id stringlengths 18 24 | number int64 2 8.48k | title stringlengths 1 290 | body stringlengths 0 228k | state stringclasses 2
values | createdAt timestamp[s]date 2020-04-14 18:18:51 2026-08-13 03:02:44 | updatedAt timestamp[s]date 2020-04-29 09:23:05 2026-08-15 04:47:12 | closedAt timestamp[s]date 2020-04-29 09:23:05 2026-07-28 10:29:56 ⌀ | url stringlengths 48 51 | author dict | labels dict | comments listlengths 0 30 |
|---|---|---|---|---|---|---|---|---|---|---|---|
I_kwDODunzps8AAAABMjWDaQ | 8,475 | `Dataset.select` raises a raw OverflowError on in-range negative indices that its own validation allows | ### Describe the bug
`Dataset.select` validates indices with [`_check_valid_indices_value`](https://github.com/huggingface/datasets/blob/836b82e0544cabf6474b25ade131b4d21e570373/src/datasets/arrow_dataset.py#L636-L638), which implements Python-style negative indexing: it only raises for a negative index when `index + ... | OPEN | 2026-08-13T03:02:44 | 2026-08-13T03:02:44 | null | https://github.com/huggingface/datasets/issues/8475 | {
"login": "codeAnqiang-ma"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABMjUQjw | 8,473 | Negative percent split boundary silently anchors to the start when it rounds to 0 (`train[-1%:]` returns the whole split) | ### Describe the bug
With the default `closest` rounding, a negative percent split boundary whose absolute value rounds to 0 loses its sign, so "relative to the end" silently becomes "relative to the start":
- `train[-1%:]` on a 30-example split returns **all 30 rows** instead of the last 1% (~0 rows, as `train[99%:]... | OPEN | 2026-08-13T02:56:52 | 2026-08-13T02:56:52 | null | https://github.com/huggingface/datasets/issues/8473 | {
"login": "codeAnqiang-ma"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABMa9c9w | 8,472 | batch(0) silently returns one batch with the whole dataset instead of raising | ### Describe the bug
`Dataset.batch()` and `IterableDataset.batch()` accept `batch_size=0` and negative sizes, and quietly return **one batch holding the whole dataset**. Asking for a zero-sized batch produces the largest possible batch, which is as close to the opposite of the request as the API can get.
### Steps o... | OPEN | 2026-08-12T07:52:28 | 2026-08-13T16:29:01 | null | https://github.com/huggingface/datasets/issues/8472 | {
"login": "shashvat-singham"
} | {
"nodes": []
} | [
"Hi @shashvat-singham i did also try and reproduced it and got the same results described :\n\n```text\nIterableDataset.batch(0/-1) → [5]\nDataset.batch(0/-1) → 1 batch\nbatch(2) → [2, 2, 1] / 3 batches\n```\n\nI’d like to give this a try\n\nThanks\n\n#self-assign"
] |
I_kwDODunzps8AAAABMa1yqQ | 8,469 | Empty column loses its dtype under numpy format (always float32) | ### Describe the bug
Under `numpy` format, a column that holds zero rows comes back as `float32` no matter what dtype it was declared with. The same column keeps its dtype as soon as it holds a single row, and the `pandas` and `arrow` formatters keep it in both cases.
This is easy to hit in normal code: any `filter` ... | OPEN | 2026-08-12T07:37:13 | 2026-08-12T07:38:05 | null | https://github.com/huggingface/datasets/issues/8469 | {
"login": "shashvat-singham"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABMau9nQ | 8,467 | PandasArrayExtensionDtype has no construct_from_string, so comparing the dtype to any string raises AssertionError | ### Describe the bug
`PandasArrayExtensionDtype` does not implement `construct_from_string`, so it inherits the base `ExtensionDtype` version, which opens with:
```python
@classmethod
def construct_from_string(cls, string):
assert isinstance(cls.name, str), (cls, type(cls.name))
```
`name` is an instance `@prope... | OPEN | 2026-08-12T07:24:51 | 2026-08-12T07:25:58 | null | https://github.com/huggingface/datasets/issues/8467 | {
"login": "shashvat-singham"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABMaaOWg | 8,465 | PandasArrayExtensionArray.take coerces fill_value before checking whether anything is filled, breaking boolean masks on integer array columns | ### Describe the bug
`PandasArrayExtensionArray.take` coerces `fill_value` to the array's value type **before** checking whether any element will actually be filled:
https://github.com/huggingface/datasets/blob/main/src/datasets/features/features.py#L965-L972
```python
if allow_fill:
fill_value = (
self.... | OPEN | 2026-08-12T06:38:18 | 2026-08-12T06:38:37 | null | https://github.com/huggingface/datasets/issues/8465 | {
"login": "shashvat-singham"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABMZe_1A | 8,463 | very bad quality images in most of them | ### Describe the bug
very bad quality images in most of them
### Steps to reproduce the bug
.
