def __init__(self):
""" Constructor for RuleInformBot class. """
SysPolicy.__init__(self)
self.cur_inform_slot_id = 0
self.cur_request_slot_id = 0
self.domains = ['Taxi']
def __init__(self,
archive_file=DEFAULT_ARCHIVE_FILE,
cuda_device=DEFAULT_CUDA_DEVICE,
model_file=None):
""" Constructor for NLU class. """
SysPolicy.__init__(self)
check_for_gpu(cuda_device)
if not os.path.isfile(archive_file):
if not model_file:
raise Exception("No model for MILU is specified!")
archive_file = cached_path(model_file)
archive = load_archive(archive_file, cuda_device=cuda_device)
dataset_reader_params = archive.config["dataset_reader"]
self.dataset_reader = DatasetReader.from_params(dataset_reader_params)
self.action_vocab = self.dataset_reader.action_vocab
self.state_encoder = self.dataset_reader.state_encoder
self.model = archive.model
self.model.eval()
def __init__(self, archive_file=DEFAULT_ARCHIVE_FILE, model_file=None):
SysPolicy.__init__(self)
if not os.path.isfile(archive_file):
if not model_file:
raise Exception("No model for Sequicity is specified!")
archive_file = cached_path(model_file)
model_dir = os.path.dirname(os.path.abspath(__file__))
if not os.path.exists(os.path.join(model_dir, 'data')):
archive = zipfile.ZipFile(archive_file, 'r')
archive.extractall(model_dir)
cfg.init_handler('tsdf-multiwoz')
torch.manual_seed(cfg.seed)
torch.cuda.manual_seed(cfg.seed)
random.seed(cfg.seed)
np.random.seed(cfg.seed)
self.m = Model('multiwoz')
self.m.count_params()
self.m.load_model()
self.reset()
def __init__(self,
archive_file=DEFAULT_ARCHIVE_FILE,
cuda_device=DEFAULT_CUDA_DEVICE,
model_file=None):
SysPolicy.__init__(self)
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 = tempfile.mkdtemp()
zip_ref = zipfile.ZipFile(archive_file, 'r')
zip_ref.extractall(temp_path)
zip_ref.close()
self.prev_state = init_state()
self.prev_active_domain = None
domain_name = 'object_division'
domain_info = domain.get_domain(domain_name)
data_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), 'data')
train_data_path = os.path.join(data_path, 'norm-multi-woz', '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 = os.path.join(data_path, 'norm-multi-woz')
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 __init__(self,
model='gpt2_v1',
model_checkpoint='./models/v1',
max_history=15,
device='cuda',
no_sample=False,
max_length=40,
min_length=1,
seed=42,
temperature=0.9,
top_k=0,
top_p=0.8):
SysPolicy.__init__(self)
if not os.path.isdir("./models"):
download_model_from_googledrive(
file_id='14bVEODjyoSV6yZbrzEz4NA8kXMuPbLJH',
dest_path='./models/models.zip')
self.model_checkpoint = model_checkpoint
self.max_history = max_history
self.max_length = max_length
self.min_length = min_length
self.temperature = temperature
self.top_k = top_k
self.top_p = top_p
self.no_sample = no_sample
self.device = device
self.seed = seed
self.domains = [
'hotel', 'restaurant', 'train', 'taxi', 'attraction', 'police',
'hospital'
]
self.cs_mapping = {
'restaurant': ['food', 'pricerange', 'name', 'area'],
'hospital': ['department', 'phone'],
'hotel': [
'name', 'area', 'parking', 'pricerange', 'stars', 'internet',
'type'
],
'attraction': ['type', 'name', 'area'],
'train':
['leaveat', 'destination', 'day', 'arriveby', 'departure'],
'taxi': ['leaveat', 'destination', 'departure', 'arriveby'],
'police': []
}
dia_act = open('./data/multiwoz/dialog_act_slot.txt', 'r')
f = dia_act.read().split('\n')
self.dia_act_dict = {}
key = ""
for i, c in enumerate(f):
if i == 0: continue # User Dialog Act case
t = c.split('\t')
if len(t) == 1:
key = t[0].lower()
self.dia_act_dict[key] = []
else:
self.dia_act_dict[key].append(t[-1].strip().lower())
random.seed(self.seed)
torch.random.manual_seed(seed)
torch.cuda.manual_seed(seed)
self.cur_dom = ''
self.prev_dom = ''
self.prev_cs = []
self.model_name = model
tokenizer_class = GPT2Tokenizer
model_class = GPT2DoubleHeadsModel
self.model = model_class.from_pretrained(self.model_checkpoint)
if 'v1' in self.model_name:
self.SPECIAL_TOKENS = SPECIAL_TOKENS_V1
else:
self.SPECIAL_TOKENS = SPECIAL_TOKENS_V4
if "gpt2" in self.model_name:
self.tokenizer = tokenizer_class.from_pretrained(
model_checkpoint, unk_token='<|unkwn|>')
SPECIAL_TOKENS_DICT = {}
for st in self.SPECIAL_TOKENS:
SPECIAL_TOKENS_DICT[st] = st
self.tokenizer.add_special_tokens(SPECIAL_TOKENS_DICT)
self.model.resize_token_embeddings(len(self.tokenizer))
else:
self.tokenizer = tokenizer_class.from_pretrained(model_checkpoint)
self.tokenizer.set_special_tokens(self.SPECIAL_TOKENS)
self.model.set_num_special_tokens(len(self.SPECIAL_TOKENS))
for s in self.SPECIAL_TOKENS:
self.tokenizer.nlp.tokenizer.add_special_case(s, [{ORTH: s}])
self.model.to(self.device)
self.model.eval()
self.count = 0
self.reset()
def __init__(self):
SysPolicy.__init__(self)
self.last_state = {}
