Replace All Missings

Synopsis

This operator handles all missing values in a data set automatically.

Description

A universal missing value handler which handles nominal values by replacing missings with a new value MISSING and replaces missings and infinites in numerical columns with the average of the non-missings. If the complete column is missing, all values will become zero. For date columns, it is using the first known date or a zero date (1 Jan 1970) again if all are missing.

The advantage of this operator is that it handles all data types as well as missing and infinite values and produces a single preprocessing model to deal with all those situations for other data sets, i.e. in scoring scenarios. While this is simpler, it is also less flexible and powerful than using the dedicated missing value handling operators.

Input

example set input

This port expects an ExampleSet for which all missing and infinite values should be replaced.

Output

example set output

The processed data without any missings or infinite values.

original

The original data set.

预处理模型

You can apply this model on new data sets with Apply Model to get the same missing value handling than for the specified data set.