The HPFOREST Procedure

PROC HPFOREST Features

The HPFOREST procedure creates an ensemble of hundreds of decision trees to predict a single target of either interval or nominal measurement level. An input variable can have an interval, ordinal, or nominal measurement level.

The HPFOREST procedure deletes from the training data any observation that has a missing target value or a FREQ variable whose value is less than or equal to 0.

Because the HPFOREST procedure is a high-performance analytical procedure, it also does the following:

enables you to run in distributed mode on a cluster of machines that distribute the data and the computations
enables you to run in single-machine mode on the server where SAS is installed
exploits all the available cores and concurrent threads, regardless of execution mode

Beginning with this release, the training algorithm uses concurrent threads whenever available. However, this release of PROC HPFOREST copies all the data to all cores in distributed mode. For more information, see the section Processing Modes in Chapter 2: Shared Concepts and Topics.