da_voc_opp)
self.belief_state_dim = 0
for domain in self.belief_domains:
for slot, value in default_state()['belief_state'][domain.lower()]['semi'].items():
self.belief_state_dim += 1
self.state_dim = self.da_opp_dim + self.da_dim + self.belief_state_dim + \
len(self.db_domains) + 6 * len(self.db_domains) + 1
def setup_class(cls):
# domain_slots will be used for checking request_state
cls.domain_slots = {}
for key, value in REF_SYS_DA.items():
new_key = key.lower()
new_value = set(value.values())
cls.domain_slots[new_key] = new_value
cls.default_belief_state = default_state()['belief_state']
cls.usr_acts = [
{
"Hotel-Inform": [["Area", "east"], ["Stars", "4"]]
},
{
"Hotel-Inform": [["Parking", "yes"], ["Internet", "yes"]]
},
{},
{
"Hotel-Inform": [["Day", "friday"]],
"Hotel-Request": [["Ref", "?"]]
},
{
"Train-Inform": [["Dest", "bishops stortford"],
["Day", "friday"], ["Depart", "cambridge"]]
},
{
"Train-Inform": [["Arrive", "19:45"]]
},
{
"Train-Request": [["Leave", "?"], ["Time", "?"],
["Ticket", "?"]]
},
{
"Hotel-Inform": [["Stay", "4"], ["Day", "monday"],
["People", "3"]]
},
]
def revert_state(model_output:dict, reversed_vocab:dict):
slotdict = {'price range':'pricerange', 'leave at':'leaveAt', 'arrive by':'arriveBy'}
valuedict = {'do not care': 'dontcare'}
belief_state = default_state()['belief_state']
for d in model_output:
for s in model_output[d]:
for v in model_output[d][s]:
domain = reversed_vocab[d]
if domain not in belief_state:
continue
slot = reversed_vocab[s]
value = reversed_vocab[v]
if slot.startswith('book '):
slot = slot[5:]
table = belief_state[domain]['book']
else:
table = belief_state[domain]['semi']
if slot in slotdict:
slot = slotdict[slot]
if value in valuedict:
value = valuedict[value]
if slot in table:
table[slot] = value
return belief_state
def __init__(
self,
archive_file=DEFAULT_ARCHIVE_FILE,
cuda_device=DEFAULT_CUDA_DEVICE,
model_file="https://convlab.blob.core.windows.net/convlab-2/larl.zip"
):
if not os.path.isfile(archive_file):
if not model_file:
raise Exception("No model for LaRL is specified!")
archive_file = cached_path(model_file)
temp_path = os.path.dirname(os.path.abspath(__file__))
#print(temp_path)
zip_ref = zipfile.ZipFile(archive_file, 'r')
zip_ref.extractall(temp_path)
zip_ref.close()
self.prev_state = default_state()
self.prev_active_domain = None
domain_name = 'object_division'
domain_info = domain.get_domain(domain_name)
self.db = Database()
data_path = os.path.join(os.path.dirname(os.path.realpath(__file__)),
'data')
train_data_path = os.path.join(data_path, 'train_dials.json')
if not os.path.exists(train_data_path):
zipped_file = os.path.join(data_path, 'norm-multi-woz.zip')
archive = zipfile.ZipFile(zipped_file, 'r')
archive.extractall(data_path)
norm_multiwoz_path = data_path
with open(
os.path.join(norm_multiwoz_path,
'input_lang.index2word.json')) as f:
self.input_lang_index2word = json.load(f)
with open(
os.path.join(norm_multiwoz_path,
'input_lang.word2index.json')) as f:
self.input_lang_word2index = json.load(f)
with open(
os.path.join(norm_multiwoz_path,
'output_lang.index2word.json')) as f:
self.output_lang_index2word = json.load(f)
with open(
os.path.join(norm_multiwoz_path,
'output_lang.word2index.json')) as f:
self.output_lang_word2index = json.load(f)
config = Pack(
seed=10,
train_path=train_data_path,
max_vocab_size=1000,
last_n_model=5,
max_utt_len=50,
max_dec_len=50,
backward_size=2,
batch_size=1,
use_gpu=True,
op='adam',
init_lr=0.001,
l2_norm=1e-05,
momentum=0.0,
grad_clip=5.0,
dropout=0.5,
max_epoch=100,
embed_size=100,
num_layers=1,
utt_rnn_cell='gru',
utt_cell_size=300,
bi_utt_cell=True,
enc_use_attn=True,
dec_use_attn=True,
dec_rnn_cell='lstm',
dec_cell_size=300,
dec_attn_mode='cat',
y_size=10,
k_size=20,
beta=0.001,
simple_posterior=True,
contextual_posterior=True,
use_mi=False,
use_pr=True,
use_diversity=False,
#
beam_size=20,
fix_batch=True,
fix_train_batch=False,
avg_type='word',
print_step=300,
ckpt_step=1416,
improve_threshold=0.996,
patient_increase=2.0,
save_model=True,
early_stop=False,
gen_type='greedy',
preview_batch_num=None,
k=domain_info.input_length(),
init_range=0.1,
pretrain_folder='2019-09-20-21-43-06-sl_cat',
forward_only=False)
config.use_gpu = config.use_gpu and torch.cuda.is_available()
self.corpus = corpora_inference.NormMultiWozCorpus(config)
self.model = SysPerfectBD2Cat(self.corpus, config)
self.config = config
if config.use_gpu:
self.model.load_state_dict(
torch.load(os.path.join(temp_path, 'larl_model/best-model')))
self.model.cuda()
else:
self.model.load_state_dict(
torch.load(os.path.join(temp_path, 'larl_model/best-model'),
map_location=lambda storage, loc: storage))
self.model.eval()
self.dic = pickle.load(
open(os.path.join(temp_path, 'larl_model/svdic.pkl'), 'rb'))
def update(self, user_act=None):
"""Update the dialog state."""
if type(user_act) is not str:
raise Exception('Expected user_act to be <class \'str\'> type, but get {}.'.format(type(user_act)))
prev_state = copy.deepcopy(self.state)
if not os.path.exists(os.path.join(self.data_dir, "results")):
os.makedirs(os.path.join(self.data_dir, "results"))
global train_batch_size
model_variables = self.model_variables
(user, sys_res, no_turns, user_uttr_len, sys_uttr_len, labels, domain_labels, domain_accuracy,
slot_accuracy, value_accuracy, value_f1, train_step, keep_prob, predictions,
true_predictions, [y, _]) = model_variables
# Note: Comment the following line since the first node is already i
# prev_state['history'] = [['sys', 'null']] if len(prev_state['history']) == 0 else prev_state['history']
assert len(prev_state['history']) > 0
first_turn = prev_state['history'][0]
if first_turn[0] != 'sys':
prev_state['history'] = [['sys', '']] + prev_state['history']
actual_history = []
assert len(prev_state['history']) % 2 == 0
for name, utt in prev_state['history']:
if not utt:
utt = 'null'
if len(actual_history)==0 or len(actual_history[-1])==2:
actual_history.append([utt])
else:
actual_history[-1].append(utt)
# actual_history[-1].append(user_act)
# actual_history = self.normalize_history(actual_history)
# if len(actual_history) == 0:
# actual_history = [['', user_act if len(user_act)>0 else 'fake user act']]
fake_dialogue = {}
turn_no = 0
for _sys, _user in actual_history:
turn = {}
turn['system'] = _sys
fake_user = {}
fake_user['text'] = _user
fake_user['belief_state'] = default_state()['belief_state']
turn['user'] = fake_user
key = str(turn_no)
fake_dialogue[key] = turn
turn_no += 1
context, actual_context = process_history([fake_dialogue], self.word_vectors, self.ontology)
batch_user, batch_sys, batch_labels, batch_domain_labels, batch_user_uttr_len, batch_sys_uttr_len, \
batch_no_turns = generate_batch(context, 0, 1, len(self.ontology)) # old feature
# run model
[pred, y_pred] = self.sess.run(
[predictions, y],
feed_dict={user: batch_user, sys_res: batch_sys,
labels: batch_labels,
domain_labels: batch_domain_labels,
user_uttr_len: batch_user_uttr_len,
sys_uttr_len: batch_sys_uttr_len,
no_turns: batch_no_turns,
keep_prob: 1.0})
# convert to str output
dialgs, _, _ = track_dialogue(actual_context, self.ontology, pred, y_pred)
assert len(dialgs) >= 1
last_turn = dialgs[0][-1]
predictions = last_turn['prediction']
new_belief_state = copy.deepcopy(prev_state['belief_state'])
# update belief state
for item in predictions:
item = item.lower()
domain, slot, value = item.strip().split('-')
value = value[::-1].split(':', 1)[1][::-1]
if slot == 'price range':
slot = 'pricerange'
if slot not in ['name', 'book']:
if domain not in new_belief_state:
raise Exception('Error: domain <{}> not in belief state'.format(domain))
slot = REF_SYS_DA[domain.capitalize( )].get(slot, slot)
assert 'semi' in new_belief_state[domain]
assert 'book' in new_belief_state[domain]
if 'book' in slot:
assert slot.startswith('book ')
slot = slot.strip().split()[1]
if slot == 'arriveby':
slot = 'arriveBy'
elif slot == 'leaveat':
slot = 'leaveAt'
domain_dic = new_belief_state[domain]
if slot in domain_dic['semi']:
new_belief_state[domain]['semi'][slot] = normalize_value(self.value_dict, domain, slot, value)
elif slot in domain_dic['book']:
new_belief_state[domain]['book'][slot] = value
elif slot.lower() in domain_dic['book']:
new_belief_state[domain]['book'][slot.lower()] = value
else:
with open('mdbt_unknown_slot.log', 'a+') as f:
f.write('unknown slot name <{}> with value <{}> of domain <{}>\nitem: {}\n\n'.format(slot, value,
domain, item))
new_request_state = copy.deepcopy(prev_state['request_state'])
# update request_state
user_request_slot = self.detect_requestable_slots(user_act)
for domain in user_request_slot:
for key in user_request_slot[domain]:
if domain not in new_request_state:
new_request_state[domain] = {}
if key not in new_request_state[domain]:
new_request_state[domain][key] = user_request_slot[domain][key]
# update state
new_state = copy.deepcopy(dict(prev_state))
new_state['belief_state'] = new_belief_state
new_state['request_state'] = new_request_state
self.state = new_state
return self.state
def init_session(self):
self.state = default_state()
if not self.param_restored:
self.restore()
def init_session(self):
"""Initialize ``self.state`` with a default state, which ``convlab2.util.multiwoz.state.default_state`` returns."""
