distilbert-base-uncased-distilled-squad-ov-fp32

Un ejemplo de cómo hacer inferencia en este modelo:

from optimum.intel.openvino import OVModelForQuestionAnswering
from transformers import AutoTokenizer, pipeline

# model_id debería estar configurado a un directorio local o un modelo disponible en el hub de HuggingFace.
model_id = "helenai/distilbert-base-uncased-distilled-squad-ov-fp32"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = OVModelForQuestionAnswering.from_pretrained(model_id)
pipe = pipeline("question-answering", model=model, tokenizer=tokenizer)
result = pipe("What is OpenVINO?", "OpenVINO is a framework that accelerates deep learning inferencing")
print(result)