Update app.py

app.py

@@ -142,12 +142,7 @@ def cut_sent(para):
 
 def embed(document):
     model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
-    pattern = re.compile(r’[\u4e00-\u9fa5]+')
-    result = pattern.search(document) 
-    if result:
-        sentences = cut_sent(document)
-    else:
-        sentences = nltk.sent_tokenize(document)
+    sentences = cut_sent(document)
     embeddings = model.encode(sentences, convert_to_tensor=True)
     #Compute the pair-wise cosine similarities
     cos_scores = util.pytorch_cos_sim(embeddings, embeddings).cpu().numpy()