Scoring API - Teradata Vantage

Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
8.10
1.1
Published
October 2019
Language
English (United States)
Last Update
2019-12-31
dita:mapPath
ima1540829771750.ditamap
dita:ditavalPath
jsj1481748799576.ditaval
dita:id
B700-4003
lifecycle
previous
Product Category
Teradata Vantageā„¢

The scoring APIs are documented in the javadoc. After installation, you can invoke Scorer by submitting scoring requests to either the file system or the API.

Scoring Requests from File System

This usage provides a simple interface to test scorer functionality. (The code blocks show high-level method calls to invoke Scorer.)

// initialize
Scorer scorer = new Scorer ();
// configure
// make sure that AML file modelFile is available on file system
scorer.configure (modelFile);
// run scorer (multiple calls)
// make sure that CSV file requestFile is available on file system
scorer.score (requestFile);

Scoring Requests Using API

Teradata recommends this usage in a production environment. By preventing the I/O overheads from populating and reading request files from the file system, it improves performance. (The code blocks show high-level method calls to invoke Scorer.)

// initialize
Scorer scorer = new Scorer ();
// configure
// make sure that AML file modelFile is available on file system
Request request = scorer.configure (modelFile);
// populate data structure request
// run scorer (multiple calls)
scorer.score (request);