def main():
# Script arguments can include path of the config
arg_parser = argparse.ArgumentParser()
arg_parser.add_argument('--config', type=str, default="chatbot.cfg")
args = arg_parser.parse_args()
# Read the config
config = configparser.ConfigParser(allow_no_value=True)
with open(args.config) as f:
config.read_file(f)
# Download and load main model
target_folder_name = download_model_folder(config)
model, tokenizer = load_model(target_folder_name, config)
# Download and load reverse model
use_mmi = config.getboolean('model', 'use_mmi')
if use_mmi:
mmi_target_folder_name = download_reverse_model_folder(config)
mmi_model, mmi_tokenizer = load_model(mmi_target_folder_name, config)
else:
mmi_model = None
mmi_tokenizer = None
# Run Telegram bot
bot = TelegramBot(model, tokenizer, config, mmi_model=mmi_model, mmi_tokenizer=mmi_tokenizer)
bot.run_chat()
def main():
spymode = False
# Script arguments can include path of the config
arg_parser = argparse.ArgumentParser()
arg_parser.add_argument('--config', type=str, default='chatbot.cfg')
args = arg_parser.parse_args()
# Read the config
config = configparser.ConfigParser(allow_no_value=True)
with open(args.config) as f:
config.read_file(f)
# Download and load main model
target_folder_name = download_model_folder(config)
(model, tokenizer) = load_model(target_folder_name, config)
# Download and load reverse model
use_mmi = config.getboolean('model', 'use_mmi')
if use_mmi:
mmi_target_folder_name = download_reverse_model_folder(config)
(mmi_model, mmi_tokenizer) = load_model(mmi_target_folder_name, config)
else:
mmi_model = None
mmi_tokenizer = None
# Run chatbot with GPT-2
# run_chat(model, tokenizer, config, mmi_model=mmi_model, mmi_tokenizer=mmi_tokenizer)
if (spymode):
chat = SpyeeChat()
chat_loop(chat,
model,
tokenizer,
config,
mmi_model=mmi_model,
mmi_tokenizer=mmi_tokenizer,
spyMode=spymode)
else:
chat = RandomChat()
chat_loop(chat,
model,
tokenizer,
config,
mmi_model=mmi_model,
mmi_tokenizer=mmi_tokenizer,
spyMode=spymode)
def main():
global translator
global num_samples
global max_turns_history
global model
global tokenizer
global mmi_model
global mmi_tokenizer
global config
global number_of_messages
global number_of_sent_messages
global number_of_servers
global history_dict
global token
token = "TOKEN_GOES_HERE" # Replace TOKEN_GOES_HERE with your discord API bot token!
history_dict = {}
# Script arguments can include path of the config
arg_parser = argparse.ArgumentParser()
arg_parser.add_argument('--config', type=str, default="chatbot.cfg")
args = arg_parser.parse_args()
# Read the config
config = configparser.ConfigParser(allow_no_value=True)
with open(args.config) as f:
config.read_file(f)
# Download and load main model
target_folder_name = download_model_folder(config)
model, tokenizer = load_model(target_folder_name, config)
# Download and load reverse model
use_mmi = config.getboolean('model', 'use_mmi')
if use_mmi:
mmi_target_folder_name = download_reverse_model_folder(config)
mmi_model, mmi_tokenizer = load_model(mmi_target_folder_name, config)
else:
mmi_model = None
mmi_tokenizer = None
# Run chatbot with GPT-2
run_chat()
def main():
# Script arguments can include path of the config
arg_parser = argparse.ArgumentParser()
arg_parser.add_argument('--config', type=str, default="chatbot.cfg")
args = arg_parser.parse_args()
# Read the config
config = configparser.ConfigParser(allow_no_value=True)
with open(args.config) as f:
config.read_file(f)
# Download model artifacts
target_dir = download_model_folder(config)
# Load model and tokenizer
model, tokenizer = load_model(target_dir, config)
# Run chatbot with GPT-2
run_chat(model, tokenizer, config)
def main():
# Script arguments can include path of the config
arg_parser = argparse.ArgumentParser()
arg_parser.add_argument('--config', type=str, default="chatbot.cfg")
args = arg_parser.parse_args()
# Read the config
config = configparser.ConfigParser(allow_no_value=True)
with open(args.config) as f:
config.read_file(f)
# Download and load main model
target_folder_name = download_model_folder(config)
# added by Weijian, avoid re-downloading the model
# data_folder = config.get('model', 'data_folder')
# model_size = config.get('model', 'model_size')
# dataset = config.get('model', 'dataset')
# from_scratch = config.getboolean('model', 'from_scratch')
# target_folder_name = model_size + "_" + dataset + ("_fs" if from_scratch else "_ft")
model, tokenizer = load_model(target_folder_name, config)
# Download and load reverse model
use_mmi = config.getboolean('model', 'use_mmi')
if use_mmi:
mmi_target_folder_name = download_reverse_model_folder(config)
mmi_model, mmi_tokenizer = load_model(mmi_target_folder_name, config)
else:
mmi_model = None
mmi_tokenizer = None
# Run Telegram bot
bot = TelegramBot(model,
tokenizer,
config,
mmi_model=mmi_model,
mmi_tokenizer=mmi_tokenizer)
bot.run_chat()
app = Flask(__name__)
# Script arguments can include path of the config
arg_parser = argparse.ArgumentParser()
arg_parser.add_argument('--config', type=str, default="emlyon-chatbot.cfg")
arg_parser.add_argument('--host', type=str, default="0.0.0.0")
arg_parser.add_argument('--port', type=str, default="5011")
args = arg_parser.parse_args()
# Read the config
config = configparser.ConfigParser(allow_no_value=True)
with open(args.config) as f:
config.read_file(f)
# Download and load main model
target_folder_name = download_model_folder(config)
model, tokenizer = load_model(target_folder_name, config)
# Download and load reverse model
use_mmi = config.getboolean('model', 'use_mmi')
if use_mmi:
mmi_target_folder_name = download_reverse_model_folder(config)
mmi_model, mmi_tokenizer = load_model(mmi_target_folder_name, config)
else:
mmi_model = None
mmi_tokenizer = None
@app.route('/query')
def query():
# Parse parameters
num_samples = config.getint('decoder', 'num_samples')
personality = random.choice(personalities)
# logger.info("Selected personality: %s", tokenizer.decode(chain(*personality)))
print(personality)
l = chain(*personality)
custom_personality_text = ['I am student from India', 'i work in the field of Computer science', 'i like playing cricket', 'In my free time I like conducting talks for students to learn',
'I am a huge fan of IPL', 'I like talking, its fun to talk!' ]
custom_personality = list(map(tokenizer.encode, custom_personality_text))
print(custom_personality)
logger.info("Selected personality: %s", tokenizer.decode(chain(*custom_personality)))
# Reddit dialogue bot part
config = configparser.ConfigParser(allow_no_value=True)
with open("chatbot.cfg") as f:
config.read_file(f)
target_dir = download_model_folder(config)
model_reddit, tokenizer_reddit = load_model(target_dir, config)
def cosine_similarity(l1, l2):
cosine = torch.mm(l1.unsqueeze(0), l2.unsqueeze(0).transpose(0, 1))
n1, n2 = l1.norm(), l2.norm()
value = cosine/(n1*n2)
return value
@app_flask.route('/get_response', methods=['GET', 'POST'])
def get_response():
print(request.json)
score = request.json['score']
history = request.json['history']
# if score < 0.5:
