Add update_predictor_exp function and its pytest function

src/analyticsdf/analyticsdataframe.py

@@ -335,3 +335,28 @@ def update_response_poly_categorical(self, predictor_name: str = None, betas: di
                 self.response_vector = numeric_vals[predictor_name]
             else:
                 self.response_vector += numeric_vals[predictor_name]
+
+    def update_predictor_exp(self, rate: float = None, 
+                             predictor_name: list = None):
+        """Update a predictor with samples from an exponential distribution.
+
+        Args:
+            rate:
+                float, must be greater than 0.
+                Lambda(λ). The rate parameter of an exponential distribution.
+            predictor_name: 
+                String, a predictor name in AnalyticsDataframe object.
+
+        Raises:
+            KeyError: If the column does not exists.
+
+        """
+        ## Check if all target predictors exist
+        _check_columns_exist(self.predictor_matrix, predictor_name)
+
+        with set_random_state(validate_random_state(self.seed)):
+            scale = 1/rate      # scale(β) = 1/lambda(λ)
+            size = self.n
+            self.predictor_matrix[predictor_name] = np.random.exponential(scale, size)
+
+

tests/test_basic.py

@@ -258,3 +258,28 @@ def test_update_response_poly_categorical():
     ad.update_response_poly_categorical(predictor_name='X6', betas={'Red': -2000, 'Blue': -1700})
     assert ad.predictor_matrix.loc[1, 'X6'] == 'Red'
     assert ad.response_vector[1] < -1900
+
+def test_predictor_exp():
+    """Test function 'update_predictor_exp'
+    Test logic:
+        The mean of an exponential distribution is also the scale parameter
+        β = 1/λ
+    """
+    ## Initialize and use the function to update
+    ad = AnalyticsDataframe(10000, 3, ["xx1", "xx2", "xx3"], "yy", 666)
+    ad.update_predictor_exp(rate = 1.5,
+                            predictor_name = "xx1")
+    pred_matrix = ad.predictor_matrix
+
+    ## Set up groud truth for testing
+    round_decimals = 2
+    exp_mean = round(1/1.5, round_decimals)
+
+    ## Test if the mean of the distribution is 1/λ
+    assert round(np.mean(pred_matrix["xx1"]), round_decimals) == exp_mean
+
+    ## Test its error cases
+    # pred_exists error: predictor name doesn't exist
+    with pytest.raises(KeyError):
+        ad.update_predictor_exp(rate = 1.5,
+                                predictor_name = "nonexistent")