---
inference: false
tags:
- onnx
- text-classification
- bert
- adapterhub:qa/boolq
- adapter-transformers
datasets:
- boolq
language:
- en
---

# ONNX export of Adapter `AdapterHub/bert-base-uncased-pf-boolq` for bert-base-uncased
## Conversion of [AdapterHub/bert-base-uncased-pf-boolq](https://huggingface.co/AdapterHub/bert-base-uncased-pf-boolq) for UKP SQuARE


## Usage
```python
onnx_path = hf_hub_download(repo_id='UKP-SQuARE/bert-base-uncased-pf-boolq-onnx', filename='model.onnx') # or model_quant.onnx for quantization
onnx_model = InferenceSession(onnx_path, providers=['CPUExecutionProvider'])

context = 'Property tax or 'house tax' is a local tax on buildings, along with appurtenant land. It is and imposed on the Possessor (not the custodian of property as per 1978, 44th amendment of constitution). It resembles the US-type wealth tax and differs from the excise-type UK rate. The tax power is vested in the states and is delegated to local bodies, specifying the valuation method, rate band, and collection procedures. The tax base is the annual rental value (ARV) or area-based rating. Owner-occupied and other properties not producing rent are assessed on cost and then converted into ARV by applying a percentage of cost, usually four percent. Vacant land is generally exempt. Central government properties are exempt. Instead a 'service charge' is permissible under executive order. Properties of foreign missions also enjoy tax exemption without requiring reciprocity. The tax is usually accompanied by service taxes, e.g., water tax, drainage tax, conservancy (sanitation) tax, lighting tax, all using the same tax base. The rate structure is flat on rural (panchayat) properties, but in the urban (municipal) areas it is mildly progressive with about 80% of assessments falling in the first two brackets.'
question = 'is house tax and property tax are same'
tokenizer = AutoTokenizer.from_pretrained('UKP-SQuARE/bert-base-uncased-pf-boolq-onnx')

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

## Architecture & Training

The training code for this adapter is available at https://github.com/adapter-hub/efficient-task-transfer.
In particular, training configurations for all tasks can be found [here](https://github.com/adapter-hub/efficient-task-transfer/tree/master/run_configs).


## Evaluation results

Refer to [the paper](https://arxiv.org/pdf/2104.08247) for more information on results.

## Citation

If you use this adapter, please cite our paper ["What to Pre-Train on? Efficient Intermediate Task Selection"](https://arxiv.org/pdf/2104.08247):

```bibtex
@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",
}
```