Error loading the dataset
#5
by
vpkprasanna
- opened
from datasets import load_dataset
dataset = load_dataset("oscar", "unshuffled_original_ta")
Error
[{'expected': SplitInfo(name='train', num_bytes=9934879441, num_examples=1263280, shard_lengths=None, dataset_name=None), 'recorded': SplitInfo(name='train', num_bytes=9933590150, num_examples=1263180, shard_lengths=[59000, 64000, 61000, 63000, 71000, 68000, 63000, 60000, 64000, 67000, 67000, 64000, 64000, 64000, 63000, 67000, 64000, 61000, 66000, 43180], dataset_name='oscar')}]
how to resolve this ?
Hi! I fixed the size info in https://huggingface.co/datasets/oscar/discussions/6.
For future reference, you can also pass verification_mode="no_checks" to load_dataset in such scenarios to ignore the verification checks.
mariosasko
changed discussion status to
closed