How to train the nlu train model in UNIX box

Rasa Open Source
1

Hi All

How to train the nlu train model in UNIX box. I have implemented 2 User Stories and it is working fine. When i do the 3rd one I am facing issue so i need to train the nlu model online in Unix server.

Thanks Kamesh

2

What kind of errors getting ?

3

image

from future import absolute_import from future import division from future import unicode_literals

import logging

from rasa.core.agent import Agent from rasa.core.policies.keras_policy import KerasPolicy from rasa.core.policies.memoization import MemoizationPolicy

if name == ‘main’: logging.basicConfig(level=‘INFO’)

training_data_file = './data/stories.md'
model_path = './models/dialogue'

#agent = Agent('weather_domain.yml', policies = [MemoizationPolicy(), KerasPolicy()])
agent = Agent('workbenchdata.yml', policies = [MemoizationPolicy(), KerasPolicy()])

data = agent.load_data(training_data_file)
agent.train(data)	
agent.persist(model_path)
4

Which version you are using for RASA ? Use latest version.

5

image

6

Ok. You are using almost new version. That’s fine.

As per my knowledge, in new version for training, you need to fire command like this rasa train. And for polices you need to write config.yml file. Check out new documentation.

usage: rasa [-h] [–version] {init,run,shell,train,interactive,test,visualize,data,x} …

Rasa command line interface. Rasa allows you to build your own conversational assistants :robot:. The ‘rasa’ command allows you to easily run most common commands like creating a new bot, training or evaluating models.

positional arguments:
  {init,run,shell,train,interactive,test,visualize,data,x}
                        Rasa commands
    init                Creates a new project, with example training data,
                        actions, and config files.
    run                 Starts a Rasa server with your trained model.
    shell               Loads your trained model and lets you talk to your
                        assistant on the command line.
    train               Trains a Rasa model using your NLU data and stories.
    interactive         Starts an interactive learning session to create new
                        training data for a Rasa model by chatting.
    test                Tests Rasa models using your test NLU data and
                        stories.
    visualize           Visualize stories.
    data                Utils for the Rasa training files.
    x                   Starts the Rasa X interface.

optional arguments:
  -h, --help            show this help message and exit
  --version             Print installed Rasa version

OR I DIDN’t EXPLORED YOUR WAY… sorry brother !

7

Thanks Shiva this did not resolve my issue. One more question how to give new Intent in unix server. Till now i was specifying Intent type directly in data.json file which i don t want to do since some times Intent is not working as expected.

image

8

You can generate demo app by RASA’s in built command. Try it. Explore new structure or RASA.

Go in empty dictionary and then fire command - rasa init. It will create demo app. You should explore it first. Then go further.

9

If your intent is not correctly predicted by RASA then, you should add more data in nlu.md file. Ultimately, you need more data to train it and yes, your config.yml file must have proper pipeline.