Cannot be loaded correctly
#1
by ZJaume - opened
The current dataset structure cannot be loaded correctly
>>> from datasets import load_dataset
>>> d = load_dataset('proxectonos/SciELO-GL')
>>> d
DatasetDict({
train: Dataset({
features: ['text'],
num_rows: 1111822
})
})
>>> d['train'][1]
{'text': 'Tuberculosis continues to be a threat worldwide.'}
all texts are loaded together without any distinction of sentence pairs whtasoever.
Luasantamaria changed discussion status to closed
Luasantamaria changed discussion status to open
Hi, thanks for the heads up. I've updated the dataset to a new version and included a jsonl file that is now linked as train split in hugginface's dataviewer.
Until now the dataset was split into txt files instead of using jsonl, which explains the error.
imdbo changed discussion status to closed