Interactive Learning in rasa_core0.12.0

Rasa Open Source
1

Hi, I am facing issue when I run interactive training. My code is this;

    import logging

    from rasa_core import utils, train
    from rasa_core.training import interactive

    logger = logging.getLogger(__name__)


    def train_agent():
        return train.train_dialogue_model(domain_file="domain.yml",
                                          stories_file="data/stories.md",
                                          output_path="model/dialogue",
                                          policy_config = "config_nlu.yml",
                                          kwargs={"batch_size": 50,
                                                  "epochs": 200,
                                                  "max_training_samples": 300
                                                  })


    if __name__ == '__main__':
        utils.configure_colored_logging(loglevel="INFO")
        agent = train_agent()
        interactive.run_interactive_learning(agent, stories="data/stories.md")

when it executed it shows the following output:

Screenshot%20from%202018-11-20%2010-35-30
Screenshot%20from%202018-11-20%2010-35-301366×768 220 KB
Screenshot%20from%202018-11-20%2010-35-42
Screenshot%20from%202018-11-20%2010-35-421366×768 198 KB

Can anyone help me to get rid of this?

2

You should update rasa_core to 0.12.1, there was a bug with the interactive learning that they’ve fixed.

3

Even after update I got the same error.

4

Resolved!!!

5

@abhishakskilrock Can you please tell how you resolved the error? I am getting the same error. But I added policy_config line and it was kind of working. But I now get an Internal Server error 500 after I enter any input. Can you please tell me how you resolved this issue? Thanks

6

@Arunkumar0258 run this command it will work: python -m rasa_core.train interactive -o models/dialogue -d domain_form.yml -s data/stories_form.md --nlu models/default/nlu --endpoints endpoints.yml

7

@abhishakskilrock I know that command works, but I want to make a Python script that does that exact thing. Btw I got the answer, I changed my code a bit and now it works. Thanks anyways :slight_smile:

8

@abhishakskilrock Hey if you want to run the file you ran first just try to import these files as well.

from future import absolute_import

from future import division

from future import print_function

from future import unicode_literals

9

Thanx @Akshit for helping me out but I have already import those libraries and Issue was not there issue was with calling stories with agent. We have to call stories file when we are calling agent

10

Hey @Arunkumar0258 did you got data in nlu.md file?
Actually I didn’t get any data in nlu.md. I got blank file.

11

@abhishakskilrock Yes I have the data in nlu.md file. I put it there myself, it’s were I put my intents. I have another file stories.md where I store my stories for all the intents.