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 copied from input_table. |
prediction | DOUBLE PRECISION | Score of the input data, given by the equation g-1(Xβ), where g-1 is the inverse link function, X the predictors, and β is the vector of coefficients estimated by the GLM2 function. For other values of Family, the scores are the expected values of dependent/response variable, conditional on the 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. |
If you specify both the regularization table and the Lambda argument, the output table has the following schema.
Column | Data Type | Description |
---|---|---|
accumulate_column | Same as in input_table | Column copied from input_table. |
prediction | DOUBLE PRECISION | Score of the input data, given by the equation g-1(Xβ), where g-1 is the inverse link function, X the predictors, and β is the vector of coefficients estimated by the GLM2 function. For other values of Family, the scores are the expected values of dependent/response variable, conditional on the predictors. |
reqlambda | DOUBLE PRECISION | Lambda value that you specified. |
matchedlambda | DOUBLE PRECISION | Lambda value used for prediction. |