Merge pull request #43 from Iron-Stark/dectree

Added more options like min_samples_split,min_samples_leaf,max_features and max_nodes in Decision Tree scikit implementation.

methods/scikit/dtc.py

@@ -50,7 +50,16 @@ def __init__(self, dataset, timeout=0, verbose=True):
     self.criterion = 'gini'
     self.max_depth = None
     self.seed = 0
-
+    self.splitter = 'best'
+    self.max_depth = None
+    self.min_samples_split = 2
+    self.min_samples_leaf = 1
+    self.min_weight_fraction_leaf = 0.0
+    self.max_features = None
+    self.random_state = None
+    self.min_impurity_split = 1e-07
+    self.class_weight = None
+    self.presort = False
   '''
   Build the model for the Decision Tree Classifier.
 
@@ -62,7 +71,15 @@ def BuildModel(self, data, labels):
     # Create and train the classifier.
     dtc = DecisionTreeClassifier(criterion=self.criterion,
                                  max_depth=self.max_depth,
-                                 random_state=self.seed)
+                                 random_state=self.seed,
+                                 splitter = self.splitter,
+                                 min_samples_split = self.min_samples_split,                                 
+                                 min_weight_fraction_leaf=self.min_weight_fraction_leaf,
+                                 max_features = self.max_features,
+                                 max_leaf_nodes = self.max_leaf_nodes,
+                                 min_impurity_split = self.min_impurity_split,
+                                 class_weight = self.class_weight,
+                                 presort = self.presort)
     dtc.fit(data, labels)
     return dtc
 
@@ -85,9 +102,18 @@ def RunDTCScikit(q):
       c = re.search("-c (\s+)", options)
       d = re.search("-d (\s+)", options)
       s = re.search("-s (\d+)", options)
-
+      mss = re.search("--min_samples_split (\d+)", options)
+      msl = re.search("--min_samples_leaf (\d+)", options)
+      mf = re.search("--max_features (\d+)", options)
+      mln = re.search("--max_leaf_nodes (\d+)", options)
+
       self.criterion = 'gini' if not c else str(c.group(1))
       self.max_depth = None if not d else int(d.group(1))
+      self.splitter = 'best' if not s else str(s.group(1))
+      self.min_samples_split = 2 if not mss else int(mss.group(1))
+      self.min_samples_leaf = 1 if not msl else int(msl.group(1))
+      self.max_features = None if not mf else int(mf.group(1))
+      self.max_leaf_nodes = None if not mln else int(mln.group(1))
       self.seed = 0 if not s else int(s.group(1))
 
       try:
@@ -148,3 +174,4 @@ def RunMetrics(self, options):
       metrics['MSE'] = Metrics.SimpleMeanSquaredError(truelabels, predictedlabels)
 
     return metrics
+