This repository is publicly accessible, but you have to accept the conditions to access its files and content.
Log in or Sign Up to review the conditions and access this dataset content.
This is a dataset for training the cross-encoder of our RAG system. It is a combination of the PIAF, FQuAD, SQuAD-French, and pandora-s-fr datasets.