self.state = default_state()
def __init__(self, data_dir=DATA_PATH, model_file='https://convlab.blob.core.windows.net/convlab-2/sumbt.tar.gz', eval_slots=multiwoz_slot_list):
DST.__init__(self)
# if not os.path.exists(data_dir):
# if model_file == '':
# raise Exception(
# 'Please provide remote model file path in config')
# resp = urllib.request.urlretrieve(model_file)[0]
# temp_file = tarfile.open(resp)
# temp_file.extractall('data')
# assert os.path.exists(data_dir)
processor = Processor(args)
self.processor = processor
label_list = processor.get_labels()
num_labels = [len(labels) for labels in label_list] # number of slot-values in each slot-type
# tokenizer
self.tokenizer = BertTokenizer.from_pretrained(args.bert_model_name, cache_dir=args.bert_model_cache_dir)
random.seed(args.seed)
np.random.seed(args.seed)
torch.manual_seed(args.seed)
self.device = torch.device("cuda" if USE_CUDA else "cpu")
self.sumbt_model = BeliefTracker(args, num_labels, self.device)
if USE_CUDA and N_GPU > 1:
self.sumbt_model = torch.nn.DataParallel(self.sumbt_model)
if args.fp16:
self.sumbt_model.half()
self.sumbt_model.to(self.device)
## Get slot-value embeddings
self.label_token_ids, self.label_len = [], []
for labels in label_list:
token_ids, lens = get_label_embedding(labels, args.max_label_length, self.tokenizer, self.device)
self.label_token_ids.append(token_ids)
self.label_len.append(lens)
self.label_map = [{label: i for i, label in enumerate(labels)} for labels in label_list]
self.label_map_inv = [{i: label for i, label in enumerate(labels)} for labels in label_list]
self.label_list = label_list
self.target_slot = processor.target_slot
## Get domain-slot-type embeddings
self.slot_token_ids, self.slot_len = \
get_label_embedding(processor.target_slot, args.max_label_length, self.tokenizer, self.device)
self.args = args
self.state = default_state()
self.param_restored = False
if USE_CUDA and N_GPU == 1:
self.sumbt_model.initialize_slot_value_lookup(self.label_token_ids, self.slot_token_ids)
elif USE_CUDA and N_GPU > 1:
self.sumbt_model.module.initialize_slot_value_lookup(self.label_token_ids, self.slot_token_ids)
self.det_dic = {}
for domain, dic in REF_USR_DA.items():
for key, value in dic.items():
assert '-' not in key
self.det_dic[key.lower()] = key + '-' + domain
self.det_dic[value.lower()] = key + '-' + domain
self.cached_res = {}
convert_to_glue_format(DATA_PATH, SUMBT_PATH)
if not os.path.isdir(os.path.join(SUMBT_PATH, args.output_dir)):
os.makedirs(os.path.join(SUMBT_PATH, args.output_dir))
self.train_examples = processor.get_train_examples(os.path.join(SUMBT_PATH, args.tmp_data_dir), accumulation=False)
self.dev_examples = processor.get_dev_examples(os.path.join(SUMBT_PATH, args.tmp_data_dir), accumulation=False)
self.test_examples = processor.get_test_examples(os.path.join(SUMBT_PATH, args.tmp_data_dir), accumulation=False)
self.eval_slots = eval_slots
self.download_model()
def __init__(self, num=1):
parser = argparse.ArgumentParser(description='S2S')
parser.add_argument('--no_cuda',
type=util.str2bool,
nargs='?',
const=True,
default=True,
help='enables CUDA training')
parser.add_argument('--seed',
type=int,
default=1,
metavar='S',
help='random seed (default: 1)')
parser.add_argument('--no_models',
type=int,
default=20,
help='how many models to evaluate')
parser.add_argument('--original',
type=str,
default='model/model/',
help='Original path.')
parser.add_argument('--dropout', type=float, default=0.0)
parser.add_argument('--use_emb', type=str, default='False')
parser.add_argument('--beam_width',
type=int,
default=10,
help='Beam width used in beamsearch')
parser.add_argument('--write_n_best',
type=util.str2bool,
nargs='?',
const=True,
default=False,
help='Write n-best list (n=beam_width)')
parser.add_argument('--model_path',
type=str,
default='model/model/translate.ckpt',
help='Path to a specific model checkpoint.')
