Update app.py

app.py

@@ -22,7 +22,7 @@ def predict(input, history=[]):
                             max_length=1000,
                             #max_length=200,
                             pad_token_id=tokenizer.eos_token_id,
-                            #no_repeat_ngram_size=3,
+                            no_repeat_ngram_size=3,
                             #do_sample=True,
                             #temperature = 0.8
                             top_p=0.92,
@@ -31,9 +31,12 @@ def predict(input, history=[]):
 
     # convert the tokens to text, and then split the responses into the right format
     response = tokenizer.decode(history[0]).split("<|endoftext|>")
+    
     response = [(response[i], response[i+1]) for i in range(0, len(response)-1, 2)]  # convert to tuples of list
-    print(response)
-    print(history)
+    
+    if len(response) > 5:
+        response.pop(0)
+            
     return response, history
 
 gr.Interface(fn=predict,