Notebook and example asset updates for pzmm

Author: smlindauer

examples/data/hmeqModels/DecisionTreeClassifier/DecisionTreeClassifier.pickle

@@ -1,16 +1,12 @@
-
-
 import math
 import pickle
 import pandas as pd
 import numpy as np
 
 import settings
+from pathlib import Path
 
-
-_thisModelFit
-
-with open(settings.pickle_path + 'DecisionTreeClassifier.pickle', 'rb') as _pFile:
+with open(Path(settings.pickle_path) / 'DecisionTreeClassifier.pickle', 'rb') as _pFile:
     _thisModelFit = pickle.load(_pFile)
 
 def scoreDecisionTreeClassifier(LOAN, MORTDUE, VALUE, YOJ, DEROG, DELINQ, CLAGE, NINQ, CLNO, DEBTINC):
@@ -20,7 +16,7 @@ def scoreDecisionTreeClassifier(LOAN, MORTDUE, VALUE, YOJ, DEROG, DELINQ, CLAGE,
         global _thisModelFit
     except NameError:
 
-        with open(settings.pickle_path + 'DecisionTreeClassifier.pickle', 'rb') as _pFile:
+        with open(Path(settings.pickle_path) / 'DecisionTreeClassifier.pickle', 'rb') as _pFile:
             _thisModelFit = pickle.load(_pFile)
 
     try:
@@ -39,9 +35,9 @@ def scoreDecisionTreeClassifier(LOAN, MORTDUE, VALUE, YOJ, DEROG, DELINQ, CLAGE,
     try:
         EM_EVENTPROBABILITY = float(prediction)
     except TypeError:
-    # If the model expects non-binary responses, a TypeError will be raised.
-    # The except block shifts the prediction to accept a non-binary response.
-        EM_EVENTPROBABILITY = float(prediction[:,1])
+        # If the prediction returns as a list of values or improper value type, a TypeError will be raised.
+        # Attempt to handle the prediction output in the except block.
+        EM_EVENTPROBABILITY = float(prediction[0])
 
     if (EM_EVENTPROBABILITY >= 0.199496644295302):
         EM_CLASSIFICATION = '1'

examples/data/hmeqModels/DecisionTreeClassifier/DecisionTreeClassifier.zip

@@ -12,5 +12,5 @@
     "eventProbVar": "EM_EVENTPROBABILITY",
     "modeler": "sasdemo",
     "tool": "Python 3",
-    "toolVersion": "3.8.3"
+    "toolVersion": "3.8.16"
 }

examples/data/hmeqModels/DecisionTreeClassifier/DecisionTreeClassifierScore.py

@@ -216,12 +216,12 @@
                 "_RASE_": 0.0,
                 "_NObs_": 4172,
                 "_GINI_": 1.0,
-                "_GAMMA_": 2.880334878688322,
+                "_GAMMA_": 2.880334878656814,
                 "_formattedPartition_": "           1",
                 "_DataRole_": "TRAIN",
                 "_MCE_": 0.0,
                 "_ASE_": 0.0,
-                "_MCLL_": 9.992007221626413e-16,
+                "_MCLL_": 2.220446049250313e-16,
                 "_KS_": 1.0,
                 "_KSPostCutoff_": null,
                 "_DIV_": 4172,
@@ -243,7 +243,7 @@
                 "_DataRole_": "TEST",
                 "_MCE_": 0.1470917225950783,
                 "_ASE_": 0.1470917225950783,
-                "_MCLL_": 5.0803882403500396,
+                "_MCLL_": 5.301723065625175,
                 "_KS_": 0.34052961140098437,
                 "_KSPostCutoff_": null,
                 "_DIV_": 1788,

examples/data/hmeqModels/DecisionTreeClassifier/ModelProperties.json

@@ -1,14 +1,14 @@
 [
     {
         "name": "EM_EVENTPROBABILITY",
-        "length": 1,
-        "type": "string",
-        "level": "nominal"
-    },
-    {
-        "name": "EM_CLASSIFICATION",
         "length": 8,
         "type": "decimal",
         "level": "interval"
+    },
+    {
+        "name": "EM_CLASSIFICATION",
+        "length": 1,
+        "type": "string",
+        "level": "nominal"
     }
 ]

examples/data/hmeqModels/DecisionTreeClassifier/dmcas_fitstat.json

@@ -1,16 +1,12 @@
-
-
 import math
 import pickle
 import pandas as pd
 import numpy as np
 
 import settings
+from pathlib import Path
 
-
-_thisModelFit
-
-with open(settings.pickle_path + 'GradientBoosting.pickle', 'rb') as _pFile:
+with open(Path(settings.pickle_path) / 'GradientBoosting.pickle', 'rb') as _pFile:
     _thisModelFit = pickle.load(_pFile)
 
 def scoreGradientBoosting(LOAN, MORTDUE, VALUE, YOJ, DEROG, DELINQ, CLAGE, NINQ, CLNO, DEBTINC):
@@ -20,7 +16,7 @@ def scoreGradientBoosting(LOAN, MORTDUE, VALUE, YOJ, DEROG, DELINQ, CLAGE, NINQ,
         global _thisModelFit
     except NameError:
 
-        with open(settings.pickle_path + 'GradientBoosting.pickle', 'rb') as _pFile:
+        with open(Path(settings.pickle_path) / 'GradientBoosting.pickle', 'rb') as _pFile:
             _thisModelFit = pickle.load(_pFile)
 
     try:
@@ -39,9 +35,9 @@ def scoreGradientBoosting(LOAN, MORTDUE, VALUE, YOJ, DEROG, DELINQ, CLAGE, NINQ,
     try:
         EM_EVENTPROBABILITY = float(prediction)
     except TypeError:
-    # If the model expects non-binary responses, a TypeError will be raised.
-    # The except block shifts the prediction to accept a non-binary response.
-        EM_EVENTPROBABILITY = float(prediction[:,1])
+        # If the prediction returns as a list of values or improper value type, a TypeError will be raised.
+        # Attempt to handle the prediction output in the except block.
+        EM_EVENTPROBABILITY = float(prediction[0])
 
     if (EM_EVENTPROBABILITY >= 0.199496644295302):
         EM_CLASSIFICATION = '1'

examples/data/hmeqModels/DecisionTreeClassifier/outputVar.json

@@ -12,5 +12,5 @@
     "eventProbVar": "EM_EVENTPROBABILITY",
     "modeler": "sasdemo",
     "tool": "Python 3",
-    "toolVersion": "3.8.3"
+    "toolVersion": "3.8.16"
 }