Hello everyone
I am a Masterstudent at the University of St.Gallen and currently developing a Chatbot for argumentative learning. Throug your amazing masterclass I’ve managed to create the basic chatbot functions, intents and also got my bot up and running on my own server with rasax. The final step, is that I need to implement a pretrained model and already defined functions(orginally from a slackbot) as a custom action. The original code for the bot is the following:
import slack
import ssl as ssl_lib
import certifi
from ArgumentBot.argument_bot import ArgumentBot
from farm.infer import Inferencer
argument_bot_sent = {}
SLACK_BOT_TOKEN = "<TOKEN>"
def start_argument_bot(web_client: slack.WebClient, user_id: str, channel: str):
argument_bot = ArgumentBot(channel)
welcome = argument_bot.get_message_payload()
response = web_client.chat_postMessage(**welcome)
argument_bot.timestamp = response["ts"]
if channel not in argument_bot_sent:
argument_bot_sent[channel] = {}
argument_bot_sent[channel][user_id] = argument_bot
@slack.RTMClient.run_on(event="message")
def message(**payload):
data = payload["data"]
web_client = payload["web_client"]
channel_id = data.get("channel")
user_id = data.get("user")
text = data.get("text")
if text and text.lower() == "start":
return start_argument_bot(web_client, user_id, channel_id)
if "subtype" not in data:
argument_bot = ArgumentBot(channel_id)
elements = predict_components(data.get("text"))
feedback = prepare_feedback(data.get("text"), elements)
response = argument_bot.get_feedback_payload(feedback)
web_client.chat_postMessage(**response)
def load_bert_model():
"""Load in the pre-trained model"""
global model
save_dir = 'Model'
model = Inferencer.load(save_dir)
def predict_components(text: str):
text_to_analyze = [{'text': '{}'.format(text)}]
result = model.inference_from_dicts(dicts=text_to_analyze)
annotated_text = [[i['label'], i['start'], i['end']] for i in result[0]['predictions'] if
i['probability'] > 0.75]
count = 0
count_claim = 0
count_premise = 0
elements = []
for ann in annotated_text:
if ann[0] != 'O':
elements.append({
'id': count,
'label': ann[0].lower(),
'start': ann[1],
'end': ann[2]
})
if ann[0].lower() == 'claim':
count_claim += 1
else:
count_premise += 1
else:
continue
count += 1
return elements, count_claim, count_premise
def prepare_feedback(text: str, elements: tuple):
feedback_text = "Hier kommt das Feedback zu Deiner Argumentation, " \
"Claims werden *fett* und Premises _kursiv_ dargestellt:\n\n\n"
before = 0
for e in elements[0]:
start = e['start']
end = e['end']
marker = '*' if e['label'] == 'claim' else '_'
feedback_text += text[before:start]
feedback_text += marker
feedback_text += text[start:end]
feedback_text += marker
before = end
if before == 0:
feedback_text += text
if elements[1] > elements[2] or elements[1] < 2:
if elements[1] < 2:
feedback_text += "\n\nIch würde dir empfehlen, deinen Text noch argumentativer zu gestalten. " \
"Versuche mindestens zwei Claims mit relevanten Premises zu stützen\n"
else:
feedback_text += "\n\nIch würde dir empfehlen, deinen Text noch argumentativer zu gestalten. " \
"Versuche Deine Claims besser mit relevanten Premises zu stützen\n"
else:
feedback_text += "\n\nIch empfinde Deine Argumentation als gelungen! " \
"Du hast mehrere Aussagen gemacht und diese mit relevanten Premises gestützt. Weiter so!\n"
return feedback_text
if __name__ == "__main__":
load_bert_model()
ssl_context = ssl_lib.create_default_context(cafile=certifi.where())
slack_token = SLACK_BOT_TOKEN
rtm_client = slack.RTMClient(token=slack_token, ssl=ssl_context)
rtm_client.start()
So tried to implement the relevant code in a custom action like this:
from rasa_sdk import Action
from farm.infer import Inferencer
class LoadBot(Action):
def name(self):
return "action_loadbot"
def run():
"""Load in the pre-trained model"""
global model
save_dir = 'Model'
model = Inferencer.load(save_dir)
class PredictComponents(Action):
def name(self):
return "action_predictcomponents"
def run(text: str):
text_to_analyze = [{'text': '{}'.format(text)}]
result = model.inference_from_dicts(dicts=text_to_analyze)
annotated_text = [[i['label'], i['start'], i['end']] for i in result[0]['predictions'] if
i['probability'] > 0.75]
count = 0
count_claim = 0
count_premise = 0
elements = []
for ann in annotated_text:
if ann[0] != 'O':
elements.append({
'id': count,
'label': ann[0].lower(),
'start': ann[1],
'end': ann[2]
})
if ann[0].lower() == 'claim':
count_claim += 1
else:
count_premise += 1
else:
continue
count += 1
return elements, count_claim, count_premise
class PrepareFeedback(Action):
def name(self):
return "action_preparefeedback"
def run(text: str, elements: tuple):
feedback_text = "Hier kommt das Feedback zu Deiner Argumentation, " \
"Claims werden *fett* und Premises _kursiv_ dargestellt:\n\n\n"
before = 0
for e in elements[0]:
start = e['start']
end = e['end']
marker = '*' if e['label'] == 'claim' else '_'
feedback_text += text[before:start]
feedback_text += marker
feedback_text += text[start:end]
feedback_text += marker
before = end
if before == 0:
feedback_text += text
if elements[1] > elements[2] or elements[1] < 2:
if elements[1] < 2:
feedback_text += "\n\nIch würde dir empfehlen, deinen Text noch argumentativer zu gestalten. " \
"Versuche mindestens zwei Claims mit relevanten Premises zu stützen\n"
else:
feedback_text += "\n\nIch würde dir empfehlen, deinen Text noch argumentativer zu gestalten. " \
"Versuche Deine Claims besser mit relevanten Premises zu stützen\n"
else:
feedback_text += "\n\nIch empfinde Deine Argumentation als gelungen! " \
"Du hast mehrere Aussagen gemacht und diese mit relevanten Premises gestützt. Weiter so!\n"
return feedback_text
But I did not work out. Can anybody help with this transformation of the code? Any steps in the right direction would be a help for me. 
Thank you!
as I have worked only on versions when rasa nlu and core were combined into rasa. Is it possible for you to upgrade? If so I highly recommend you to upgrade to the latest version.