--- a/feature_selection.xml	Tue May 22 19:31:59 2018 -0400
+++ b/feature_selection.xml	Wed May 30 08:25:49 2018 -0400
@@ -25,6 +25,8 @@
 
 @COLUMNS_FUNCTION@
 
+@FEATURE_SELECTOR_FUNCTION@
+
 input_json_path = sys.argv[1]
 params = json.load(open(input_json_path, "r"))
 
@@ -57,42 +59,10 @@
 y=y.ravel()
 
 ## Create feature selector
-selector = params["feature_selection_algorithms"]["selected_algorithm"]
-selector = getattr(sklearn.feature_selection, selector)
-options = params["feature_selection_algorithms"]["options"]
-
-if params['feature_selection_algorithms']['selected_algorithm'] == 'SelectFromModel':
-    if not options['threshold'] or options['threshold'] == 'None':
-        options['threshold'] = None
-        if 'extra_estimator' in params['feature_selection_algorithms'] and params['feature_selection_algorithms']['extra_estimator']['has_estimator'] == 'no_load':
-            fitted_estimator = pickle.load(open("params['feature_selection_algorithms']['extra_estimator']['fitted_estimator']", 'r'))
-            new_selector = selector(fitted_estimator, prefit=True, **options)
-        else:
-            estimator=params["feature_selection_algorithms"]["estimator"]
-            if params["feature_selection_algorithms"]["extra_estimator"]["has_estimator"]=='no':
-                estimator=params["feature_selection_algorithms"]["extra_estimator"]["new_estimator"]
-            estimator=eval(estimator.replace('__dq__', '"').replace("__sq__","'"))
-            new_selector = selector(estimator, **options)
-            new_selector.fit(X, y)
-
-elif params['feature_selection_algorithms']['selected_algorithm'] in ['RFE', 'RFECV']:
-    if 'scoring' in options and (not options['scoring'] or options['scoring'] == 'None'):
-        options['scoring'] = None
-    estimator=params["feature_selection_algorithms"]["estimator"]
-    if params["feature_selection_algorithms"]["extra_estimator"]["has_estimator"]=='no':
-        estimator=params["feature_selection_algorithms"]["extra_estimator"]["new_estimator"]
-    estimator=eval(estimator.replace('__dq__', '"').replace("__sq__","'"))
-    new_selector = selector(estimator, **options)
-    new_selector.fit(X, y)
-
-elif params['feature_selection_algorithms']['selected_algorithm'] == "VarianceThreshold":
-    new_selector = selector(**options)
-    new_selector.fit(X, y)
-
-else:
-    score_func = params["feature_selection_algorithms"]["score_func"]
-    score_func = getattr(sklearn.feature_selection, score_func)
-    new_selector = selector(score_func, **options)
+new_selector = feature_selector(params['feature_selection_algorithms'])
+if params['feature_selection_algorithms']['selected_algorithm'] != 'SelectFromModel' or \
+        'extra_estimator' not in params['feature_selection_algorithms'] or \
+        params['feature_selection_algorithms']['extra_estimator']['has_estimator'] != 'no_load' :
     new_selector.fit(X, y)
 
 ## Transform to select features