Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add set.seed ability to CLI when splitting NLU data #4647

Closed
3 tasks
neelkes opened this issue Oct 21, 2019 · 4 comments · Fixed by #4842
Closed
3 tasks

Add set.seed ability to CLI when splitting NLU data #4647

neelkes opened this issue Oct 21, 2019 · 4 comments · Fixed by #4842
Assignees
Labels
help wanted type:enhancement ✨ Additions of new features or changes to existing ones, should be doable in a single PR

Comments

@neelkes
Copy link

neelkes commented Oct 21, 2019

Description of Problem:
The user is not able to set a seed when using the CLI to split NLU data into test and training data sets (rasa split data nlu command). This feature will enhance reproducibility when required.

Overview of the Solution:
Add set.seed() to split_nlu_data.py or train_test_split.py

Definition of Done:

  • Tests are added
  • Feature described the docs
  • Feature mentioned in the changelog
@neelkes neelkes added the type:enhancement ✨ Additions of new features or changes to existing ones, should be doable in a single PR label Oct 21, 2019
@sara-tagger
Copy link
Collaborator

Thanks for submitting this feature request 🚀@Ghostvv will get back to you about it soon!✨

Caw! 🐧

@tabergma
Copy link
Contributor

Great idea 👍 We could add an argument like --random-seed to rasa split data nlu to set the seed. Needs to be added here: https://github.com/RasaHQ/rasa/blob/master/rasa/cli/data.py. And the seed should be forwarded to https://github.com/RasaHQ/rasa/blob/master/rasa/nlu/training_data/training_data.py#L400 in order to set it.

@neelkes Do you want to work on this feature yourself and submit a PR?

@joaorobson
Copy link
Contributor

I found this enhancement pretty useful. Can I be assigned to this issue? (Supposing that @neelkes is not working with this feature anymore)

@tabergma
Copy link
Contributor

@joaorobson Sure, feel free to work on it. Thanks! Let me know if you need help/if you have questions.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
help wanted type:enhancement ✨ Additions of new features or changes to existing ones, should be doable in a single PR
Projects
None yet
Development

Successfully merging a pull request may close this issue.

5 participants