parser.add_argument('--model_dir',
type=str,
default='data/multi-woz/model/model/')
parser.add_argument('--model_name', type=str, default='translate.ckpt')
parser.add_argument('--valid_output',
type=str,
default='model/data/val_dials/',
help='Validation Decoding output dir path')
parser.add_argument('--decode_output',
type=str,
default='model/data/test_dials/',
help='Decoding output dir path')
args = parser.parse_args([])
args.cuda = not args.no_cuda and torch.cuda.is_available()
torch.manual_seed(args.seed)
self.device = torch.device("cuda" if args.cuda else "cpu")
with open(
os.path.join(os.path.dirname(__file__),
args.model_path + '.config'), 'r') as f:
add_args = json.load(f)
# print(add_args)
for k, v in add_args.items():
setattr(args, k, v)
# print(args)
args.mode = 'test'
args.load_param = True
args.dropout = 0.0
assert args.dropout == 0.0
# Start going through models
args.original = args.model_path
args.model_path = args.original
self.model = loadModel(num, args)
self.dial = {"cur": {"log": []}}
self.prev_state = default_state()
self.prev_active_domain = None
def main():
dialogue = {
"log": [{
"text":
"am looking for a place to to stay that has cheap price range it should be in a type of hotel",
"metadata": {}
}, {
"text": "Okay, do you have a specific area you want to stay in?",
"metadata": {
"taxi": {
"book": {
"booked": []
},
"semi": {
"leaveAt": "",
"destination": "",
"departure": "",
"arriveBy": ""
}
},
"police": {
"book": {
"booked": []
},
"semi": {}
},
"restaurant": {
"book": {
"booked": [],
"time": "",
"day": "",
"people": ""
},
"semi": {
"food": "",
"pricerange": "",
"name": "",
"area": ""
}
},
"hospital": {
"book": {
"booked": []
},
"semi": {
"department": ""
}
},
"hotel": {
"book": {
"booked": [],
"stay": "",
"day": "",
"people": ""
},
"semi": {
"name": "not mentioned",
"area": "not mentioned",
"parking": "not mentioned",
"pricerange": "cheap",
"stars": "not mentioned",
"internet": "not mentioned",
"type": "hotel"
}
},
"attraction": {
"book": {
"booked": []
},
"semi": {
"type": "",
"name": "",
"area": ""
}
},
"train": {
"book": {
"booked": [],
"people": ""
},
"semi": {
"leaveAt": "",
"destination": "",
"day": "",
"arriveBy": "",
"departure": ""
}
}
}
}, {
"text":
"no, i just need to make sure it's cheap. oh, and i need parking",
"metadata": {}
}, {
"text":
"I found 1 cheap hotel for you that includes parking. Do you like me to book it?",
"metadata": {
"taxi": {
"book": {
"booked": []
},
"semi": {
"leaveAt": "",
"destination": "",
"departure": "",
"arriveBy": ""
}
},
"police": {
"book": {
"booked": []
},
"semi": {}
},
"restaurant": {
"book": {
"booked": [],
"time": "",
"day": "",
"people": ""
},
"semi": {
"food": "",
"pricerange": "",
"name": "",
"area": ""
}
},
"hospital": {
"book": {
"booked": []
},
"semi": {
"department": ""
}
},
"hotel": {
"book": {
"booked": [],
"stay": "",
"day": "",
"people": ""
},
"semi": {
"name": "not mentioned",
"area": "not mentioned",
"parking": "yes",
"pricerange": "cheap",
"stars": "not mentioned",
"internet": "not mentioned",
"type": "hotel"
}
},
"attraction": {
"book": {
"booked": []
},
"semi": {
"type": "",
"name": "",
"area": ""
}
},
"train": {
"book": {
"booked": [],
"people": ""
},
"semi": {
"leaveAt": "",
"destination": "",
"day": "",
"arriveBy": "",
"departure": ""
}
}
}
}, {
"text": "Yes, please. 6 people 3 nights starting on tuesday.",
"metadata": {}
}, {
"text":
"I am sorry but I wasn't able to book that for you for Tuesday. Is there another day you would like to stay or perhaps a shorter stay?",
"metadata": {
"taxi": {
"book": {
"booked": []
},
"semi": {
"leaveAt": "",
"destination": "",
"departure": "",
"arriveBy": ""
}
},
"police": {
"book": {
"booked": []
},
"semi": {}
},
"restaurant": {
"book": {
"booked": [],
"time": "",
"day": "",
"people": ""
},
"semi": {
"food": "",
"pricerange": "",
"name": "",
"area": ""
}
},
"hospital": {
"book": {
"booked": []
