Fix dataset reading error
#1
by
vatolinalex
- opened
Deleted "Unnamed: 0" column to fix the dataset reading error
>>> ds = datasets.load_dataset("Vikhrmodels/habr_qa_sbs")
...
CastError: Couldn't cast
Unnamed: 0: int64
question: string
best: string
bad: string
-- schema metadata --
huggingface: '{"info": {"features": {"Unnamed: 0": {"dtype": "int64", "_t' + 161
to
{'question': Value(dtype='string', id=None), 'best': Value(dtype='string', id=None), 'bad': Value(dtype='string', id=None)}
because column names don't match
I tested the dataset reads after my fix, the error is no longer present
>>> ds = datasets.load_dataset("Vikhrmodels/habr_qa_sbs", revision="ea1067a55eb5191750b54a819d0ddc91777804d8")
>>> ds
DatasetDict({
train: Dataset({
features: ['question', 'best', 'bad'],
num_rows: 102558
})
})
vatolinalex
changed pull request status to
open
AlexWortega
changed pull request status to
merged