Update app.py

app.py

@@ -1,15 +1,20 @@
 import gradio as gr
+from transformers import pipeline
 from transformers import AutoModelForQuestionAnswering, pipeline,AutoTokenizer
-import torch
-from optimum.onnxruntime import ORTModelForQuestionAnswering
+from optimum.intel import OVModelForQuestionAnswering
+model_checkpoint = "letrunglinh/qa_pnc"
+global model_convert,tokenizer
+tokenizer = AutoTokenizer.from_pretrained(model_checkpoint)
+model_convert = OVModelForQuestionAnswering.from_pretrained(model_checkpoint,export=True)
+model_convert.half()
+model_convert.compile()
+
 def question_answer(context, question):
-  AUTH_TOKEN = "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"
-  model_checkpoint = "letrunglinh/qa_pnc"
-  device = "cuda" if torch.cuda.is_available() else "cpu"
-  tokenizer = AutoTokenizer.from_pretrained(model_checkpoint, use_auth_token=AUTH_TOKEN)
-  model = ORTModelForQuestionAnswering.from_pretrained(model_checkpoint, from_transformers=True)
-  model = pipeline('question-answering', model=model,
-  tokenizer=tokenizer, use_auth_token=AUTH_TOKEN)
+  
+  model = pipeline('question-answering', model=model_convert,
+  tokenizer=tokenizer)
+
+
   to_predict = [
     {
         "question": question,