fatf.transparency.sklearn.linear_model.linear_classifier_coefficients

fatf.transparency.sklearn.linear_model.linear_classifier_coefficients(clf: sklearn.base.BaseEstimator) → numpy.ndarray[source]

Extracts coefficients (feature importances) of a linear scikit-learn model.

New in version 0.0.2.

Note

Please note that for the coefficients (feature importances) to be comparable the values of all features had to be normalised to the same range before training the model.

Parameters
clfsklearn.base.BaseEstimator

A linear scikit-learn model.

Returns
coefficientsnumpy.ndarray

A numpy array that holds coefficients of the clf linear model. (The order of the coefficients corresponds to the order of the features in the training data array).

Raises
sklearn.exceptions.NotFittedError

The scikit-learn package (sklearn.utils.validation.check_is_fitted function) will raise this exception if the model is not fitted.

TypeError

The clf classifier is not a scikit-learn linear model.