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

Data type specification and checking #103

Open
robertness opened this issue Jan 26, 2023 · 1 comment
Open

Data type specification and checking #103

robertness opened this issue Jan 26, 2023 · 1 comment
Labels
good first issue Good for newcomers UI Issues related to user interface

Comments

@robertness
Copy link
Collaborator

Is your feature request related to a problem? Please describe.
The type and domain of the variables in the data should be a first class citizen

Describe the solution you'd like

  • Way to explicitly specify type (and possibly range) of variables in the context variable
  • Informative errors when a method doesn't work with a provided data type

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
This is also related to making assumptions first class citizens.

@robertness robertness added good first issue Good for newcomers UI Issues related to user interface labels Jan 26, 2023
@adam2392
Copy link
Collaborator

I think we can follow a similar approach to scikit-learn and assume continuous by default and allow users to pass in a categorical mask (e.g. https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html).

Idk if range of variables is important tho?

Then, we could have private attributes for each method _supports_categorical, _supports_mixed, _supports_continuous that is checked during fit(...)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
good first issue Good for newcomers UI Issues related to user interface
Projects
None yet
Development

No branches or pull requests

2 participants