Merge pull request #5 from drcrook1/prk

inference code update & Readme enhancements

Project_One-Tests/test_ml_metrics.py

@@ -5,6 +5,11 @@
 from inference_code.model_class import MyModel
 from sklearn.metrics.regression import r2_score
 
+
+class Super_secret_data:
+    [x] = {""}
+    [y] = {""}
+
 class TestMLMetrics(object):
     """
     testing of the model
@@ -14,16 +19,24 @@ def setUp(self):
 
     def test_r2_within_business_value(self):
         m = MyModel()
+        print("initing Model")
         m.init()
 
         # Add code to load your super secret cross validation stuff
         # use secrets encoded in a variable group available in the ADO Pipeline
         # cuz scientists can't access that and game the system.
-        super_secret_data = None
 
-        y_predicted = m.predict(super_secret_data[x])
+        x1 = '{"age": 37, "hours-per-week": 40.0, "sex": "Female", "occupation" : "Exec-managerial"}'
+        y1 = [200000, 190000]
+        # We will now make the prediction and since we have a single sample, will create 2 points by adding and subtracting 10
+        # just for creating passable data to r2_score - note that this is a meamingless exercise just to demonsrtate the process
+        # In real case, you will have more meaningful tests and asserts
+
+        y_predicted = m.predict(x1)[0][0]
+        ypred = [y_predicted+10, y_predicted-10]
+        print (ypred)
+        r2score = r2_score(y1, ypred)
 
-        r2_score(super_secret_data[y], y_predicted)
+        assert(-2 < r2score < 2 )
 
-        assert(r2_score > 2 < r2_score)
 

Project_One/inference_code/model_class.py

@@ -20,6 +20,7 @@ def init(self):
         root_path = "./assets/"
         with open(root_path + "x_scaler.pkl", "rb") as xfile:
             self.x_scaler = pickle.load(xfile)
+            print (self.x_scaler)
         with open(root_path + "y_scaler.pkl", "rb") as yfile:
             self.y_scaler = pickle.load(yfile)
         with open(root_path + "model.pkl", "rb") as mfile:
@@ -29,10 +30,14 @@ def predict(self, input_package):
         """
         input_package: json formatted string of the form
         {"age": integer, "hours-per-week" : double, "sex" : string, "occupation" string}
-
+        
         returns json formatted string of the form: {"estimated_wages" : float}
         """
+        #input_package = '{"age": 37, "hours-per-week" : 40.0, "sex" : "Female", "occupation" : "Exec-managerial"}'
+
+
         x = transform_input(input_package)
+        print (x)
         x = self.x_scaler.transform(x)
         y = self.model.predict(x)
         y = self.y_scaler.inverse_transform(y)

Project_One/inference_code/utility.py

@@ -14,10 +14,12 @@ def transform_input(input_package):
     input_package: raw json input package as agreed upon
     returns: numpy array of correct format without pre-processing
     """
+    print ("loading json")
     d = json.loads(input_package)
+    print(d)
     # Add extra processing for some reason.
     x = np.array([d["age"], d["hours-per-week"]]).transpose()
-    return x
+    return x.reshape(-1,2)
 
 def transform_output(y):
     """