midlearn.MIDConditional

class midlearn.MIDConditional(estimator: MIDRegressor | MIDExplainer, variable: str, pred_type: str = 'response', **kwargs)[source]

MID Conditional Expectations.

This object is returned by the MIDRegressor.conditional() method and contains data for plotting conditional dependence.

__init__(estimator: MIDRegressor | MIDExplainer, variable: str, pred_type: str = 'response', **kwargs)[source]

Initialize the MIDConditional object.

Parameters:

  • estimator (MIDRegressor or MIDExplainer) – The fitted MID model instance to use.

  • variable (str) – The name of the feature for which to calculate conditional dependence.

  • pred_type (str) – The scale on which to calculate and plot the conditional expectations.

  • **kwargs (dict) – Additional keyword arguments passed to the midr::mid.conditional() function in R.

Methods

__init__(estimator, variable[, pred_type])

Initialize the MIDConditional object.

plot([style, theme, var_color, dots, reference])

Visualize Individual Conditional Expectation (ICE) plots or Centered ICE (c-ICE) plots with plotnine.

terms(**kwargs)

Extract term labels from the fitted model.

Attributes

conditional

pd.DataFrame of the hypothetical observations and their corresponding predictions.

observed

pd.DataFrame of the original observations used.

values

A vector or the sample points for the 'variable' used in the ICE calculation.