Merge pull request #23 from ngoix/fix_check_estimator

[MRG] Fix check estimator

Author: ngoix

doc/index.rst

@@ -25,7 +25,7 @@ estimator with unit tests, along with examples and benchmarks.
        auto_examples/index
        ...
 
-See the `README <https://github.com/skope-rules/skope-rules/blob/master/README.md>`_
+See the `README <https://github.com/scikit-learn-contrib/skope-rules/blob/master/README.md>`_
 for more information.
 
 

setup.py

@@ -17,7 +17,7 @@
 setup(name='skope-rules',
       version='1.0.0',
       description='Machine Learning with Interpretable Rules',
-      url='https://github.com/skope-rules/skope-rules',
+      url='https://github.com/scikit-learn-contrib/skope-rules',
       author='see AUTHORS.rst',
       license='BSD 3 clause',
       packages=find_packages(),

skrules/rule.py

@@ -1,15 +1,17 @@
 import re
 
+
 def replace_feature_name(rule, replace_dict):
     def replace(match):
         return replace_dict[match.group(0)]
 
     rule = re.sub('|'.join(r'\b%s\b' % re.escape(s) for s in replace_dict),
-           replace, rule)
+                  replace, rule)
     return rule
 
+
 class Rule:
-    """ An object modelizing a logical rule and add factorization methods.
+    """ An object modelling a logical rule and add factorization methods.
     It is used to simplify rules and deduplicate them.
 
     Parameters
@@ -66,4 +68,3 @@ def __repr__(self):
                 [feature, symbol, str(self.agg_dict[(feature, symbol)])])
                 for feature, symbol in sorted(self.agg_dict.keys())
                 ])
-

skrules/skope_rules.py

@@ -17,8 +17,9 @@
 INTEGER_TYPES = (numbers.Integral, np.integer)
 BASE_FEATURE_NAME = "__C__"
 
+
 class SkopeRules(BaseEstimator):
-    """ An easy-interpretable classifier optimizing simple logical rules.
+    """An easy-interpretable classifier optimizing simple logical rules.
 
     Parameters
     ----------
@@ -161,8 +162,6 @@ 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_depth_duplication = max_depth_duplication
         self.max_features = max_features
         self.min_samples_split = min_samples_split
@@ -251,7 +250,7 @@ def fit(self, X, y, sample_weight=None):
 
         # default columns names :
         feature_names_ = [BASE_FEATURE_NAME + x for x in
-                            np.arange(X.shape[1]).astype(str)]
+                          np.arange(X.shape[1]).astype(str)]
         if self.feature_names is not None:
             self.feature_dict_ = {BASE_FEATURE_NAME + str(i): feat
                                   for i, feat in enumerate(self.feature_names)}
@@ -263,7 +262,10 @@ def fit(self, X, y, sample_weight=None):
         clfs = []
         regs = []
 
-        for max_depth in self.max_depths:
+        self._max_depths = self.max_depth \
+            if isinstance(self.max_depth, Iterable) else [self.max_depth]
+
+        for max_depth in self._max_depths:
             bagging_clf = BaggingClassifier(
                 base_estimator=DecisionTreeClassifier(
                     max_depth=max_depth,
@@ -362,10 +364,6 @@ def fit(self, X, y, sample_weight=None):
             for rule in
             [Rule(r, args=args) for r, args in rules_]]
 
-
-
-
-
         # keep only rules verifying precision_min and recall_min:
         for rule, score in rules_:
             if score[0] >= self.precision_min and score[1] >= self.recall_min:
@@ -393,7 +391,7 @@ def fit(self, X, y, sample_weight=None):
 
         # Replace generic feature names by real feature names
         self.rules_ = [(replace_feature_name(rule, self.feature_dict_), perf)
-                  for rule, perf in self.rules_]
+                       for rule, perf in self.rules_]
 
         return self