remove fp16 load

handler.py

@@ -6,7 +6,7 @@ class EndpointHandler:
     def __init__(self, path="facebook/musicgen-large"):
         # load model and processor from path
         self.processor = AutoProcessor.from_pretrained(path)
-        self.model = MusicgenForConditionalGeneration.from_pretrained(path, torch_dtype=torch.float16).to("cuda")
+        self.model = MusicgenForConditionalGeneration.from_pretrained(path).to("cuda")
 
     def __call__(self, data: Dict[str, Any]) -> Dict[str, str]:
         """