updated to model path in handler

handler.py

@@ -10,8 +10,8 @@ class EndpointHandler():
         # Preload all the elements you are going to need at inference.
         # pseudo:
         # self.model= load_model(path)
-        self.model = AutoModelForVision2Seq.from_pretrained("microsoft/kosmos-2-patch14-224").to("cuda")
-        self.processor = AutoProcessor.from_pretrained("microsoft/kosmos-2-patch14-224")
+        self.model = AutoModelForVision2Seq.from_pretrained(path).to("cuda")
+        self.processor = AutoProcessor.from_pretrained(path)
 
         # prompt = "<grounding>An image of"