Update model card

README.md

@@ -33,8 +33,8 @@ The linear modules **roberta.encoder.layer.16.output.dense**, **roberta.encoder.
 ### Load with Intel® Neural Compressor:
 
 ```python
-from neural_compressor.utils.load_huggingface import OptimizedModel
-int8_model = OptimizedModel.from_pretrained(
-    'Intel/roberta-base-squad2-int8-static',
-)
+from optimum.intel.neural_compressor import IncQuantizedModelForSequenceClassification
+
+model_id = "Intel/roberta-base-squad2-int8-static"
+int8_model = IncQuantizedModelForSequenceClassification.from_pretrained(model_id)
 ```