},
"semi": {
"department": ""
}
},
"hotel": {
"book": {
"booked": [],
"stay": "3",
"day": "tuesday",
"people": "6"
},
"semi": {
"name": "not mentioned",
"area": "not mentioned",
"parking": "yes",
"pricerange": "cheap",
"stars": "not mentioned",
"internet": "not mentioned",
"type": "hotel"
}
},
"attraction": {
"book": {
"booked": []
},
"semi": {
"type": "",
"name": "",
"area": ""
}
},
"train": {
"book": {
"booked": [],
"people": ""
},
"semi": {
"leaveAt": "",
"destination": "",
"day": "",
"arriveBy": "",
"departure": ""
}
}
}
}, {
"text": "how about only 2 nights.",
"metadata": {}
}, {
"text":
"Booking was successful.\nReference number is : 7GAWK763. Anything else I can do for you?",
"metadata": {
"taxi": {
"book": {
"booked": []
},
"semi": {
"leaveAt": "",
"destination": "",
"departure": "",
"arriveBy": ""
}
},
"police": {
"book": {
"booked": []
},
"semi": {}
},
"restaurant": {
"book": {
"booked": [],
"time": "",
"day": "",
"people": ""
},
"semi": {
"food": "",
"pricerange": "",
"name": "",
"area": ""
}
},
"hospital": {
"book": {
"booked": []
},
"semi": {
"department": ""
}
},
"hotel": {
"book": {
"booked": [{
"name": "the cambridge belfry",
"reference": "7GAWK763"
}],
"stay":
"2",
"day":
"tuesday",
"people":
"6"
},
"semi": {
"name": "not mentioned",
"area": "not mentioned",
"parking": "yes",
"pricerange": "cheap",
"stars": "not mentioned",
"internet": "not mentioned",
"type": "hotel"
}
},
"attraction": {
"book": {
"booked": []
},
"semi": {
"type": "",
"name": "",
"area": ""
}
},
"train": {
"book": {
"booked": [],
"people": ""
},
"semi": {
"leaveAt": "",
"destination": "",
"day": "",
"arriveBy": "",
"departure": ""
}
}
}
}, {
"text": "No, that will be all. Good bye.",
"metadata": {}
}, {
"text": "Thank you for using our services.",
"metadata": {
"taxi": {
"book": {
"booked": []
},
"semi": {
"leaveAt": "",
"destination": "",
"departure": "",
"arriveBy": ""
}
},
"police": {
"book": {
"booked": []
},
"semi": {}
},
"restaurant": {
"book": {
"booked": [],
"time": "",
"day": "",
"people": ""
},
"semi": {
"food": "",
"pricerange": "",
"name": "",
"area": ""
}
},
"hospital": {
"book": {
"booked": []
},
"semi": {
"department": ""
}
},
"hotel": {
"book": {
"booked": [{
"name": "the cambridge belfry",
"reference": "7GAWK763"
}],
"stay":
"2",
"day":
"tuesday",
"people":
"6"
},
"semi": {
"name": "not mentioned",
"area": "not mentioned",
"parking": "yes",
"pricerange": "cheap",
"stars": "not mentioned",
"internet": "not mentioned",
"type": "hotel"
}
},
"attraction": {
"book": {
"booked": []
},
"semi": {
"type": "",
"name": "",
"area": ""
}
},
"train": {
"book": {
"booked": [],
"people": ""
},
"semi": {
"leaveAt": "",
"destination": "",
"day": "",
"arriveBy": "",
"departure": ""
}
}
}
}]
}
dialogue2 = {
"log": [{
"text":
"am looking for a place to to stay that has cheap price range it should be in a type of hotel"
}, {
"text": "Okay, do you have a specific area you want to stay in?",
"metadata": default_state()['belief_state']
}, {
"text":
"no, i just need to make sure it's cheap. oh, and i need parking"
}, {
"text":
"I found 1 cheap hotel for you that includes parking. Do you like me to book it?",
"metadata": default_state()['belief_state']
}, {
"text": "Yes, please. 6 people 3 nights starting on tuesday."
}, {
"text":
"I am sorry but I wasn't able to book that for you for Tuesday. Is there another day you would like to stay or perhaps a shorter stay?",
"metadata": default_state()['belief_state']
}, {
"text": "how about only 2 nights."
}, {
"text":
"Booking was successful.\nReference number is : 7GAWK763. Anything else I can do for you?",
"metadata": default_state()['belief_state']
}, {
"text": "No, that will be all. Good bye."
}, {
"text": "Thank you for using our services.",
"metadata": default_state()['belief_state']
}]
}
create_data(dialogue2)
print(
'Create WOZ-like dialogues. Get yourself a coffee, this might take a while.'
