-
Parameterize a PMML Scoring analysis to score a Logistic Regression model which predicts a discrete outcome as follows:
- Select File Name — RegressionDiscretePMML.xml
- Selected Tables — twm_customer_analysis
- Index Columns — cust_idd
- Result Table Name — twm_pmml_score_reg_2
-
Model Output Options
- P_ccacct1 — Enabled
- P_ccacct0 — Enabled
- Run the analysis.
-
Click on Results when it completes.
For this example, the PMML Scoring analysis generated the following results. A single click on each page name populates Results with the item. SQL is always produced and displayed, but is not shown here for brevity.
Tutorial #3 - PMML Scoring Report Resulting Scored Table Name Number of Rows in Scored File score_reg_2 747 Tutorial #3 - PMML Scoring Data cust_id ccacct P_ccacct1 P_ccacct0 1362527 0 0.125740481096571 0.874259518903429 1363078 1 0.861086203667224 0.138913796332776 1362588 1 0.723429148501114 0.276570851498886 1363486 0 0.125034199014627 0.874965800985373 1362752 0 0.419312298702164 0.580687701297836 1362893 1 0.970060355675886 0.0299396443241139 1363017 1 0.999980678896465 1.93211035354291E-05 1363444 0 0.173538764837706 0.826461235162294 1362548 0 0.265964538752992 0.734035461247008 1362487 1 0.872345777062174 0.127654222937826 ... ... ... ... ... ... ... ... ... ... ... ...