GLM2Predict Output - 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 Vantage™

The output table schema depends on whether you specify the regularization table and Lambda argument.

Output Table Schema, Regularization Table or Lambda Omitted

If you omit the regularization table, or specify the regularization table but omit the Lambda argument, the output table has the following schema.

Column Data Type Description
accumulate_column Same as in input table [Column appears once for each specified accumulate_column.] Column copied from input table.
prediction DOUBLE PRECISION Score of input data, given by equation g-1(Xβ), where g-1 is the inverse link function, X is the predictors, and β is the vector of coefficients estimated by the GLM2 function.

For BINOMIAL classification, a predicted value close to 1 indicates a high probability of class 1. A predicted value close to 0 indicates a high probability of class 0. For other values of Family, scores are expected values of dependent/response variable, conditional on predictors.

lambda DOUBLE PRECISION Lambda value used for prediction.

If you omit both the regularization table and Lambda argument, this value is the minLambda value from the model table.

Output Table Schema, Regularization Table and Lambda Specified

Column Data Type Description
accumulate_column Same as in input table [Column appears once for each specified accumulate_column.] Column copied from input table.
prediction DOUBLE PRECISION Score of input data, given by equation g-1(Xβ), where g-1 is the inverse link function, X is the predictors, and β is the vector of coefficients estimated by the GLM2 function.

For BINOMIAL classification, a predicted value close to 1 indicates a high probability of class 1. A predicted value close to 0 indicates a high probability of class 0. For other values of Family, scores are expected values of dependent/response variable, conditional on predictors.

reqlambda DOUBLE PRECISION Lambda value that you specified.
matchedlambda DOUBLE PRECISION Lambda value used for prediction.