Skip to content
This repository has been archived by the owner on Jun 30, 2022. It is now read-only.

calling fit returns Status details: Error : Unknown label type: continuous #6

Open
shadiakiki1986 opened this issue Jan 21, 2018 · 3 comments

Comments

@shadiakiki1986
Copy link

shadiakiki1986 commented Jan 21, 2018

Hello. When I call fit from the mljar python API, it fails with the following status on each of the experiments: Status details: Error : Unknown label type: continuous

Edit: also, the fit call just hangs without returning any error

@pplonski
Copy link
Contributor

can you provide steps to reproduce?

@shadiakiki1986
Copy link
Author

The steps are just the default model = Mljar(...) followed by model.fit(...).
Here is the exact code

model = Mljar(
  project='g2-ml take2/ex3',
  #experiment='Get 0.58 A',
  experiment='RF enc-on-raw A',
  metric='auc',
  algorithms=['rfc'],
  validation_kfolds=15,
  validation_shuffle=False,
  validation_stratify=True,
  validation_train_split=None,
  tuning_mode='Sport',
  create_ensemble=True,
  single_algorithm_time_limit='10'
)

model.fit(train_features, train_target)

@shadiakiki1986
Copy link
Author

It seems that this was related to #8 where my target was accidentally continuous whereas the experiment was a binary classification. Feel free to close this issue

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants