-
Notifications
You must be signed in to change notification settings - Fork 3.3k
Pull requests: huggingface/datasets
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Stop
Dataset.sort() emitting a pyarrow FutureWarning
#8471
opened Aug 12, 2026 by
shashvat-singham
Loading…
Keep the dtype of an empty column in the numpy formatter
#8470
opened Aug 12, 2026 by
shashvat-singham
Loading…
Implement
construct_from_string on PandasArrayExtensionDtype
#8468
opened Aug 12, 2026 by
shashvat-singham
Loading…
Raise a clear error for a deeply nested mesh file
#8466
opened Aug 12, 2026 by
shashvat-singham
Loading…
Make
PandasArrayExtensionDtype._metadata a tuple
#8464
opened Aug 12, 2026 by
shashvat-singham
Loading…
Add test for tokenizer stride with Dataset.map (issue #4769)
#8461
opened Aug 11, 2026 by
imjimit07
Loading…
Fix Dataset.to_json() truncating/corrupting non-millisecond temporal columns
#8459
opened Aug 10, 2026 by
aakif-ms
Loading…
Fix Dataset.map(batched=True, drop_last_batch=True) returning unchanged input when dataset is smaller than batch_size
#8456
opened Aug 9, 2026 by
mayuriphad
Loading…
close input stream in _in_memory_arrow_table_from_file to prevent file descriptor leaks
#8455
opened Aug 9, 2026 by
HassanRady
Loading…
Coerce ClassLabel.names to plain str to fix invalid YAML in README.md
#8454
opened Aug 9, 2026 by
RudrenduPaul
Loading…
Reject features with different columns in IterableDataset.cast
#8453
opened Aug 9, 2026 by
LeSingh1
Loading…
Use the documented default batch_size in the batched IterableDataset exports
#8451
opened Aug 9, 2026 by
LeSingh1
Loading…
Raise a clear error for values that cannot be encoded as an image
#8450
opened Aug 9, 2026 by
LeSingh1
Loading…
Reject an unknown column in IterableDataset.cast_column
#8448
opened Aug 9, 2026 by
LeSingh1
Loading…
Reject a non-positive batch_size in the Dataset export methods
#8446
opened Aug 9, 2026 by
LeSingh1
Loading…
Validate by_column and fix the error message in Dataset.batch
#8443
opened Aug 9, 2026 by
LeSingh1
Loading…
Accept numpy integers as train_size / test_size in train_test_split
#8442
opened Aug 9, 2026 by
LeSingh1
Loading…
Reject a non-positive buffer_size in IterableDataset.shuffle
#8441
opened Aug 9, 2026 by
LeSingh1
Loading…
Previous Next
ProTip!
no:milestone will show everything without a milestone.