Does the training data format has changed a lot in the new version?

I want to use ‘rasa data validate’ to validate the data first, but it doesn’t have an argument for me pass the domain/nlu/core file to the command line.

rasa data --help
usage: rasa data [-h] [-v] [-vv] [--quiet]
                 {convert,split,validate,migrate} ...

Does it always use a default data path?

@lingvisa nice question, but I guess rasa data validate command by default take all the mentioned files for validation i.e

Data validation verifies that no mistakes or major inconsistencies appear in your domain, NLU data, or story data. To validate your data, have your CI run this command:

rasa data validate

I’d recommend to see this rasa doc OR I guess you already saw this ? : Testing Your Assistant


(Command Line Interface)

If you have any further doubt please asked :slight_smile:

I hope this will help you.

So it validates files recursively from its current directory and all its subdirectories. I think it would be great if it can provide a --data Data option, just as the rasa train command.

@lingvisa it’s a nice suggestion, hope the rasa team in near future will update this :slight_smile: . If my suggestion helps you can I request you to please close this thread or if you have any burning questions please go ahead. Thanks