def __init__(self, config):
self.config = config
self._path = config.data_dir[0]
self.max_utt_len = config.max_utt_len
self.tokenize = get_chat_tokenize()
self.train_corpus = self._read_file(self._path, 'Train')
self.valid_corpus = self._read_file(self._path, 'Valid')
self.test_corpus = self._read_file(self._path, 'Test')
self._extract_vocab()
print("Done loading corpus")
def __init__(self, config):
self.config = config
self._path = config.data_dir[0]
self.max_utt_len = config.max_utt_len
self.tokenize = get_chat_tokenize()
self.train_corpus = self._read_file(os.path.join(self._path, 'kvret_train_public.json'))
self.valid_corpus = self._read_file(os.path.join(self._path, 'kvret_dev_public.json'))
self.test_corpus = self._read_file(os.path.join(self._path, 'kvret_test_public.json'))
self._build_vocab(config.max_vocab_cnt)
print("Done loading corpus")
def __init__(self, config):
self.config = config
self.config = config
self._path = config.data_dir[0]
self.max_utt_len = config.max_utt_len
self.tokenize = get_chat_tokenize()
self.train_corpus = self._read_file(
os.path.join(self._path, 'train_sent_emo.csv'))
# print(self.train_corpus[:3])
self.valid_corpus = self._read_file(
os.path.join(self._path, 'dev_sent_emo.csv'))
self.test_corpus = self._read_file(
os.path.join(self._path, 'test_sent_emo.csv'))
self._build_vocab(config.max_vocab_cnt)
print("Done loading corpus")
def __init__(self, corpus, config, action2name):
super(DialogExchanger, self).__init__()
self.config = config
self.vocab = corpus.vocab
self.vocab_dict = corpus.vocab_dict
self.rev_vocab = corpus.rev_vocab
self.unk_id = corpus.unk_id
self.vocab_size = len(self.vocab)
self.tokenize = get_chat_tokenize()
self.backward_size = config.backward_size
# self.name2action = revert_action_dict(action2name)
# actionencoder returns the corresponsing latent action ids given natural languages
# self.action_encoder = ActionEncoder(corpus, config, self.name2action)
self.max_utt_size = config.max_utt_len
self.unk_word = {'PLACEHOLDE'}
self.gamma = 0.95
def __init__(self, config):
self.config = config
self._path = config.data_dir[0]
self.max_utt_len = config.max_utt_len
self.tokenize = get_chat_tokenize()
self.temp_corpus = self._read_file(self._path)