Im using the CLI to train my RASA NLU model and was wondering if it is possible to set the seed when doing rasa data split nlu so that i dont have to keep the train and test seed but can reproduce it if needed?
Unfortunately you currently can’t do that. Would you mind creating a feature request in the
rasa repository or do a PR with the required changes? Would be really cool to get this into
Hi @Tobias_Wochinger I added a feature request, and if time allows will look at potentially doing a PR. Thanks for responding