flexibility for cpu or cuda ep

handler.py

@@ -11,15 +11,26 @@ if torch.backends.cudnn.is_available():
 
 class EndpointHandler():
     def __init__(self, path=""):
+
+        on_cuda = torch.cuda.is_available()
         # load the optimized model
+
+        provider = "CPUExecutionProvider"
+        if on_cuda:
+            provider = "CUDAExecutionProvider"
+        
         model = ORTModelForSequenceClassification.from_pretrained(
             path,
             export=False,
-            provider="CUDAExecutionProvider",
+            provider=provider,
         )
         tokenizer = AutoTokenizer.from_pretrained(path)
+        
+        device = -1
+        if on_cuda:
+            device = 0
         # create inference pipeline
-        self.pipeline = pipeline("text-classification", model=model, tokenizer=tokenizer, device=0)
+        self.pipeline = pipeline("text-classification", model=model, tokenizer=tokenizer, device=device)
 
 
     def __call__(self, data: Any) -> List[List[Dict[str, float]]]: