search

cnellington / Contextualized

An SKLearn-style toolbox for estimating and analyzing models, distributions, and functions with context-specific parameters.
http://contextualized.ml/
GNU General Public License v3.0
65 stars 9 forks source link

add base predictor option to regression and easy modules #89 #92

Closed cnellington closed 2 years ago

cnellington commented 2 years ago

Added a base predictor kwarg to ContextualizedRegressionBase and enabled for all regression and easy classes. Added base predictor kwarg quicktests to regression and easy Resolves #89 for regression and easy

cnellington commented 2 years ago

Added the base_y_predictor and base_param_predictor kwargs for contextualized linear models. Adding a base_y_predictor makes the contextualized module predict only non-base effects, while adding a base_param_predictor will include the base effects in the contextualized parameter estimates. Or similarly, adding a base_param_predictor may help with learning heterogeneous effects but still includes homogeneous (base) effects in predictions, while adding a base_y_predictor will only learn/predict heterogeneous (non-base) effects.

These kwargs should be mutually exclusive in practice, but they're both used in the tests since the dummy classes are only there to help check tensor shapes and make sure we don't break torch autograd.

cnellington commented 2 years ago

Agreed, I think this type of warning extends to many of our kwargs so I'll make an issue to sanity-check arguments in all contextualized models.