TableGPT: Few-shot Table-to-Text Generation with Table Structure Reconstruction and Content Matching

Hi everyone, I came across a paper called ‘TableGPT: Few-shot Table-to-Text Generation
with Table Structure Reconstruction and Content Matching’. (https://aclanthology.org/2020.coling-main.179.pdf) This paper uses transforms and GPT-2, the creator has put his code online on github (GitHub - syw1996/TableGPT). There is no documentation for the code, is there anyone who has already worked with this code and knows how I can give a table as input to get a sentence back?

Thanks in advance and Kind regards Vito

team, it will be helpful if you consider this request.

Alguma evolução neste tópico? Estou interessado também, se eu tiver algum progresso posso publicar aqui.

Translated:

Any developments on this topic? I’m interested too, if I have any progress I can post it here.