app.py

from transformers import AutoTokenizer, pipeline
from optimum.onnxruntime import ORTModelForQuestionAnswering
import gradio as gr
model = ORTModelForQuestionAnswering.from_pretrained("optimum/roberta-base-squad2")
tokenizer = AutoTokenizer.from_pretrained("deepset/roberta-base-squad2")

onnx_qa = pipeline("question-answering", model=model, tokenizer=tokenizer)

question = "What's my name?"
context = "My name is Philipp and I live in Nuremberg."
def get_answer(question):
    pred = onnx_qa(question, context)
    return pred


demo = gr.Blocks()

with demo():
    with gr.Row():
        question = gr.Textbox(label='Question', lines= 3)
        b1 = gr.Button('Get Answer')
        answer = gr.Textbox(label='Answer', lines=4)
    b1.click(fn = get_answer, inputs=question, outputs=answer)

demo.launch()