Update myagent.py

myagent.py

@@ -63,12 +63,14 @@ class LocalLlamaModel:
     def generate(self, prompt: str, max_new_tokens=512, **kwargs):
         try:
             # Generate answer using the provided prompt - following the recommended pattern
-            input_ids = self.tokenizer.apply_chat_template(
-                [{"role": "user", "content": str(prompt)}],
-                add_generation_prompt=True,
-                return_tensors="pt",
-                tokenize=True,
-            ).to(self.model.device)
+            # input_ids = self.tokenizer.apply_chat_template(
+            #     [{"role": "user", "content": str(prompt)}],
+            #     add_generation_prompt=True,
+            #     return_tensors="pt",
+            #     tokenize=True,
+            # ).to(self.model.device)
+            
+            inputs = self.tokenizer(prompt, return_tensors="pt").to(self.model.device)
             
             # Generate output - exactly as in recommended code
             output = self.model.generate(