Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
pandas
License:

Details of caseHOLD data

#4
by iRanadheer - opened

Thanks for this dataset, really cool and challenging. I was exploring the caseHOLD dataset, it took me a while to understand (?) about the data.

you have a column where the context is, with holding token (mask). another column is sequences list, and the third column is label, which is the index of the correct holding in the list of sequences. Am i assuming correct?

Sign up or log in to comment