You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The issue is the division between multiclass and continuous.
The quick solution is to filter continuous by checking whether labels cover the integer of [0, C).
However, it causes the following issues:
numerical class labels of random integers are not allowed
1-indexed class labels are not allowed
rank regression will be mis-recognized as multiclass
To address those issues, we need to enhance input_validator or preprocessing BEFORE instantiating dataset class.
At least, we know that the given task is either classification or regression, so we should use this knowledge.
The text was updated successfully, but these errors were encountered:
The issue is the division between multiclass and continuous.
The quick solution is to filter continuous by checking whether labels cover the integer of [0, C).
However, it causes the following issues:
To address those issues, we need to enhance input_validator or preprocessing BEFORE instantiating dataset class.
At least, we know that the given task is either classification or regression, so we should use this knowledge.
The text was updated successfully, but these errors were encountered: