Datasets:

Modalities:
Text
Formats:
text
Languages:
Serbian
Libraries:
Datasets
License:

Dataset loader does not work, and dataset's raw file has only 336 lines

#2
by nljubesi - opened

The dataset loader does not work, and when I download the dataset manually (file SrpKorNews.txt), it has only 336 lines. The file seems to have around 446 million words, but the segmentation into (many more than 336) documents seems not to be present.

Društvo za jezičke resurse i tehnologije org

True. I will look into it.

Btw. I was wrong, the dataset loads after pip install -U datasets, but loads the 336 entries only.

Društvo za jezičke resurse i tehnologije org

FIXED. Both here and for https://huggingface.co/datasets/procesaur/kisobran, where this corpus is also included :)

Works well now! Thanks!

nljubesi changed discussion status to closed

Sign up or log in to comment