Capability: the estimator can handle missing data, e.g,, NaNs.

  • String name: "capability:missing_values"

  • Public capability tag

  • Values: boolean, True / False

  • Example: True

  • Default: False

  • Alias: handles-missing-data (forecasters, transformations)

If the tag is True, the estimator can handle missing data, e.g., NaNs in input data.

This applies to main and secondary input data where applicable, e.g., X in fit of transformations and classifieres, or y in forecasters, but not to target labels in the case of labelling of entire time series, such as in classification or regression.

If the tag is False, the estimator cannot handle missing data, and will raise an error if missing data is encountered.