### Expected behavior
.
### Environment info
. | OPEN | 2026-08-12T03:49:23 | 2026-08-12T03:49:23 | null | https://github.com/huggingface/datasets/issues/8463 | {
"login": "anis-mselmi"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABMJnM8A | 8,458 | Replace httpx with httpx2 | ### Feature request
Replace the httpx library with the [httpx2](https://github.com/pydantic/httpx2) fork, maintained by Pydantic.
### Motivation
Httpx has not been released since December 2024, and has closed all public access to Issues / Discussions / PRs, this raises questions about the actual state of maintenance... | OPEN | 2026-08-10T13:01:44 | 2026-08-15T04:47:12 | null | https://github.com/huggingface/datasets/issues/8458 | {
"login": "Aegdesil"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"Hi @lhoestq can I work on this",
"I think if we would have to discuss in a https://github.com/huggingface/huggingface_hub/ issue about it and `datasets` could align with what's decided there"
] |
I_kwDODunzps8AAAABLwbj6Q | 8,410 | _legacy_no_dict_keys_sorting has no reader since #7817, so the 2.15.0 legacy cache lookup computes the wrong config_id | ### Describe the bug
`Pickler._legacy_no_dict_keys_sorting` has a definition and a writer, but since #7817 it has no reader, so the flag does nothing.
`DatasetBuilder._check_legacy_cache2` patches it on to recompute the `config_id` that 2.15.0 would have produced, because 2.15.0 hashed dicts without sorting their ite... | OPEN | 2026-08-06T18:40:14 | 2026-08-06T18:40:14 | null | https://github.com/huggingface/datasets/issues/8410 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABLaT_hA | 8,394 | IterableDataset.repeat() drops the Arrow fast path and widens numpy dtypes | ### Describe the bug
`RepeatExamplesIterable` is the only wrapper in `iterable_dataset.py` that does not forward `iter_arrow`, `is_typed` and `features` to the iterable it wraps. Two things follow from that: `.repeat()` drops the Arrow fast path, and under `numpy` format it silently widens the declared dtypes.
`SkipE... | OPEN | 2026-08-04T13:01:08 | 2026-08-04T13:01:08 | null | https://github.com/huggingface/datasets/issues/8394 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABLXe4fQ | 8,393 | CPU Bottleneck in `IterableDataset.state_dict()` when shuffling multiple shards | **Describe the bug**
There is a significant CPU bottleneck in `IterableDataset` dataloading when `buffer_size` and `max_buffer_input_shards` are greater than 1 (i.e. highly parallelized streaming with a shuffle buffer).
Currently, `_BaseExamplesIterable.state_dict()` falls back to performing a full recursive `copy.dee... | OPEN | 2026-08-04T07:16:30 | 2026-08-04T07:16:30 | null | https://github.com/huggingface/datasets/issues/8393 | {
"login": "yuxin00j"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABLW1gtA | 8,391 | Nanosecond timestamps are silently truncated even when Value("timestamp[ns]") is requested | ### Describe the bug
`Dataset.from_dict` / `from_list` / `map` silently drop sub-microsecond precision, even when the user explicitly declares a nanosecond feature type. The resulting dataset reports `timestamp[ns]` but the data has already been truncated to microseconds.
### Steps to reproduce the bug
```python
imp... | OPEN | 2026-08-04T05:36:19 | 2026-08-06T02:08:27 | null | https://github.com/huggingface/datasets/issues/8391 | {
"login": "LeSingh1"
} | {
"nodes": []
} | [
"#self-assign"
] |
I_kwDODunzps8AAAABLW1aHQ | 8,390 | to_json cannot round-trip temporal columns: all timestamps written as epoch milliseconds | ### Describe the bug
`Dataset.to_json()` serializes every temporal column as a bare epoch integer at a single fixed resolution (pandas' `date_unit="ms"` default), with no unit recorded in the output. Timestamps therefore do not survive a `to_json` → `from_json` round trip: they either come back silently wrong or raise... | OPEN | 2026-08-04T05:36:00 | 2026-08-10T23:10:37 | null | https://github.com/huggingface/datasets/issues/8390 | {
"login": "LeSingh1"
} | {
"nodes": []
} | [
"#self-assign\n\nOpened a fix in #8459."
] |
I_kwDODunzps8AAAABLWGFhg | 8,386 | map(batched=True, drop_last_batch=True) returns the input unchanged when the dataset is smaller than batch_size | ### Describe the bug
When `drop_last_batch=True` and the dataset is smaller than `batch_size`, `Dataset.map(batched=True, ...)` returns the **input unchanged** instead of an empty dataset. The mapped function is never called, but the original rows are passed through with the original (unbatched) schema.
The iterable ... | OPEN | 2026-08-04T03:21:59 | 2026-08-09T22:06:42 | null | https://github.com/huggingface/datasets/issues/8386 | {
"login": "LeSingh1"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABLS-nkg | 8,381 | IterableDataset.to_dict() and to_polars() return an empty generator instead of the data | ### Describe the bug
`IterableDataset.to_dict()` and `IterableDataset.to_polars()` each have a bare `yield` in their `batched=True` branch, which makes the whole method a generator function. Calling either one the way its docstring shows returns a generator that yields nothing, and the value the `else` branch returns ... | OPEN | 2026-08-03T18:35:13 | 2026-08-03T18:35:25 | null | https://github.com/huggingface/datasets/issues/8381 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [
"#take"
] |
I_kwDODunzps8AAAABK4yh9Q | 8,375 | `PandasArrayExtensionDtype._metadata` should be a tuple, not a string | ### Describe the bug
There's a small typo in `PandasArrayExtensionDtype` (`src/datasets/features/features.py`):
```python
_metadata = "value_type"
```
Pandas expects `_metadata` to be a tuple of attribute names, like `("value_type",)`, not a plain string ([API documentation ref](https://pandas.pydata.org/docs/refer... | OPEN | 2026-07-31T01:50:02 | 2026-08-12T06:30:29 | null | https://github.com/huggingface/datasets/issues/8375 | {
"login": "kohankhaki"
} | {
"nodes": []
} | [
"#self-assign"
] |
I_kwDODunzps8AAAABKqkD8g | 8,373 | TypeError in _dill.py _batch_setitems with Python 3.14 | ### Describe the bug
When using `load_dataset` on Python 3.14, a `TypeError` is raised during the fingerprinting process inside the bundled `_dill.py`. This happens because Python 3.14 changed the signature of the internal C-level `_Pickler._batch_setitems` to require 3 arguments (`self, items, obj`), while `datasets/... | OPEN | 2026-07-29T15:00:19 | 2026-08-01T19:30:07 | null | https://github.com/huggingface/datasets/issues/8373 | {
"login": "Vinello28"
} | {
"nodes": []
} | [
"Hey Mates, this works: In src/datasets/utils/_dill.py, starting around line 71, update the _batch_setitems overriding to pass the obj argument downstream.\n\n def _batch_setitems(self, items, obj=None):\n if self._legacy_no_dict_keys_sorting:\n return super()._batch_setitems(items,... |
I_kwDODunzps8AAAABKJdn3g | 8,365 | to_json/to_csv/to_sql cast nullable integer columns to float, corrupting values past 2**53 | ### Describe the bug
`Dataset.to_json`, `to_csv`, and `to_sql` write a nullable integer column as floating point rather than integers. All three writers (`io/json.py`, `io/csv.py`, `io/sql.py`) call `batch.to_pandas()` with pyarrow's default `integer_object_nulls=False`, which casts any integer column containing a nul... | CLOSED | 2026-07-25T13:38:01 | 2026-07-28T10:29:56 | 2026-07-28T10:29:56 | https://github.com/huggingface/datasets/issues/8365 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABJ_L-6w | 8,362 | ArrayXD fixed-shape integer column with a null row is silently cast to float on the python read path | ### Describe the bug
A fixed-shape `ArrayXD` column (for example `Array2D(shape=(1, 2), dtype="int64")`) that contains at least one `None` row is silently cast to float on the python read path. `to_dict()` / `to_pylist()` return every value as a float, and integers above 2**53 lose precision, purely because an unrelat... | CLOSED | 2026-07-24T04:32:35 | 2026-07-24T11:59:13 | 2026-07-24T11:59:13 | https://github.com/huggingface/datasets/issues/8362 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABJ-7a3Q | 8,359 | IterableDataset.filter() checkpoint/resume silently drops the rest of the in-flight Arrow table | ### Describe the bug
Resuming a streaming `IterableDataset.filter()` from a `state_dict()` checkpoint silently drops rows: the remainder of the in-flight Arrow table is skipped. For a single-shard dataset this loses up to a full row group per checkpoint, and for a single-table shard it can drop every unconsumed row.
... | CLOSED | 2026-07-24T03:36:32 | 2026-07-24T12:06:21 | 2026-07-24T12:06:21 | https://github.com/huggingface/datasets/issues/8359 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [
"#self-assign"
] |
I_kwDODunzps8AAAABJ33lPg | 8,351 | with_format('numpy') promotes flat float64/int columns with nulls to dtype=object (float32 stays numeric) | ### Describe the bug
`NumpyArrowExtractor._arrow_array_to_numpy` (used by `Dataset.with_format("numpy")`) promotes a flat, homogeneous numeric column to `dtype=object` when the column contains nulls, but only for `float64`/`int64` columns. A `float32` column is returned as a numeric array. The result is a silent wrong... | CLOSED | 2026-07-23T09:41:06 | 2026-07-23T17:32:25 | 2026-07-23T17:32:25 | https://github.com/huggingface/datasets/issues/8351 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABJU4jRg | 8,343 | Dataset.to_pandas() returns raw JSON strings for Json() columns instead of decoded objects | ### Describe the bug
`Dataset.to_pandas()` returns the raw Arrow storage strings for `Json()` feature columns, while every other in-memory export path decodes them to Python objects. `to_dict()`, `to_list()`, `to_json()` and `with_format("pandas")` all run `get_json_field_paths_from_feature` + `json_decode_field`, but... | CLOSED | 2026-07-19T01:44:53 | 2026-07-24T13:19:08 | 2026-07-24T13:19:08 | https://github.com/huggingface/datasets/issues/8343 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABJR6Akw | 8,341 | concatenate_datasets(axis=1) with Arrow format drops the first source's columns | ### Describe the bug
The Arrow fast path of horizontal concatenation (`concatenate_datasets([...], axis=1)` consumed through `.with_format("arrow")`, or any Arrow/tensor consumer) drops the columns of the first source. The plain-Python path is correct; only the Arrow path diverges.
`HorizontallyConcatenatedMultiSourc... | CLOSED | 2026-07-18T09:41:43 | 2026-07-22T18:35:31 | 2026-07-22T18:35:31 | https://github.com/huggingface/datasets/issues/8341 | {
"login": "ebarkhordar"
} | {
"nodes": []
} | [
"#self-assign"
] |
I_kwDODunzps8AAAABIxjmTA | 8,331 | datasets-server.huggingface.co returning 503 on every endpoint (whole host down, not one route) | ### Describe the bug
The entire datasets-server.huggingface.co backend (the dataset-viewer API) is
returning 503 Service Temporarily Unavailable on every route I've tried, not
just one endpoint. It reproduces on public datasets with no auth token, so it
doesn't appear to be account- or dataset-specific. huggingface.co... | CLOSED | 2026-07-14T13:59:29 | 2026-07-17T12:09:35 | 2026-07-17T12:09:34 | https://github.com/huggingface/datasets/issues/8331 | {
"login": "ChristianDenniss"
} | {
"nodes": []
} | [
"Hi ! Thanks for the ping, yes we are aware of the issue and actively working on it. It's taking some time though since we want to make sure it will be 100% back and for good.\n\nThis impacts both the API and the Viewer on HF which is based on it",
"Seems to be up and running properly now I'll conclude this issue... |
I_kwDODunzps8AAAABIxG3Ww | 8,330 | Dataset Studio and Viewer down | ### Describe the bug
Hello!
I want to report a bug on the datasets viewer and data studio pages. I have attempted to access it on various datasets from various devices and it will not let me view the dataset this way. I have tried restarting, login in and out but to no success. I use Hugging Face as a repository th... | CLOSED | 2026-07-14T13:06:53 | 2026-07-24T15:26:15 | 2026-07-24T15:26:15 | https://github.com/huggingface/datasets/issues/8330 | {
"login": "Taryn-Cail"
} | {
"nodes": []
} | [
"Hi ! Thanks for the ping, yes we are aware of the issue and actively working on it. It's taking some time though since we want to make sure it will be 100% back and for good"
] |
I_kwDODunzps8AAAABIvhOLA | 8,328 | Streaming datasets fail with 403 SignatureError: invalid key pair id from xet-bridge-us (Colab, latest libraries) | ### Describe the bug
I'm unable to stream datasets from the Hugging Face Hub. The issue affects both public datasets (e.g. beans) and gated datasets (ILSVRC/imagenet-1k).
