physbo.misc.permutation_importance module

physbo.misc.permutation_importance.get_permutation_importance(model, X, t, n_perm: int, comm=None, split_features_parallel=False, query_only=False)

Calculating permutation importance of model

パラメータ:

• X (numpy.ndarray) -- inputs

• t (numpy.ndarray) -- target (label)

• n_perm (int) -- number of permutations

• comm (MPI.Comm) -- MPI communicator

• query_only (bool) -- If True, call model.get_post_fmean with a single argument (query points only), for models like BLM where training is already set by prepare(). If False, use two arguments (training, query) as in GP. Default is False.

戻り値:

• numpy.ndarray -- importance_mean

• numpy.ndarray -- importance_std