Merge pull request #717 from py-why/fix/avoid_retrain_estimator

Fix avoid retrain estimator on causal_model API

dowhy/causal_model.py

@@ -73,6 +73,7 @@ class CausalModel:
         self._proceed_when_unidentifiable = proceed_when_unidentifiable
         self._missing_nodes_as_confounders = missing_nodes_as_confounders
         self.logger = logging.getLogger(__name__)
+        self._estimator_cache = {}
 
         if graph is None:
             self.logger.warning("Causal Graph not provided. DoWhy will construct a graph based on data inputs.")
@@ -302,21 +303,26 @@ class CausalModel:
                 method_params = {}
 
             identified_estimand.set_identifier_method(identifier_name)
-            causal_estimator = causal_estimator_class(
-                self._data,
-                identified_estimand,
-                self._treatment,
-                self._outcome,  # names of treatment and outcome
-                control_value=control_value,
-                treatment_value=treatment_value,
-                test_significance=test_significance,
-                evaluate_effect_strength=evaluate_effect_strength,
-                confidence_intervals=confidence_intervals,
-                target_units=target_units,
-                effect_modifiers=effect_modifiers,
-                **method_params,
-                **extra_args,
-            )
+
+            if not fit_estimator and method_name in self._estimator_cache:
+                causal_estimator = self._estimator_cache[method_name]
+            else:
+                causal_estimator = causal_estimator_class(
+                    self._data,
+                    identified_estimand,
+                    self._treatment,
+                    self._outcome,  # names of treatment and outcome
+                    control_value=control_value,
+                    treatment_value=treatment_value,
+                    test_significance=test_significance,
+                    evaluate_effect_strength=evaluate_effect_strength,
+                    confidence_intervals=confidence_intervals,
+                    target_units=target_units,
+                    effect_modifiers=effect_modifiers,
+                    **method_params,
+                    **extra_args,
+                )
+                self._estimator_cache[method_name] = causal_estimator
 
         return estimate_effect(
             self._treatment,