Update app.py

app.py

@@ -66,7 +66,7 @@ def generate_paraphrase(comment, max_length=128):
     return paraphrase
 
 # RLHF Loop
-max_iterations = 5
+max_iterations = 2
 reward_threshold = 0.2  # Target for acceptable paraphrases (based on dataset range -0.25 to 0.24)
 results = []