DecisionTree Example 2: Alternate Input Format - Teradata Vantage

Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
8.00
1.0
Published
May 2019
Language
English (United States)
Last Update
2019-11-22
dita:mapPath
blj1506016597986.ditamap
dita:ditavalPath
blj1506016597986.ditaval
dita:id
B700-4003
lifecycle
previous
Product Category
Teradata Vantageā„¢

Input

The InputTable is a combination of the attribute table and response table of DecisionTree Example 1.

InputTable: iris_altinput
pid attribute attrvalue response
1 petal_width 0.2 1
1 petal_length 1.4 1
1 sepal_width 3.5 1
1 sepal_length 5.1 1
2 petal_width 0.2 1
2 petal_length 1.4 1
2 sepal_width 3 1
2 sepal_length 4.9 1
3 petal_width 0.2 1
3 petal_length 1.3 1
3 sepal_width 3.2 1
3 sepal_length 4.7 1
4 petal_width 0.2 1

SQL Call

DROP TABLE iris_attribute_output_2;
DROP TABLE splits_small_2;

SELECT * FROM DecisionTree (
  ON iris_altinput AS InputTable
  OUT TABLE OutputTable (iris_attribute_output_2)
  OUT TABLE IntermediateSplitsTable (splits_small_2)
  USING
  NumSplits (3)
  SplitMeasure ('gini')
  MaxDepth (10)
  IDColumns ('pid')
  AttributeNameColumns ('attribute')
  AttributeValueColumn ('attrvalue')
  ResponseColumn ('response')
  MinNodeSize (10)
  ApproxSplits ('false')
) AS dt;

Output

message
Output model table is successfully stored in the table specified in OutputTable argument.
Depth of the tree is:6

These queries return the attribute and the splits table (which are the same as in DecisionTree Example 1):

SELECT * FROM iris_attribute_output_2 ORDER BY 1;
SELECT * FROM splits_small_2 ORDER BY 1;