[SPARK-30543] RandomForest add Param bootstrap to control sampling method

• Type: Improvement
• Status: Resolved
• Priority: Minor
• Resolution: Resolved
• Affects Version/s: 3.0.0
• Component/s: ML, PySpark
• Assignee: Ruifeng Zheng
• Reporter: Ruifeng Zheng
• Created: 17/Jan/20 07:47
• Updated: 23/Jan/20 08:48
• Resolved: 23/Jan/20 08:48

Description

Current RF with numTrees=1 will directly build a tree using the orignial dataset,

while with numTrees>1 it will use bootstrap samples to build trees.

This design is to train a DecisionTreeModel by the impl of RandomForest, however, it is somewhat strange.

In Scikit-Learn, there is a param bootstrap to control bootstrap samples are used.

Issue Links

• links to: GitHub Pull Request #27254