Fixed nltk bug

summarizer.py

@@ -3,6 +3,7 @@ from transformers import BartTokenizer, TFBartForConditionalGeneration
 from Utils import get_input_chunks
 import networkx as nx
 from nltk.tokenize import sent_tokenize
+import nltk
 from sklearn.feature_extraction.text import TfidfVectorizer
 import community
 from title_generator import T5Summarizer
@@ -47,7 +48,11 @@ class BARTSummarizer:
     def preprocess_for_auto_chapters(self, text: str):
        
         # Tokenize the text into sentences
-        sentences = sent_tokenize(text)
+        try:
+            sentences = sent_tokenize(text)
+        except:
+            nltk.download('punkt')
+            sentences = sent_tokenize(text)
 
         # Filter out empty sentences and sentences with less than 5 words
         sentences = [sentence for sentence in sentences if len(sentence.strip()) > 0 and len(sentence.split(" ")) > 4]