This is an
tapas-base model, trained on the lookup queries of
wikisql dataset. It was trained to take tables and questions as input to extract answers from the table.
One can use this model to predict answers for natural language queries given a table. Biases associated with pre-training of tapas-base and wikisql dataset may be present.
One can use this model directly in the
PrimeQA framework as in this example
notebook.
1@misc{herzig2020tapas,
2 title={TAPAS: Weakly Supervised Table Parsing via Pre-training},
3 author={Jonathan Herzig and Paweł Krzysztof Nowak and Thomas Müller and Francesco Piccinno and Julian Martin Eisenschlos},
4 year={2020},
5 eprint={2004.02349},
6 archivePrefix={arXiv},
7 primaryClass={cs.IR}
8}
9
