Hi Gaurav,
Thanks for your reply. I was also thinking that it might be an authentication issue, after going through a few posts. But, the same code is able to generate a valid PMML (that I can use for forecasting with logistic regression), when I have 3 variables (e.g., with the example in the pdf doc). Should not it be the case that it would also fail, if it was due to authentication issue? Please find the code below for your reference, thanks again for helping out.
---------------------------------------------------------------------------------------------------------------------
--- Working Code generating valid PMML (from example, with 3 independent variables)
------------------------------------------------------------------------------------------------------------------------
SET SCHEMA DM_PAL;
DROP TYPE PAL_LOGISTICR_DATA_T;
CREATE TYPE PAL_LOGISTICR_DATA_T AS TABLE("V1" VARCHAR (50),"V2" DOUBLE,"V3" INTEGER,"CATEGORY" INTEGER);
DROP TYPE PAL_LOGISTICR_RESULT_T;
CREATE TYPE PAL_LOGISTICR_RESULT_T AS TABLE("Coefficient" varchar(50),"CoefficientValue" DOUBLE,"ZSCORE" DOUBLE, "Pr(>|Z|)" DOUBLE);
DROP TYPE PAL_CONTROL_T;
CREATE TYPE PAL_CONTROL_T AS TABLE("NAME" VARCHAR(50), "INTARGS" INTEGER, "DOUBLEARGS" DOUBLE, "STRINGARGS" VARCHAR(100));
DROP TYPE PAL_LOGISTICR_PMMLMODEL_T;
CREATE TYPE PAL_LOGISTICR_PMMLMODEL_T AS TABLE( "ID" INTEGER, "PMMLMODEL" VARCHAR(5000));
DROP table PAL_LOGISTICR_PDATA_TBL;
CREATE column table PAL_LOGISTICR_PDATA_TBL("POSITION" INTEGER, "SCHEMA_NAME" VARCHAR(100), "TYPE_NAME" VARCHAR(100), "PARAMETER_TYPE" VARCHAR(100));
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (1,'DM_PAL', 'PAL_LOGISTICR_DATA_T','IN');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (2,'DM_PAL', 'PAL_CONTROL_T','IN');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (3,'DM_PAL', 'PAL_LOGISTICR_RESULT_T','OUT');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (4,'DM_PAL', 'PAL_LOGISTICR_PMMLMODEL_T','OUT');
CALL SYS.AFLLANG_WRAPPER_PROCEDURE_DROP('DM_PAL', 'PAL_LOGISTICR_PROC');
CALL SYS.AFLLANG_WRAPPER_PROCEDURE_CREATE('AFLPAL','LOGISTICREGRESSION', 'DM_PAL', 'PAL_LOGISTICR_PROC',PAL_LOGISTICR_PDATA_TBL);
DROP TABLE PAL_LOGISTICR_DATA_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_DATA_TBL LIKE PAL_LOGISTICR_DATA_T;
INSERT INTO PAL_LOGISTICR_DATA_TBL VALUES ('B',2.62,0,1);
INSERT INTO PAL_LOGISTICR_DATA_TBL VALUES ('B',2.875,0,1);
INSERT INTO PAL_LOGISTICR_DATA_TBL VALUES ('A',2.32,1,1);
--- insert some more data ---
DROP TABLE #PAL_CONTROL_TBL;
CREATE LOCAL TEMPORARY COLUMN TABLE #PAL_CONTROL_TBL ("NAME" VARCHAR(50), "INTARGS" INTEGER, "DOUBLEARGS" DOUBLE, "STRINGARGS" VARCHAR(100));
INSERT INTO #PAL_CONTROL_TBL VALUES ('VARIABLE_NUM',3,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('METHOD', 0, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('EXIT_THRESHOLD',null,0.000001,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('THREAD_NUMBER',8,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('MAX_ITERATION',1000,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('CATEGORY_COL', 2, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('STAT_INF', 1, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('PMML_EXPORT', 1, null, null);
DROP TABLE PAL_LOGISTICR_RESULTS_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_RESULTS_TBL LIKE PAL_LOGISTICR_RESULT_T;
DROP TABLE PAL_LOGISTICR_PMMLMODEL_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_PMMLMODEL_TBL LIKE PAL_LOGISTICR_PMMLMODEL_T;