The failure occurs before any examples are yielded. The request is redirected to the Xet CDN (xet-bridge-us) and fails with:
```
403 Forbidden
Au... | CLOSED | 2026-07-14T09:21:43 | 2026-07-24T15:26:24 | 2026-07-24T15:26:24 | https://github.com/huggingface/datasets/issues/8328 | {
"login": "mahmoud-mannes"
} | {
"nodes": []
} | [
"Hi ! I couldn't reproduce the issue on my side. I can see two options, let me know if any helps you:\n\n1) this was a transient issue (we restarted some machines recently but this shouldn't have impacted you tbh), can you try again just in case ?\n2) something is wrong with your token smh, can you logout and login... |
I_kwDODunzps8AAAABIoqexg | 8,327 | `Dataset.from_pandas` allocates ~1 MB per Arrow chunk, Out of Memory errors | ### Describe the bug
`Dataset.from_pandas` memory usage scales with the chunk count of the backing Arrow array (around 1 MB per chunk).
`Dataset.shuffle()` followed by `to_pandas()` produces string columns with one chunk per row, so a 25k-row DataFrame holding ~32 MB of text allocates ~30 GB on the way back through ... | CLOSED | 2026-07-13T14:13:38 | 2026-07-22T17:28:51 | 2026-07-22T17:28:51 | https://github.com/huggingface/datasets/issues/8327 | {
"login": "nenadom"
} | {
"nodes": []
} | [
"The measurements in the issue point at `from_pandas`, but I think the allocation is actually in fingerprinting, and not in the Arrow conversion at all.\n\nOn datasets 5.0.0 I get:\n\n| step | peak |\n|---|---|\n| `pa.Table.from_pandas(df)` | ~0 MB |\n| `Dataset.from_pandas(df)` | 1499 MB |\n| `generate_fingerprint... |
I_kwDODunzps8AAAABIiJwaQ | 8,324 | Path traversal / arbitrary file read via the `file_name` metadata field in `*folder` dataset builders (no `trust_remote_code`) | ### Describe the bug
reported via email on 9 June 2026 - no acknowledgement or response:
Package: datasets (PyPI). Affected file: src/datasets/packaged_modules/folder_based_builder/folder_based_builder.py
Affected Versions: confirmed on 5.0.0 (PyPI) and HEAD 63e103e.
CVSS: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N... | CLOSED | 2026-07-12T15:38:21 | 2026-07-23T15:37:57 | 2026-07-23T15:37:57 | https://github.com/huggingface/datasets/issues/8324 | {
"login": "geo-chen"
} | {
"nodes": []
} | [
"#self-assign"
] |
I_kwDODunzps8AAAABHy6FNw | 8,308 | Streaming IterableDataset state stops advancing after first resume — second resume restarts from the beginning |
## Description
For a streaming `IterableDataset`, calling `load_state_dict()` and then continuing to iterate leaves `state_dict()` frozen at the initial position (`shard_idx=0`, `shard_example_idx=0`). Reading resumes correctly the first time, but the tracked state never advances again.
This causes a second consecut... | CLOSED | 2026-07-06T10:12:27 | 2026-07-24T07:58:49 | 2026-07-23T13:12:33 | https://github.com/huggingface/datasets/issues/8308 | {
"login": "francesco-bertolotti"
} | {
"nodes": []
} | [
"@francesco-bertolotti Hi maintainers! I am an NLP intern at Sunbird AI and a first-time contributor. I would love to work on this issue. Could you please assign it to me?",
"Thank you for your interest @SsemuliJoseph \n\nThe fix was merged 2 weeks ago https://github.com/huggingface/datasets/pull/8295\nI should p... |
I_kwDODunzps8AAAABHmFpKA | 8,302 | Turkish DIK-style Syntactic Structure Benchmark Dataset | ### Feature request
A structured dataset proposal introducing a Turkish syntactic structure benchmark designed for evaluating LLMs on fine-grained grammatical reasoning in a morphologically rich language.
