Machine Learning: Prediction On Random Forest Tree Is Slow (Doc ID 2780390.1)

Last updated on JULY 20, 2024

Applies to:

Oracle Data Mining - Version 12.1.0.2 and later
Information in this document applies to any platform.

SQL query against large Random Forest mining model takes too long

Predicting from a created random forest tree model is very slow.

A query with 1 or about 1 Million rows takes up to 10 minutes, Query is expected to run within seconds,

The data mining scoring code uses SQL operators (PREDICTION_SET, FEATURE_SET,CLUSTER_SET) that return VARRAYs of objects,

The unnesting operation is a serial bottleneck and does not parallelize. This represents a serious performance issue.

	To view full details, sign in with your My Oracle Support account.
	Don't have a My Oracle Support account? Click to get started!

In this Document

Symptoms

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.