default=CONFIG_DEFAULT,
help="Configuration file for the bot, default=" +
CONFIG_DEFAULT)
parser.add_argument("-m", "--model", required=True, help="Model to load")
parser.add_argument("--sample",
default=False,
action='store_true',
help="Enable sampling mode")
prog_args = parser.parse_args()
conf = configparser.ConfigParser()
if not conf.read(os.path.expanduser(prog_args.config)):
log.error("Configuration file %s not found", prog_args.config)
sys.exit()
emb_dict = data.load_emb_dict(os.path.dirname(prog_args.model))
log.info("Loaded embedded dict with %d entries", len(emb_dict))
rev_emb_dict = {idx: word for word, idx in emb_dict.items()}
end_token = emb_dict[data.END_TOKEN]
net = model.PhraseModel(emb_size=model.EMBEDDING_DIM,
dict_size=len(emb_dict),
hid_size=model.HIDDEN_STATE_SIZE)
net.load_state_dict(torch.load(prog_args.model))
def bot_func(bot, update, args):
text = " ".join(args)
words = utils.tokenize(text)
seq_1 = data.encode_words(words, emb_dict)
input_seq = model.pack_input(seq_1, net.emb)
enc = net.encode(input_seq)
def process_string(s, emb_dict, rev_emb_dict, net, use_sampling=False):
out_words = words_to_words(words, emb_dict, rev_emb_dict, net, use_sampling=use_sampling)
print(" ".join(out_words))
if __name__ == "__main__":
logging.basicConfig(format="%(asctime)-15s %(levelname)s %(message)s", level=logging.INFO)
parser = argparse.ArgumentParser()
parser.add_argument("-m", "--model", required=True, help="Model name to load")
parser.add_argument("-s", "--string", help="String to process, otherwise will loop")
parser.add_argument("--sample", default=False, action="store_true", help="Enable sampling generation instead of argmax")
parser.add_argument("--self", type=int, default=1, help="Enable self-loop mode with given amount of phrases.")
args = parser.parse_args()
emb_dict = data.load_emb_dict(os.path.dirname(args.model))
net = model.PhraseModel(emb_size=model.EMBEDDING_DIM, dict_size=len(emb_dict), hid_size=model.HIDDEN_STATE_SIZE)
net.load_state_dict(torch.load(args.model))
rev_emb_dict = {idx: word for word, idx in emb_dict.items()}
while True:
if args.string:
input_string = args.string
else:
input_string = input(">>> ")
if not input_string:
break
words = utils.tokenize(input_string)
for _ in range(args.self):
