0 ×

Spark Random Forests Learner

KNIME Extension for Apache Spark core infrastructure version 2.4.0.v201811301556 by KNIME AG, Zurich, Switzerland

This node applies the Apache Spark Random Forests algorithm.

Please note that all data must be numeric, including the label column. Use the Spark Category To Number nodes to convert nominal values to numeric columns. The mapping dictionary of the converter node is the input of the second in port of this node.

Use the Spark Predictor node to apply the learned model to unseen data.

Options

Number of trees
Number of trees in the random forest.
Max depth
Maximum depth of the tree. E.g., depth 0 means 1 leaf node; depth 1 means 1 internal node + 2 leaf nodes.
Max number of bins
Maximum number of bins used for splitting features.
Feature selection strategy
Number of features to consider for splits at each node. Supported:
  • auto
  • all
  • sqrt
  • log2
  • onethird
If "auto" is set, this parameter is set based on numTrees: if numTrees == 1, set to "all"; if numTrees > 1 (forest) set to "sqrt".
Is classification
Select this option for binary or multiclass classification.
Quality measure
Criterion used for information gain calculation. Available methods:
  • gini (recommended)
  • entropy
For more details on the available methods see the MLlib documentation. Only available for classifications.
Seed
Random seed for bootstrapping and choosing feature subsets.
Class column
The classification column. Must be numeric.
Feature Columns
The feature columns to learn the model from. Supports only numeric columns.

Input Ports

Input Spark DataFrame/RDD
PMML with the nominal values mapping dictionary

Output Ports

Spark MLlib Random Forests Model

Best Friends (Incoming)

Best Friends (Outgoing)

Workflows