Update app.py

app.py

@@ -710,24 +710,18 @@ Provide clear, actionable advice while emphasizing the need for professional med
     
     
     def _generate_response(self, query: str, context: str) -> str:
-    """Enhanced response generation using model.generate() to avoid DynamicCache errors"""
-    if self.llm is None or self.tokenizer is None:
-        return self._generate_fallback_response(query, context)
-
-    prompt = f"""{self.system_prompt}
-
-MEDICAL KNOWLEDGE CONTEXT:
-{context}
-
-PATIENT QUESTION: {query}
-
-RESPONSE (provide practical, Gaza-appropriate medical guidance):"""
-
-    try:
-        # Tokenize input and move to correct device
-        inputs = self.tokenizer(prompt, return_tensors="pt").to(self.llm.device)
+        """Enhanced response generation using model.generate() to avoid DynamicCache errors"""
+        if self.llm is None or self.tokenizer is None:
+            return self._generate_fallback_response(query, context)
+            prompt = f"""{self.system_prompt} 
+            MEDICAL KNOWLEDGE CONTEXT:
+            {context}
+            PATIENT QUESTION: {query}
+            RESPONSE (provide practical, Gaza-appropriate medical guidance):"""
+        
+        try:
+            inputs = self.tokenizer(prompt, return_tensors="pt").to(self.llm.device)
 
-        # Generate output
         outputs = self.llm.generate(
             **inputs,
             max_new_tokens=800,