hidimstat.CPI

class hidimstat.CPI(estimator, imputation_model, n_permutations: int = 50, loss: callable = <function root_mean_squared_error>, score_proba: bool = False, random_state: int = None, n_jobs: int = 1)

Conditional Permutation Importance (CPI) algorithm. CHAMMA et al.[1] and for group-level see Chamma et al.[2].

Parameters:

estimator: scikit-learn compatible estimator

The predictive model.

imputation_model: scikit-learn compatible estimator or list of estimators

The model(s) used to estimate the covariates. If a single estimator is provided, it will be cloned for each covariate group. Otherwise, a list of potentially different estimators can be provided, the length of the list must match the number of covariate groups.

n_permutations: int, default=50

Number of permutations to perform.

loss: callable, default=root_mean_squared_error

Loss function to evaluate the model performance.

score_proba: bool, default=False

Whether to use the predict_proba method of the estimator.

random_state: int, default=None

Random seed for the permutation.

n_jobs: int, default=1

Number of jobs to run in parallel.

References

[1]

Ahmad CHAMMA, Denis A. Engemann, and Bertrand Thirion. Statistically valid variable importance assessment through conditional permutations. In A. Oh, T. Naumann, A. Globerson, K. Saenko, M. Hardt, and S. Levine, editors, Advances in Neural Information Processing Systems, volume 36, 67662–67685. Curran Associates, Inc., 2023. URL: https://proceedings.neurips.cc/paper_files/paper/2023/file/d60e14c19cd6e0fc38556ad29ac8fbc9-Paper-Conference.pdf.

[2]

Ahmad Chamma, Bertrand Thirion, and Denis Engemann. Variable importance in high-dimensional settings requires grouping. Proceedings of the AAAI Conference on Artificial Intelligence, 38(10):11195–11203, Mar. 2024. URL: https://ojs.aaai.org/index.php/AAAI/article/view/28997, doi:10.1609/aaai.v38i10.28997.

__init__(estimator, imputation_model, n_permutations: int = 50, loss: callable = <function root_mean_squared_error>, score_proba: bool = False, random_state: int = None, n_jobs: int = 1)

Methods

__init__(estimator, imputation_model[, ...])
fit(X[, y, groups])	Fit the covariate estimators to predict each group of covariates from the others.
get_metadata_routing()	Get metadata routing of this object.
get_params([deep])	Get parameters for this estimator.
predict(X[, y])	Compute the prediction of the model with perturbed data for each group.
score(X, y)	Compute the importance scores for each group of covariates.
set_fit_request(*[, groups])	Request metadata passed to the fit method.
set_params(**params)	Set the parameters of this estimator.