Try passing model repo [deploy failing]

It's really strange: philschmid fork loads fine, and duplicating
philschmid fork in this repo doesn't.

handler.py

@@ -14,7 +14,7 @@ def mean_pooling(model_output, attention_mask):
 class EndpointHandler():
     def __init__(self, path=""):
         # load the optimized model
-        self.model = ORTModelForFeatureExtraction.from_pretrained(path, file_name="model-quantized.onnx")
+        self.model = ORTModelForFeatureExtraction.from_pretrained("jpohhhh/MiniLM-L6-v2-optimum-embeddings", file_name="model-quantized.onnx")
         self.tokenizer = AutoTokenizer.from_pretrained(path)
 
     def __call__(self, data: Any) -> List[List[float]]: