fix pep8

Author: Florian Gardin

skrules/skope_rules.py

@@ -14,6 +14,7 @@
 
 INTEGER_TYPES = (numbers.Integral, np.integer)
 
+
 class SkopeRules(BaseEstimator):
     """ An easy-interpretable classifier optimizing simple logical rules.
 
@@ -158,7 +159,8 @@ def __init__(self,
         self.bootstrap = bootstrap
         self.bootstrap_features = bootstrap_features
         self.max_depth = max_depth
-        self.max_depths = max_depth if isinstance(max_depth, Iterable) else [max_depth]
+        self.max_depths = max_depth \
+            if isinstance(max_depth, Iterable) else [max_depth]
         self.max_depth_duplication = max_depth_duplication
         self.max_features = max_features
         self.min_samples_split = min_samples_split
@@ -361,7 +363,6 @@ def fit(self, X, y, sample_weight=None):
         # count representation of feature
         if self.max_depth_duplication is not None:
             self.rules_ = self.deduplicate(self.rules_)
-            # TODO : Factorize disjoints performing rules (ex : c0 > 0 and c1 > 1  , c0 > 0 and c1 <= 1)
         return self
 
     def predict(self, X):
@@ -592,10 +593,12 @@ def _eval_rule_perf(self, rule, X, y):
         return y_detected.mean(), float(true_pos) / pos
 
     def deduplicate(self, rules):
-        return [max(rules_set, key=self.f1_score) for rules_set in self._find_similar_rulesets(rules)]
+        return [max(rules_set, key=self.f1_score)
+                for rules_set in self._find_similar_rulesets(rules)]
 
     def _find_similar_rulesets(self, rules):
-        """Create clusters of rules using a decision tree based on the terms of the rules
+        """Create clusters of rules using a decision tree based
+        on the terms of the rules
 
         Parameters
         ----------
@@ -635,11 +638,11 @@ def split_with_best_feature(rules, depth, exceptions=[]):
                     rules_splitted[1].append(rule)
                 else:
                     rules_splitted[2].append(rule)
-
+            new_exceptions = exceptions+[most_represented_term]
             # Choose best term
             return [split_with_best_feature(ruleset,
                                             depth-1,
-                                            exceptions=exceptions+[most_represented_term])
+                                            exceptions=new_exceptions)
                     for ruleset in rules_splitted]
 
         def breadth_first_search(rules, leaves=None):
@@ -656,4 +659,5 @@ def breadth_first_search(rules, leaves=None):
         return leaves
 
     def f1_score(self, x):
-        return 2 * x[1][0] * x[1][1] / (x[1][0] + x[1][1]) if (x[1][0] + x[1][1]) > 0 else 0
+        return 2 * x[1][0] * x[1][1] / \
+               (x[1][0] + x[1][1]) if (x[1][0] + x[1][1]) > 0 else 0

skrules/tests/test_skope_rules.py

@@ -132,7 +132,7 @@ def test_skope_rules_works():
     rules_vote = clf.rules_vote(X_test)
     score_top_rules = clf.score_top_rules(X_test)
     pred = clf.predict(X_test)
-    pred_score_top_rules = clf.predict_top_rules(X_test,1)
+    pred_score_top_rules = clf.predict_top_rules(X_test, 1)
     # assert detect outliers:
     assert_greater(np.min(decision_func[-2:]), np.max(decision_func[:-2]))
     assert_greater(np.min(rules_vote[-2:]), np.max(rules_vote[:-2]))
@@ -141,6 +141,7 @@ def test_skope_rules_works():
     assert_array_equal(pred, 6 * [0] + 2 * [1])
     assert_array_equal(pred_score_top_rules, 6 * [0] + 2 * [1])
 
+
 def test_deduplication_works():
     # toy sample (the last two samples are outliers)
     X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1], [6, 3], [4, -7]]
@@ -154,7 +155,7 @@ def test_deduplication_works():
     rules_vote = clf.rules_vote(X_test)
     score_top_rules = clf.score_top_rules(X_test)
     pred = clf.predict(X_test)
-    pred_score_top_rules = clf.predict_top_rules(X_test,1)
+    pred_score_top_rules = clf.predict_top_rules(X_test, 1)
 
 
 def test_performances():