Imputing With Mode Value

Imputing with most frequently observed value. It fills missing values (None) in selected columns with most frequently observed value in the corresponding column, in the incoming DataFrame.

Type

transform

Class

fire.nodes.ml.NodeReplaceMissingValueWithMode

Fields

Name	Title	Description
colNames	Columns	Columns to be processed for imputing the missing values.

Details

This node imputes the missing values in the specified columns by mode of the values in the column.

Examples

Incoming Dataframe has following rows and [AGE] column has missing value / [NULL] for some rows:

CUST_CD    |    CUST_NAME    |    AGE
------------------------------------------
C01        |    MATT         |    50
C02        |    LISA         |    45
C03        |    ROBIN        |
C04        |    MARCUS       |
C05        |    MARK         |    30
C06        |    DAVID        |    30
C07        |    ANTHONY      |    90

If imputingwithmodevalue node is configured to Impute [AGE] with mode value then missing values in [AGE] column would be replaced with 30 which is the mode of [AGE] column.

Outgoing Dataframe would result as below:

CUST_CD    |    CUST_NAME    |    AGE
------------------------------------------
C01        |    MATT         |    50
C02        |    LISA         |    45
C03        |    ROBIN        |    30
C04        |    MARCUS       |    30
C05        |    MARK         |    30
C06        |    DAVID        |    30
C07        |    ANTHONY      |    90