)
delex_data = createData()
print('Divide dialogues...')
divideData(delex_data)
def init_session(self):
self.dial = {"cur": {"log": []}}
self.prev_state = default_state()
self.prev_active_domain = None
def reset():
self.prev_state = default_state()
def train(model, reader, optimizer, writer, config, local_rank, evaluator):
iterator = reader.make_batch(reader.train)
inform_rate = 0
success_rate = 0
inform_test = 0
success_test = 0
dialog_count = 0
if local_rank == 0: # only one process prints something
t = tqdm(enumerate(iterator), total=train.max_iter, ncols=250)
else:
t = enumerate(iterator)
for batch_idx, batch in t:
# try:
inputs, contexts, context_lengths, dial_ids = reader.make_input(batch)
batch_size = len(contexts[0])
turns = len(inputs)
gate_loss = 0
value_loss = 0
action_loss = 0
response_loss = 0
slot_acc = 0
joint_acc = 0
perplexity = 0
batch_count = 0 # number of batches
beliefs = [] # [batch, slots, len] * turn => list
belief_gens = [] # [batch, slots, len] * turn => list
response_gens = [] # [batch, len] * turn => list
distributed_batch_size = math.ceil(batch_size / config.num_gpus)
action_history = [[] for i in range(distributed_batch_size)]
model.module.states = [
default_state() for i in range(distributed_batch_size)
] # user_action, system_action, belief_state, request_state, terminated, history
for b_idx in range(distributed_batch_size):
model.module.states[b_idx]["history"].append(["sys", "null"])
for turn_idx in range(turns):
context_len = contexts[turn_idx].size(1)
# distribute batches to each gpu
for key, value in inputs[turn_idx].items():
if key != "prev_gate":
inputs[turn_idx][key] = distribute_data(
value, config.num_gpus)[local_rank]
contexts[turn_idx] = distribute_data(contexts[turn_idx],
config.num_gpus)[local_rank]
context_lengths[turn_idx] = distribute_data(
context_lengths[turn_idx], config.num_gpus)[local_rank]
# if turn_idx > 0:
# inputs[turn_idx]["prev_gate"] = prev_gate
contexts[turn_idx] = contexts[
turn_idx][:, :context_lengths[turn_idx].max()]
first_turn = (turn_idx == 0)
# teacher_forcing = 1 if np.random.rand() >= 0.5 else 0
teacher_forcing = 0
optimizer.zero_grad()
gate_loss_, value_loss_, acc_belief, belief_gen, action_history_ = model.forward(inputs[turn_idx], contexts[turn_idx], context_lengths[turn_idx], \
action_history, teacher_forcing)
loss = gate_loss_ + value_loss_
gate_loss += gate_loss_ * distributed_batch_size
value_loss += value_loss_ * distributed_batch_size
slot_acc += acc_belief.sum(dim=1).sum(dim=0)
joint_acc += (acc_belief.mean(dim=1) == 1).sum(dim=0).float()
batch_count += distributed_batch_size
action_history = action_history_
# gate_loss_, value_loss_, action_loss_, response_loss_, acc_belief, ppl, prev_gate, belief_gen, response_gen, _, _, _ = \
# model.forward(inputs[turn_idx], contexts[turn_idx], context_lengths[turn_idx], action_history, teacher_forcing)
# # [1], [1], [1], [1], [batch, slots], [1], [batch, slots]
# loss = gate_loss_ + value_loss_ + action_loss_ + response_loss_
# gate_loss += gate_loss_ * distributed_batch_size
# value_loss += value_loss_ * distributed_batch_size
# action_loss += action_loss_ * distributed_batch_size
# response_loss += response_loss_ * distributed_batch_size
# slot_acc += acc_belief.sum(dim=1).sum(dim=0)
# joint_acc += (acc_belief.mean(dim=1) == 1).sum(dim=0)
# perplexity += ppl * distributed_batch_size
# batch_count += distributed_batch_size
loss.backward()
torch.nn.utils.clip_grad_norm_(model.parameters(), 10)
optimizer.step()
torch.cuda.empty_cache()
# belief = [] # [batch, slots, len] => list
# for belief_batch in inputs[turn_idx]["belief"]:
# belief_ = []
# for belief_slot in belief_batch:
# belief_.append(reader.vocab.decode(belief_slot[belief_slot != 0].tolist()[:-1])) # remove <eos>
# belief.append(belief_)
# beliefs.append(belief)
# belief_gens.append(belief_gen)
# response_gens.append([reader.vocab.decode(response) for response in response_gen])
# mean of all gpus
# dist.all_reduce(gate_loss)
# gate_loss = gate_loss / config.num_gpus
# dist.all_reduce(value_loss)
# value_loss = value_loss / config.num_gpus
# dist.all_reduce(action_loss)
# action_loss = action_loss / config.num_gpus
# dist.all_reduce(response_loss)
# response_loss = response_loss / config.num_gpus
# dist.all_reduce(slot_acc)
# slot_acc = slot_acc / config.num_gpus
# dist.all_reduce(joint_acc)
# joint_acc = joint_acc / config.num_gpus
# dist.all_reduce(perplexity)
# perplexity = perplexity / config.num_gpus
gate_loss = gate_loss.item() / batch_count
value_loss = value_loss.item() / batch_count
