Update handler.py

handler.py

@@ -11,8 +11,8 @@ class EndpointHandler():
         self.device = 'cuda' if torch.cuda.is_available() else 'cpu'
         try:
             logger.info(f"Loading model and tokenizer from path: {path}")
-            self.model = AutoModelForSeq2SeqLM.from_pretrained(f"{path}/model_v2/").to(self.device)
-            self.tokenizer = AutoTokenizer.from_pretrained(f"{path}/model_v2/")
+            self.model = AutoModelForSeq2SeqLM.from_pretrained(f"{path}").to(self.device)
+            self.tokenizer = AutoTokenizer.from_pretrained(f"{path}")
         except Exception as e:
             logger.error(f"Error loading model or tokenizer from path {path}: {e}")
             # Handle error (e.g., exit or set model/tokenizer to None)