multi-languages in BEIR format
#2
by
d-rau
- opened
Hey, I only find the arabic subset for this dataset in beir format. Same holds for other beir format datasets that exist in other languages.
Hey, in the sub folders are all the different versions. You are right, the configurations are not correctly added. I will look into it soon.
This dataset is mostly designed for running it with the MTEB benchmark. This works correctly, since we implemented the loading mechanism on folder level.