# action_loss = action_loss.item() / batch_count
# response_loss = response_loss.item() / batch_count
slot_acc = slot_acc.item() / batch_count / len(
ontology.all_info_slots) * 100
joint_acc = joint_acc.item() / batch_count * 100
# perplexity = perplexity.item() / batch_count
total_loss = gate_loss + value_loss #+ action_loss + response_loss
train.global_step += 1
# dial_ids = distribute_data(dial_ids, config.num_gpus)[local_rank]
# inform_rate_, success_rate_, inform_test_, success_test_ = evaluator.evaluate(beliefs, belief_gens, response_gens, dial_ids)
# inform_rate += inform_rate_
# success_rate += success_rate_
# inform_test += inform_test_
# success_test += success_test_
# dialog_count += distributed_batch_size
if local_rank == 0:
writer.add_scalar("Train/loss", total_loss, train.global_step)
t.set_description("iter: {}, total loss: {:.4f}, gate loss: {:.4f}, value loss: {:.4f}, joint accuracy: {:.4f}, slot accuracy: {:.4f}"\
.format(batch_idx+1, total_loss, gate_loss, value_loss, joint_acc, slot_acc))
def predict_response(self, state):
history = []
for i in range(len(state['history'])):
history.append(state['history'][i][1])
e_idx = len(history)
s_idx = max(0, e_idx - self.config.backward_size)
context = []
for turn in history[s_idx:e_idx]:
# turn = pad_to(config.max_utt_len, turn, do_pad=False)
context.append(turn)
if len(state['history']) == 1:
self.prev_state = default_state()
prepared_data = {}
prepared_data['context'] = []
prepared_data['response'] = {}
prev_action = deepcopy(self.prev_state['user_action'])
prev_bstate = deepcopy(self.prev_state['belief_state'])
state_history = state['history']
action = deepcopy(state['user_action'])
bstate = deepcopy(state['belief_state'])
# mark_not_mentioned(prev_state)
#active_domain = self.get_active_domain_convlab(self.prev_active_domain, prev_bstate, bstate)
active_domain = self.get_active_domain(self.prev_active_domain,
prev_bstate, bstate)
#print(active_domain)
domain_mark_not_mentioned(bstate, active_domain)
top_results, num_results = None, None
for usr in context:
words = usr.split()
usr = delexicalize.delexicalise(' '.join(words), self.dic)
# parsing reference number GIVEN belief state
usr = delexicaliseReferenceNumber(usr, bstate)
# changes to numbers only here
digitpat = re.compile('\d+')
usr = re.sub(digitpat, '[value_count]', usr)
# add database pointer
pointer_vector, top_results, num_results = addDBPointer(
bstate, self.db)
#print(top_results)
# add booking pointer
pointer_vector = addBookingPointer(bstate, pointer_vector)
belief_summary = get_summary_bstate(bstate)
usr_utt = [BOS] + usr.split() + [EOS]
packed_val = {}
packed_val['bs'] = belief_summary
packed_val['db'] = pointer_vector
packed_val['utt'] = self.corpus._sent2id(usr_utt)
prepared_data['context'].append(packed_val)
prepared_data['response']['bs'] = prepared_data['context'][-1]['bs']
prepared_data['response']['db'] = prepared_data['context'][-1]['db']
results = [
Pack(context=prepared_data['context'],
response=prepared_data['response'])
]
data_feed = prepare_batch_gen(results, self.config)
outputs = self.model_predict(data_feed)
if active_domain is not None and active_domain in num_results:
num_results = num_results[active_domain]
else:
num_results = 0
if active_domain is not None and active_domain in top_results:
top_results = {active_domain: top_results[active_domain]}
else:
top_results = {}
state_with_history = deepcopy(bstate)
state_with_history['history'] = deepcopy(state_history)
response = self.populate_template(outputs, top_results, num_results,
state_with_history)
return response, active_domain
def validate(model, reader, evaluator, config, local_rank):
model.eval()
val_loss = 0
slot_acc = 0
joint_acc = 0
perplexity = 0
batch_count = 0
inform_rate = 0
success_rate = 0
inform_test = 0
success_test = 0
dialog_count = 0
with torch.no_grad():
iterator = reader.make_batch(reader.dev)
if local_rank == 0:
t = tqdm(enumerate(iterator), total=validate.max_iter, ncols=150)
else:
t = enumerate(iterator)
for batch_idx, batch in t:
inputs, contexts, context_lengths, dial_ids = reader.make_input(
batch)
batch_size = len(contexts[0])
beliefs = [] # [batch, slots, len] * turn => list
belief_gens = [] # [batch, slots, len] * turn => list
response_gens = [] # [batch, len] * turn => list
action_gens = []
restores = []
turn_domains = []
turns = len(inputs)
distributed_batch_size = math.ceil(batch_size / config.num_gpus)
action_history = [[] for i in range(distributed_batch_size)]
model.module.states = [
default_state() for i in range(distributed_batch_size)
]
for b_idx in range(distributed_batch_size):
model.module.states[b_idx]["history"].append(["sys", "null"])
for turn_idx in range(turns):
