pyproteonet.imputation.fancyimpute.iterative_svd_impute

pyproteonet.imputation.fancyimpute.iterative_svd_impute(dataset: Dataset, molecule: str, column: str, result_column: str | None = None, min_value: float | None = None, rank: float | int = 0.2, **kwargs) → Dataset

Impute missing values using iterative singular value decomposition.

Parameters:

• dataset (Dataset) – Dataset to impute.

• molecule (str) – Molecular type to impute.

• column (str) – Value column to impute.

• result_column (Optional[str], optional) – If given imputed results are stored under this name. Defaults to None.

• min_value (Optional[float], optional) – If given imputation are restricted by this value, otherwise the minimum of the dataset is used. Defaults to None.

• rank (Union[float, int], optional) – Rank of the low dimensional representation, either given as integer or float <= 1.0 representing the fraction of number of samples used as rank. Defaults to 0.2.

Returns:

imputed values

Return type:

pd.Series