CALL DM_PAL.PAL_LOGISTICR_PROC(PAL_LOGISTICR_DATA_TBL, "#PAL_CONTROL_TBL", PAL_LOGISTICR_RESULTS_TBL, PAL_LOGISTICR_PMMLMODEL_TBL) WITH OVERVIEW;
DROP TYPE PAL_LOGISTICR_RESULT_T;
CREATE TYPE PAL_LOGISTICR_RESULT_T AS TABLE("Coefficient" varchar(50),"CoefficientValue" DOUBLE,"ZSCORE" DOUBLE, "Pr(>|Z|)" DOUBLE);
DROP TYPE PAL_CONTROL_T;
CREATE TYPE PAL_CONTROL_T AS TABLE("NAME" VARCHAR(50), "INTARGS" INTEGER, "DOUBLEARGS" DOUBLE, "STRINGARGS" VARCHAR(100));
DROP TYPE PAL_LOGISTICR_PMMLMODEL_T;
CREATE TYPE PAL_LOGISTICR_PMMLMODEL_T AS TABLE( "ID" INTEGER, "PMMLMODEL" VARCHAR(5000));
DROP table PAL_LOGISTICR_PDATA_TBL;
CREATE column table PAL_LOGISTICR_PDATA_TBL("POSITION" INTEGER, "SCHEMA_NAME" VARCHAR(100), "TYPE_NAME" VARCHAR(100), "PARAMETER_TYPE" VARCHAR(100));
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (1,'DM_PAL', 'PAL_LOGISTICR_DATA_T','IN');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (2,'DM_PAL', 'PAL_CONTROL_T','IN');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (3,'DM_PAL', 'PAL_LOGISTICR_RESULT_T','OUT');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (4,'DM_PAL', 'PAL_LOGISTICR_PMMLMODEL_T','OUT');
CALL SYS.AFLLANG_WRAPPER_PROCEDURE_DROP('DM_PAL', 'PAL_LOGISTICR_PROC');
CALL SYS.AFLLANG_WRAPPER_PROCEDURE_CREATE('AFLPAL','LOGISTICREGRESSION', 'DM_PAL', 'PAL_LOGISTICR_PROC',PAL_LOGISTICR_PDATA_TBL);
DROP TABLE #PAL_CONTROL_TBL;
CREATE LOCAL TEMPORARY COLUMN TABLE #PAL_CONTROL_TBL ("NAME" VARCHAR(50), "INTARGS" INTEGER, "DOUBLEARGS" DOUBLE, "STRINGARGS" VARCHAR(100));
INSERT INTO #PAL_CONTROL_TBL VALUES ('VARIABLE_NUM',30,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('METHOD', 0, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('EXIT_THRESHOLD',null,0.000001,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('THREAD_NUMBER',8,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('MAX_ITERATION',1000,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('CATEGORY_COL', 2, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('STAT_INF', 1, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('PMML_EXPORT', 1, null, null);
DROP TABLE PAL_LOGISTICR_RESULTS_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_RESULTS_TBL LIKE PAL_LOGISTICR_RESULT_T;
DROP TABLE PAL_LOGISTICR_PMMLMODEL_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_PMMLMODEL_TBL LIKE PAL_LOGISTICR_PMMLMODEL_T;
CALL DM_PAL.PAL_LOGISTICR_PROC(PAL_LOGISTICR_DATA_TBL, "#PAL_CONTROL_TBL", PAL_LOGISTICR_RESULTS_TBL, PAL_LOGISTICR_PMMLMODEL_TBL) WITH OVERVIEW;
-------------------------------------------------------------------------------------------------------------------------------
-- The following code fails to genrate PMML with PMML_EXPORT=1 but works fine with PMML_EXPORT=0 (with 30 independent variables)
--------------------------------------------------------------------------------------------------------------------------------
DROP TYPE PAL_LOGISTICR_DATA_T;
CREATE TYPE PAL_LOGISTICR_DATA_T AS TABLE("V_City" varchar(50), "V_Age" integer, "V_Gender" varchar(5),
"V_Race" varchar(50), "V_Company" varchar(50), "V_Transport" varchar(50), "V_Drive" varchar(5),
"V_Skills" varchar(50), "V_Interests" varchar(50), "V_Familiar_SFO" varchar(5), "V_Familiar_SCSJ" varchar(5),
"V_Interest_Lead" varchar(5), "V_Interest_Bay" varchar(5), "V_Comfort_Feat" varchar(5), "V_Comfort_Conv" varchar(5),
"V_Other_Lang" varchar(50), "V_Pref_Loc" varchar(50), "V_Hr_Past" varchar(10), "V_Dietary_Needs" varchar(50),
"V_Behind_Scenes" varchar(5),