for key, value in inputs[turn_idx].items():
if key != "prev_gate":
inputs[turn_idx][key] = distribute_data(
value, config.num_gpus)[local_rank]
contexts[turn_idx] = distribute_data(
contexts[turn_idx], config.num_gpus)[local_rank]
context_lengths[turn_idx] = distribute_data(
context_lengths[turn_idx], config.num_gpus)[local_rank]
# if turn_idx > 0:
# inputs[turn_idx]["prev_gate"] = prev_gate
contexts[turn_idx] = contexts[
turn_idx][:, :context_lengths[turn_idx].max()]
first_turn = (turn_idx == 0)
teacher_forcing = 0
gate_loss_, value_loss_, acc_belief, belief_gen, action_history_ = model.forward(inputs[turn_idx], contexts[turn_idx], context_lengths[turn_idx], \
action_history, teacher_forcing)
# gate_loss_, value_loss_, action_loss_, response_loss_, acc_belief, ppl, prev_gate, belief_gen, response_gen, action_gen, restored, turn_domain = \
# model.forward(inputs[turn_idx], contexts[turn_idx], context_lengths[turn_idx], teacher_forcing)
loss = gate_loss_ + value_loss_ #+ action_loss_ + response_loss_
val_loss += loss * distributed_batch_size
slot_acc += acc_belief.sum(dim=1).sum(dim=0)
joint_acc += (acc_belief.mean(dim=1) == 1).sum(dim=0).float()
# perplexity += ppl * distributed_batch_size
batch_count += distributed_batch_size
action_history = action_history_
torch.cuda.empty_cache()
# belief = [] # [batch, slots, len] => list
# for belief_batch in inputs[turn_idx]["belief"]:
# belief_ = []
# for belief_slot in belief_batch:
# belief_.append(reader.vocab.decode(belief_slot[belief_slot != 0].tolist()[:-1])) # remove <eos>
# belief.append(belief_)
# beliefs.append(belief)
# belief_gens.append(belief_gen)
# response_gens.append([reader.vocab.decode(response) for response in response_gen])
# action_gens.append([reader.vocab.decode(action) for action in action_gen])
# restores.append(restored)
# turn_domains.append([ontology.all_domains[domain] for domain in turn_domain])
if local_rank == 0:
t.set_description("iter: {}".format(batch_idx + 1))
# dial_ids = distribute_data(dial_ids, config.num_gpus)[local_rank]
# inform_rate_, success_rate_, inform_test_, success_test_ = evaluator.evaluate(beliefs, belief_gens, response_gens, dial_ids)
# inform_rate += inform_rate_
# success_rate += success_rate_
# inform_test += inform_test_
# success_test += success_test_
# dialog_count += distributed_batch_size
# if turns > 1:
# for i in range(min(5,len(response_gens[1]))):
# print(local_rank, reader.vocab.decode(inputs[1]["user"][i][inputs[1]["user"][i]!=0].tolist()), "@@", action_gens[1][i], "@@", response_gens[1][i], "@@", \
# restores[1][i], "@@", turn_domains[1][i])
dist.all_reduce(val_loss)
val_loss = val_loss / config.num_gpus
dist.all_reduce(slot_acc)
slot_acc = slot_acc / config.num_gpus
dist.all_reduce(joint_acc)
joint_acc = joint_acc / config.num_gpus
# dist.all_reduce(perplexity)
# perplexity = perplexity / config.num_gpus
# dist.all_reduce(inform_rate)
# inform_rate = inform_rate / config.num_gpus
# dist.all_reduce(success_rate)
# success_rate = success_rate / config.num_gpus
# dist.all_reduce(inform_test)
# inform_test = inform_test / config.num_gpus
# dist.all_reduce(success_test)
# success_test = success_test / config.num_gpus
model.train()
val_loss = val_loss.item() / batch_count
slot_acc = slot_acc.item() / batch_count / len(
ontology.all_info_slots) * 100
joint_acc = joint_acc.item() / batch_count * 100
# perplexity = perplexity.item() / batch_count
# inform_rate = inform_rate / dialog_count * 100
# success_rate = success_rate / dialog_count * 100
# inform_test = inform_test / dialog_count * 100
# success_test = success_test / dialog_count * 100
return val_loss, joint_acc, slot_acc #, perplexity, inform_rate, success_rate, inform_test, success_test
from convlab2.nlg.template.multiwoz.nlg import TemplateNLG
user_nlg = TemplateNLG(is_user=True, mode='manual')
sys_nlg = TemplateNLG(is_user=False, mode='manual')
from convlab2.util.multiwoz.state import default_state
user_policy.init_session()
sys_policy.init_session()
print(user_policy.goal)
print(user_policy.agenda)
user_act = user_policy.predict([])
print(user_act)
user_utt = user_nlg.generate(user_act)
print(user_utt)
state = default_state()
state['user_action'] = user_act
sys_act = sys_policy.predict(state)
sys_act.append(["Request", "Restaurant", "Price", "?"])
print(sys_act)
user_act = user_policy.predict(sys_act)
print(user_act)
user_utt = user_nlg.generate(user_act)
print(user_utt)
sys_act = sys_policy.predict(state)
print(sys_act)
user_act = user_policy.predict(sys_act)
print(user_act)
user_utt = user_nlg.generate(user_act)
def init_session(self):
self.state = default_state()