GLML1L2 Example 2: Factor Table as InputTable - 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 input table is admissions_factor_table, output by GLML1L2 Example 1: Ridge Regression, Family ('BINOMIAL'). In admissions_factor_table, categorical predictors were converted to integers.

SQL Call

Because the admissions_factor_table has only integers, this call does not specify CategoricalColumns. Because Randomization was true in the SQL call that created admissions_factor_table, this call does not need it.

DROP TABLE admissions_model_2;

CREATE MULTISET TABLE admissions_model_2 AS (
  SELECT * FROM GLML1L2 (
    ON admissions_factor_table AS InputTable
    USING
    FeatureColumns ('[0:5]')
    ResponseColumn ('admitted') 
    Family ('BINOMIAL') 
    Alpha (0)
    Lambda (0.02)   
  ) AS dt
) WITH DATA;

Output

admissions_model_2
attribute estimate information
(Intercept) 0.383816240766463 p
masters -1.26525302726537 p
stats 0.0806346550146325 p
stats -0.0267165533072416 p
programming -1.02594302137308 p
programming -0.0820786516340258 p
gpa 0.383464234338727 p
Family   Binomial
Regularization   Ridge
Alpha 0  
Lambda 0.02  
Iterations # 28  
Converged   true
Rows # 40  
Features # 6  
AIC 15.2192798193498  
BIC 27.0414359981473