Split name and organization

#2
by n8rob - opened

Hello,

It seems the only "split" available for this dataset is "train", which is a bit confusing since none of this is actually train data. It appears that the split labeled as "train" is actually FLORES+ "devtest" with "dev" concatenated on the end. Could you make this clear in the documentation? Or even better, reorganize to have the different splits instead of a "train" split?

Sign up or log in to comment