Empty test cases
#2
by
StDestiny
- opened
The output of the test set are empty.
Hi! The source dataset was the tau/scrolls dataset on huggingface which i believe purposefully drops all test examples. If pulling/comparing on the test set is relevant to your use case, I’d recommend directly using the source dataset: https://github.com/Yale-LILY/QMSum
pszemraj
changed discussion status to
closed
ok thanks