Fixed bugs with deleting nessisary variable.

src/classifiers/nearestneighbors.py

@@ -25,6 +25,7 @@ def main(options: list):
     data, labels = map(np.array,
                        zip(*[(FeatureVector(p, features).get(), p.user)
                              for p in samples]))
+    num_users = len(np.unique([s.user for s in samples]))
     avg, p = classify(data, labels, num_users, args)
     print("Overall Accuracy: %f" % avg)
     if args.p_value:

src/classifiers/randomforest.py

@@ -25,6 +25,7 @@ def main(options: list):
     data, labels = map(np.array,
                        zip(*[(FeatureVector(p, features).get(), p.user)
                              for p in samples]))
+    num_users = len(np.unique([s.user for s in samples]))
     avg, p = classify(data, labels, num_users, args)
     print("Overall Accuracy: %f" % avg)
     if args.p_value:
@@ -54,12 +55,12 @@ def parse_args(args: list):
 def classify(data, labels, num_users: int, args):
     s = np.arange(data.shape[0])
     np.random.shuffle(s)
-    res = randomForest(data[s], labels[s],
-                       n=args.folds, verbose=args.verbose, fn=args.criterion
+    res = random_forest(data[s], labels[s],
+                       n=args.folds, verbose=args.verbose, fn=args.criterion,
                        estimators=args.estimators)
     return (np.average(res), t_test(res, num_users)[1] / 2)
 
-def randomForest(data: list, labels: list, n=5, verbose=0, estimators=100,
+def random_forest(data: list, labels: list, n=5, verbose=0, estimators=100,
                  fn="gini"):
     from sklearn.ensemble import RandomForestClassifier
     folds = KFold(n_splits=n)