search

wenhuchen / LogicNLG

The data and code for ACL2020 paper "Logical Natural Language Generation from Open-Domain Tables"
MIT License
166 stars 22 forks source link

Question about predictions #16

Closed ixn872 closed 3 years ago

ixn872 commented 3 years ago

Thank you for this repository. I would like to ask you, what is the correct way to use the GPT-2 model version to generate summaries over new table.

wenhuchen commented 3 years ago

Could you encode your table as a JSON file (the same format as the ones in the repo)? Then maybe you revise the https://github.com/wenhuchen/LogicNLG/blob/4b8b4f10bbadf93ad22a69fd31862b2061761a3d/GPT2.py#L211 to specifically read your table file as the input.