Fixed issues #202

Author: dpuenteramirez

instance_selection/_RNN.py

@@ -48,8 +48,8 @@ def filter(self, samples, y):
         samples, y = cnn.filter(samples, y)
         samp_set = transform(samples, y)
 
-        data = deque([x for x in samp_set["data"]])
-        target = deque([x for x in samp_set["target"]])
+        data = deque(list(samp_set['data']))
+        target = deque(list(samp_set['target']))
 
         for instance in zip(samp_set.data, samp_set.target):
             (sample, class_sample) = instance

semisupervised/CoTraining.py

@@ -83,7 +83,11 @@ def __init__(
             else:
                 configs.append(GaussianNB())
 
-        self.h1, self.h2 = configs
+        try:
+            self.h1, self.h2 = configs
+        except ValueError:
+            raise AttributeError("Classifiers and/or params were not "
+                                 "correctly passed.")
 
     def fit(self, samples, y):
         """

semisupervised/DemocraticCoLearning.py

@@ -91,7 +91,11 @@ def __init__(
             else:
                 configs.append(default_classifiers[index]())
 
-        self.h1, self.h2, self.h3 = configs
+        try:
+            self.h1, self.h2, self.h3 = configs
+        except ValueError:
+            raise AttributeError("Classifiers and/or params were not "
+                                 "correctly passed.")
 
     def fit(self, samples, y):
         """

semisupervised/DensityPeaks.py

@@ -145,9 +145,25 @@ def __local_density(self):
         """
 
         def gauss_func(dij, dc):
+            """
+            > The function takes in a distance value and a cutoff value, and
+            returns the value of the Gaussian function at that point
+
+            :param dij: distance between two nodes
+            :param dc: The cutoff distance
+            :return: the value of the gaussian function.
+            """
             return math.exp(-((dij / dc) ** 2))
 
         def cutoff_func(dij, dc):
+            """
+            If the distance between two atoms is less than the cutoff distance,
+            return 1, otherwise return 0
+
+            :param dij: distance between atoms i and j
+            :param dc: cutoff distance
+            :return: 1 if dij < dc, else 0
+            """
             return 1 if dij < dc else 0
 
         func = gauss_func if self.gauss_cutoff else cutoff_func

semisupervised/TriTraining.py

@@ -84,7 +84,11 @@ def __init__(
             else:
                 configs.append(default_classifiers[index]())
 
-        self.hj, self.hk, self.hi = configs
+        try:
+            self.hj, self.hk, self.hi = configs
+        except ValueError:
+            raise AttributeError("Classifiers and/or params were not "
+                                 "correctly passed.")
 
         self.random_state = (
             random_state