Datasets:

Helsinki-NLP

/

tatoeba_mt

like 54

Follow

Language Technology Research Group at the University of Helsinki 226

When downloading the German to English subset, I get the following error.

NonMatchingSplitsSizesError: [{
'expected': SplitInfo(name='validation', num_bytes=25137584, num_examples=289778, dataset_name='tatoeba_mt'),
'recorded': SplitInfo(name='validation', num_bytes=28610541, num_examples=289748, dataset_name='tatoeba_mt')
}]

Reproduce: load_dataset("Helsinki-NLP/tatoeba_mt", "deu-eng") on datasets==2.4.0