"S_Start" integer, "S_Slot" varchar(20), "S_Duration" integer, "S_Location" varchar(50), "S_Area" varchar(50),
"S_Role" varchar(50),
"SV_Role_Matches_Interests" integer, "SV_Role_Matches_Skills" integer,
"P_Avail_Prop" double, "P_Work_Prop" double,
"Showup" integer);
DROP TABLE PAL_LOGISTICR_DATA_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_DATA_TBL like PAL_LOGISTICR_DATA_T;
INSERT INTO PAL_LOGISTICR_DATA_TBL VALUES ('Milpitas',29,'F','Asian','','Public_Transport','','CustSrv','Youth','Y','Y','Y','Y','Y','N','Korean','San Francisco','>200','NonVeg','N',17,'Afternoon',180,'San Jose','Station','CustSrv',0,1,0.48,0.76,1);
INSERT INTO PAL_LOGISTICR_DATA_TBL VALUES ('Campbell',49,'F','Unknown','Levi Strauss & Co.','Other','','Trport','Hosp','N','N','N','Y','Y','N','Punjabi','San Francisco','1-100','NonVeg','Y',16,'Afternoon',120,'Santa Clara','Airport','Hosp',1,0,0.78,0.74,0);
INSERT INTO PAL_LOGISTICR_DATA_TBL VALUES ('San Jose',30,'F','White','','Public_Transport','','Youth','SocMed','N','N','N','N','Y','N','German','San Jose','1-100','Veg','N',8,'Morning',150,'San Francisco','Airport','Mgmt',0,0,0.03,0.55,1);
--- insert some more data ----
DROP TYPE PAL_LOGISTICR_RESULT_T;
CREATE TYPE PAL_LOGISTICR_RESULT_T AS TABLE("Coefficient" varchar(50),"CoefficientValue" DOUBLE,"ZSCORE" DOUBLE, "Pr(>|Z|)" DOUBLE);
DROP TYPE PAL_CONTROL_T;
CREATE TYPE PAL_CONTROL_T AS TABLE("NAME" VARCHAR(50), "INTARGS" INTEGER, "DOUBLEARGS" DOUBLE, "STRINGARGS" VARCHAR(100));
DROP TYPE PAL_LOGISTICR_PMMLMODEL_T;
CREATE TYPE PAL_LOGISTICR_PMMLMODEL_T AS TABLE( "ID" INTEGER, "PMMLMODEL" VARCHAR(5000));
DROP table PAL_LOGISTICR_PDATA_TBL;
CREATE column table PAL_LOGISTICR_PDATA_TBL("POSITION" INTEGER, "SCHEMA_NAME" VARCHAR(100), "TYPE_NAME" VARCHAR(100), "PARAMETER_TYPE" VARCHAR(100));
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (1,'DM_PAL', 'PAL_LOGISTICR_DATA_T','IN');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (2,'DM_PAL', 'PAL_CONTROL_T','IN');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (3,'DM_PAL', 'PAL_LOGISTICR_RESULT_T','OUT');
INSERT INTO PAL_LOGISTICR_PDATA_TBL VALUES (4,'DM_PAL', 'PAL_LOGISTICR_PMMLMODEL_T','OUT');
CALL SYS.AFLLANG_WRAPPER_PROCEDURE_DROP('DM_PAL', 'PAL_LOGISTICR_PROC');
CALL SYS.AFLLANG_WRAPPER_PROCEDURE_CREATE('AFLPAL','LOGISTICREGRESSION', 'DM_PAL', 'PAL_LOGISTICR_PROC',PAL_LOGISTICR_PDATA_TBL);
DROP TABLE #PAL_CONTROL_TBL;
CREATE LOCAL TEMPORARY COLUMN TABLE #PAL_CONTROL_TBL ("NAME" VARCHAR(50), "INTARGS" INTEGER, "DOUBLEARGS" DOUBLE, "STRINGARGS" VARCHAR(100));
INSERT INTO #PAL_CONTROL_TBL VALUES ('VARIABLE_NUM',30,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('METHOD', 0, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('EXIT_THRESHOLD',null,0.000001,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('THREAD_NUMBER',8,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('MAX_ITERATION',1000,null,null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('CATEGORY_COL', 2, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('STAT_INF', 1, null, null);
INSERT INTO #PAL_CONTROL_TBL VALUES ('PMML_EXPORT', 1, null, null);
DROP TABLE PAL_LOGISTICR_RESULTS_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_RESULTS_TBL LIKE PAL_LOGISTICR_RESULT_T;
DROP TABLE PAL_LOGISTICR_PMMLMODEL_TBL;
CREATE COLUMN TABLE PAL_LOGISTICR_PMMLMODEL_TBL LIKE PAL_LOGISTICR_PMMLMODEL_T;
CALL DM_PAL.PAL_LOGISTICR_PROC(PAL_LOGISTICR_DATA_TBL, "#PAL_CONTROL_TBL", PAL_LOGISTICR_RESULTS_TBL, PAL_LOGISTICR_PMMLMODEL_TBL) WITH OVERVIEW;