grassp.pp.impute_knn

impute_knn(data, k=20, weights='uniform', missing_values=0, inplace=True)

Impute missing values using KNN imputation.

This function imputes missing values (zeros) in the data matrix using KNN imputation. The parameters of the KNN are derived from the observed (non-zero) values. :type data: AnnData :param data: Annotated data matrix with proteins as observations (rows). :type k: int (default: 20) :param k: Number of neighbors to use for imputation. :type weights: str | callable (default: 'uniform') :param weights: Weight function to use for imputation. :type missing_values: int | float | str | np.nan (default: 0) :param missing_values: Value to use for missing values.

Return type:

np.ndarray | None

Parameters:

data AnnData

k int

weights str | callable

missing_values int | float | str | np.nan

inplace bool