diff --git a/postagger.py b/postagger.py
index 8483594a52b6c8ff3c8d1ae812fe7422b7544ab1..5bfc6af2af40a60767ab357c4128053cfc158d6e 100644
--- a/postagger.py
+++ b/postagger.py
@@ -119,6 +119,16 @@ def to_bag_of_words(corpus):
 #fun fact: len(bag_of_words) is 25325 for corpus.csv
 
 
+def extract(corpus_instance, pos_vocab):
+    "nimmt einzelnes dict, und gibt featurevector der größe len(bag-bigram) zurück"
+    pass
+
+
+def get_pos_vocabulary(corpus):
+    "geht über ganzes corpus, gibt bigram-bag zurück"
+    pass
+
+
 if __name__ == '__main__':
     corpus = read_corpus("minicorpus.csv")
     tagged_corpus = corpus_pos_tagger(corpus)
@@ -130,3 +140,4 @@ if __name__ == '__main__':
     
     for vector in corpus_vector:
         print(vector)
+   
diff --git a/training_testing.py b/training_testing.py
index 01af8fb15083597c56bf8d944eb43e663146281c..8d5625e2c5be86e78837d462fa190711d51573a4 100644
--- a/training_testing.py
+++ b/training_testing.py
@@ -5,6 +5,7 @@ import ngram_feature
 import numpy as np
 from sklearn import svm
 from sklearn.tree import DecisionTreeClassifier
+import postagger
 
 
 def create_vector(corpus_instance, vocabulary=None):
@@ -14,6 +15,7 @@ def create_vector(corpus_instance, vocabulary=None):
     single corpus instance)
     """
     f1 = ngram_feature.extract(corpus_instance, vocabulary)
+    # f2 = postagger.to_bigram_vector(corpus_instance, pos_vocab)
     f4 = sent_rating_feature.extract(corpus_instance)
 
     return np.concatenate((f1,f4))
@@ -31,6 +33,10 @@ if __name__ == '__main__':
     # vocabularies
     unigram_vocab = ngram_feature.get_vocabulary(train_set, 1)
     bigram_vocab = ngram_feature.get_vocabulary(train_set, 2)
+    
+    # pos_bags
+    # bigram_pos_vocab = postagger.get_pos_vocabulary(train_set) (entspricht corpus in postagger.py)
+    
 
     # inputs:
     train_inputs = [create_vector(el, unigram_vocab)