Upload handler.py

handler.py

@@ -6,7 +6,7 @@ from typing import Dict, List, Any
 class EndpointHandler():
     def __init__(self, path=""):
         self.model = AutoModelForCausalLM.from_pretrained(path,
-                                                          torch_dtype=torch.float16,
+                                                          torch_dtype=torch.bfloat16,
                                                           device_map="auto",
                                                           trust_remote_code=True)
         self.tokenizer = AutoTokenizer.from_pretrained(path)