gemseo.mlearning.linear_model_fitting.base_linear_model_fitter module

Base class for linear model fitting algorithms.

class BaseLinearModelFitter(settings=None)

Bases: Generic[FitterType, SettingsType]

Base class for linear model fitting algorithms.

Initialize self. See help(type(self)) for accurate signature.

Parameters:

settings (SettingsType | None) -- The settings of the linear model fitting algorithm. If None, use a default instance of Settings.

fit(input_data, output_data, *extra_data)

Fit a linear model to data.

Parameters:

• input_data (RealArray) -- The features matrix, shaped as (n_samples, n_features).

• output_data (RealArray) -- The observations matrix, shaped as (n_samples, n_targets).

• *extra_data (tuple[RealArray, RealArray]) -- Additional pairs "(features matrix, observations matrix)". where a features matrix has n_features columns, an observations matrix as n_targets columns and the matrices of a pair have the same number of rows. E.g., Jacobian observations. This argument cannot be used when the option fit_intercept is True.

Returns:

The coefficients of the linear model, shaped as (n_targets, n_features).

Return type:

RealArray

Settings: ClassVar[type[BaseLinearModelFitter_Settings]]

The class for defining the settings of the linear model fitting algorithm.