Spark Predictor (Classification)

This node classifies/labels input data using a previously learned Spark ML classification model. Please note that all feature columns selected during model training must be present in the ingoing DataFrame.

Note: This node is not compatible with Spark MLlib models. For these models please use the Spark Predictor node.

This node requires at least Apache Spark 2.0.


Change prediction column name
When set, you can change the name of the prediction column. The default name is "Prediction (targetcolumn)".
Prediction Column
The desired name for the prediction column.
Append individual class probabilities
Select to append the class probability of each class to the output. For each class, a new column with name "P (targetcolumn=class)" will appended.
Append individual class probabilities
If class probabilities are appended, the suffix allows you to avoid duplicate column names. Can be empty.

Input Ports

Spark ML classification model to use.
Spark DataFrame containing the input data to classify.

Output Ports

Input DataFrame with appended prediction column and, if selected, columns for the class probabilities.


This node has no views




You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.