ODSC 68406: AutoMLx Global Explainer (#1071)

Author: codeloop

ads/opctl/operator/lowcode/forecast/model/automlx.py

@@ -249,17 +249,18 @@ def _generate_report(self):
                 self.explain_model()
 
                 global_explanation_section = None
-                if self.spec.explanations_accuracy_mode != SpeedAccuracyMode.AUTOMLX:
-                    # Convert the global explanation data to a DataFrame
-                    global_explanation_df = pd.DataFrame(self.global_explanation)
 
-                    self.formatted_global_explanation = (
-                        global_explanation_df / global_explanation_df.sum(axis=0) * 100
-                    )
-                    self.formatted_global_explanation = self.formatted_global_explanation.rename(
-                        {self.spec.datetime_column.name: ForecastOutputColumns.DATE},
-                        axis=1,
-                    )
+                # Convert the global explanation data to a DataFrame
+                global_explanation_df = pd.DataFrame(self.global_explanation)
+
+                self.formatted_global_explanation = (
+                    global_explanation_df / global_explanation_df.sum(axis=0) * 100
+                )
+
+                self.formatted_global_explanation.rename(
+                    columns={self.spec.datetime_column.name: ForecastOutputColumns.DATE},
+                    inplace=True,
+                )
 
                 aggregate_local_explanations = pd.DataFrame()
                 for s_id, local_ex_df in self.local_explanation.items():
@@ -428,7 +429,9 @@ def explain_model(self):
                     # Use the MLExplainer class from AutoMLx to generate explanations
                     explainer = automlx.MLExplainer(
                         self.models[s_id]["model"],
-                        self.datasets.additional_data.get_data_for_series(series_id=s_id)
+                        self.datasets.additional_data.get_data_for_series(
+                            series_id=s_id
+                        )
                         .drop(self.spec.datetime_column.name, axis=1)
                         .head(-self.spec.horizon)
                         if self.spec.additional_data
@@ -463,6 +466,13 @@ def explain_model(self):
 
                     # Store the explanations in the local_explanation dictionary
                     self.local_explanation[s_id] = explanations_df
+
+                    self.global_explanation[s_id] = dict(
+                        zip(
+                            self.local_explanation[s_id].columns,
+                            np.nanmean((self.local_explanation[s_id]), axis=0),
+                        )
+                    )
                 else:
                     # Fall back to the default explanation generation method
                     super().explain_model()

ads/opctl/operator/lowcode/forecast/model/base_model.py

@@ -503,6 +503,7 @@ def _save_report(
                         f2.write(f1.read())
 
         # forecast csv report
+        # todo: add test data into forecast.csv
         # if self.spec.test_data is not None:
         #     test_data_dict = test_data.get_dict_by_series()
         #     for series_id, test_data_values in test_data_dict.items():
@@ -772,14 +773,6 @@ def explain_model(self):
                     logger.warn(
                         "No explanations generated. Ensure that additional data has been provided."
                     )
-                elif (
-                    self.spec.model == SupportedModels.AutoMLX
-                    and self.spec.explanations_accuracy_mode
-                    == SpeedAccuracyMode.AUTOMLX
-                ):
-                    logger.warning(
-                        "Global explanations not available for AutoMLX models with inherent explainability"
-                    )
                 else:
                     self.global_explanation[s_id] = dict(
                         zip(