diff --git a/code.py b/code.py
index 8c61c2c..644c61f 100644
--- a/code.py
+++ b/code.py
@@ -28,10 +28,17 @@ def calc_word_count(in_path, expected_path):
     word_counts = {'paranormal':defaultdict(int), 'skeptic': defaultdict(int)}
     with open(in_path,encoding='utf-8') as in_file, open(expected_path,encoding='utf-8')  as  expected_file:
         for line, exp in zip(in_file, expected_file):
-            class_ = exp.rstrip('\n').replace(' ','')
+            class_ = exp.rstrip('\n')
+                #.replace(' ','')
             text, timestamp = line.rstrip('\n').split('\t')
+            text = re.sub(r'\\n+', " ", text)
+            text = re.sub(r'http\S+', " ", text)
+            text = re.sub(r'\/[a-z]\/', " ", text)
+            text = re.sub(r'[^a-z]', " ", text)
+            text = re.sub(r'\s{2,}', " ", text)
             text = re.sub(r'(\s+|\\n)', ' ', text)
-            text = re.sub(r'(https?:|www)\S+(\s+|$)', ' URL ', text)
+            text = re.sub(r'\W\w{1,3}\W|\A\w{1,3}\W', " ", text)
+            text = re.sub(r'^\s', "", text)
             tokens = text.lower().split(' ')
             for token in tokens:
                 if class_ == 'P':
diff --git a/code_prediction.py b/code_prediction.py
index 2d22cfe..4ea86a8 100644
--- a/code_prediction.py
+++ b/code_prediction.py
@@ -11,13 +11,21 @@ paranomal_class_logprob, skeptic_class_logprob, word_logprobs = pickle_loaded
 #Niektórych słów nie bezie w zbiorze treningowym dev-0 i dev-A
 def prediction(input,output):
     output_file = open(output,'w')
+    pickle_load = pickle.load(open('naive_base_model.pkl', 'rb'))
+    paranormal_class_logprob, skeptic_class_logprob, word_logprob = pickle_load
     with open(input,encoding='utf-8') as in_file:
         for line in in_file:
             temp_paranormal_logprob = paranomal_class_logprob
             temp_skeptic_logprob = skeptic_class_logprob
             text, timestamp = line.rstrip('\n').split('\t')
+            text = re.sub(r'\\n+', " ", text)
+            text = re.sub(r'http\S+', " ", text)
+            text = re.sub(r'\/[a-z]\/', " ", text)
+            text = re.sub(r'[^a-z]', " ", text)
+            text = re.sub(r'\s{2,}', " ", text)
             text = re.sub(r'(\s+|\\n)', ' ', text)
-            text = re.sub(r'(https?:|www)\S+(\s+|$)', ' URL ', text)
+            text = re.sub(r'\W\w{1,3}\W|\A\w{1,3}\W', " ", text)
+            text = re.sub(r'^\s', "", text)
             tokens = text.lower().split(' ')
             for token in tokens:
                 if token not in word_logprobs['paranormal']: