With the Simple Linear Regression model, users can evaluate the model using the accuracy score. This results in a hard crash with the following error:
ValueError: Classification metrics can't handle a mix of continuous and binary targets
Expected Behavior
This method of evaluation does not work with this type of model. Maybe remove it as an option or provide an error message stating that this type of evaluation does not work with the model we have selected.
Steps To Reproduce
Navigate through the website process until the training/model selection step.
Select Simple Linear Regression.
Continue navigating the form. We used the default values for hyperparameters.
For model evaluation, select accuracy score and submit.
It's a very valid Issue. Thanks for bringing it to our notice. I fixed it by providing some extra information to select a relevant metric according to the selected model.
Is there an existing issue for this?
Current Behavior
With the Simple Linear Regression model, users can evaluate the model using the accuracy score. This results in a hard crash with the following error:
ValueError: Classification metrics can't handle a mix of continuous and binary targets
Expected Behavior
This method of evaluation does not work with this type of model. Maybe remove it as an option or provide an error message stating that this type of evaluation does not work with the model we have selected.
Steps To Reproduce
Environment
No response
Anything else?
No response