ONNX 导出版适配器 AdapterHub/bert-base-uncased-pf-quoref（适用于 bert-base-uncased）

将 AdapterHub/bert-base-uncased-pf-quoref 转换为 UKP SQuARE 格式

使用方法

onnx_path = hf_hub_download(repo_id='UKP-SQuARE/bert-base-uncased-pf-quoref-onnx', filename='model.onnx') # 或使用 model_quant.onnx 进行量化
onnx_model = InferenceSession(onnx_path, providers=['CPUExecutionProvider'])

context = 'ONNX is an open format to represent models. The benefits of using ONNX include interoperability of frameworks and hardware optimization.'
question = 'What are advantages of ONNX?'
tokenizer = AutoTokenizer.from_pretrained('UKP-SQuARE/bert-base-uncased-pf-quoref-onnx')

inputs = tokenizer(question, context, padding=True, truncation=True, return_tensors='np')
inputs_int64 = {key: np.array(inputs[key], dtype=np.int64) for key in inputs}
outputs = onnx_model.run(input_feed=dict(inputs_int64), output_names=None)

架构与训练

该适配器的训练代码可在 https://github.com/adapter-hub/efficient-task-transfer 获取。 具体而言，所有任务的训练配置可以在这里找到。

评估结果

有关结果的更多信息，请参阅论文。

引用

如果您使用此适配器，请引用我们的论文《What to Pre-Train on? Efficient Intermediate Task Selection》：

@inproceedings{poth-etal-2021-pre,
    title = "{W}hat to Pre-Train on? {E}fficient Intermediate Task Selection",
    author = {Poth, Clifton  and
      Pfeiffer, Jonas  and
      R{\"u}ckl{'e}, Andreas  and
      Gurevych, Iryna},
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2021",
    address = "Online and Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.emnlp-main.827",
    pages = "10585--10605",
}