Current NLP datasets in Hugging Face focus heavily on surface-level semantic tasks or English-centric benchmarks.... | OPEN | 2026-07-03T15:33:39 | 2026-08-09T12:19:58 | null | https://github.com/huggingface/datasets/issues/8302 | {
"login": "nane100503-gif"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"Hi! I'd be interested in helping with this benchmark. I have experience working with NLP datasets and Hugging Face tooling, and I'd be happy to contribute on the engineering side (dataset packaging, loading script, validation, documentation, evaluation utilities, or testing). If you're planning to move forward, I'... |
I_kwDODunzps8AAAABHOI25g | 8,296 | [security] Symlink-Following Arbitrary File Write via Archive Extraction in huggingface/datasets | ### Describe the bug
## Summary
A symlink-following vulnerability exists in the archive extraction subsystem of `huggingface/datasets` that allows a local attacker with write access to a shared HuggingFace cache directory to write arbitrary files with the victim's privileges, by pre-planting a symlink at a predictabl... | OPEN | 2026-06-30T18:46:34 | 2026-07-04T19:06:53 | null | https://github.com/huggingface/datasets/issues/8296 | {
"login": "AAtomical"
} | {
"nodes": []
} | [
"thanks for reporting, I'm not entirely sure how someone could exploit this but a fix would be welcome for sure ! Option A looks good to me",
"@lhoestq Thanks for the quick response! I opened a PR #8303 implementing Option A. The threat model is a shared cache. Any user who can write into downloads/extracted/ — b... |
I_kwDODunzps8AAAABHI57og | 8,293 | Support concatenating multiple streaming datasets while preserving the sum of shards | ### Feature request
It would be useful if concatenating multiple `IterableDataset`s could preserve the shard information from each input dataset, so that:
- the total number of shards equals the sum of the input shards;
- `split_dataset_by_node()` and worker sharding continue to work efficiently;
- downstream distrib... | OPEN | 2026-06-30T05:44:23 | 2026-07-03T13:40:43 | null | https://github.com/huggingface/datasets/issues/8293 | {
"login": "muyihao"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"it's already the case afaik:\n\n```python\nIn [1]: from datasets import IterableDataset, concatenate_datasets\n\nIn [2]: ds1 = IterableDataset.from_dict({\"i\": [0, 1]})\n\nIn [3]: ds2 = IterableDataset.from_dict({\"i\": [2, 3]})\n\nIn [4]: ds = concatenate_datasets([ds1, ds2])\n\nIn [5]: list(ds)\nOut[5]: [{'i': ... |
I_kwDODunzps8AAAABFaF9XA | 8,269 | MetadataConfigs drops parquet shards when exported config rows are non-consecutive | ## Bug
`_from_exported_parquet_files_and_dataset_infos()` uses `itertools.groupby()` on unsorted `exported_parquet_files`.
If the same `config` appears again after another config, the earlier shard URLs are lost.
## Repro
```python
from itertools import groupby
from operator import itemgetter
exported = [
{"co... | OPEN | 2026-06-14T04:07:09 | 2026-06-14T04:07:09 | null | https://github.com/huggingface/datasets/issues/8269 | {
"login": "KirtiRamchandani"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABFREIKQ | 8,261 | Feature Proposal: Add FunASR Chinese/Multilingual ASR Benchmark Datasets | <!-- funasr-ops:accuracy-note-20260714 -->
> [!NOTE]
> **License and capability clarification (2026-07-14):** FunASR is a toolkit, not a single checkpoint. The [FunASR](https://github.com/modelscope/FunASR#license) and [SenseVoice](https://github.com/FunAudioLLM/SenseVoice#license) repository source code is MIT; model ... | OPEN | 2026-06-12T10:23:14 | 2026-07-14T05:32:24 | null | https://github.com/huggingface/datasets/issues/8261 | {
"login": "LauraGPT"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABE9PTZA | 8,256 | Dataset Viewer fails on TSFile datasets | ### Describe the bug
## Description
The Dataset Viewer fails when trying to display a dataset stored in TSFile format.
The error shown by the viewer is:
```python
ModuleNotFoundError: No module named 'tsfile'
### Steps to reproduce the bug
# Dataset Viewer fails to load TSFile dataset due to missing `tsfile` depe... | OPEN | 2026-06-10T03:15:32 | 2026-07-06T01:52:51 | null | https://github.com/huggingface/datasets/issues/8256 | {
"login": "gengziyand"
} | {
"nodes": []
} | [
"Thanks for reporting ! I'm working on it :)\n\nThe viewer doesn't use `datasets` V5 yet, I have to do some minor adjustments first",
"I opened a PR that addresses the missing optional tsfile dependency error by replacing the raw ModuleNotFoundError with a clearer ImportError and adding a regression test.\n\nThis... |
I_kwDODunzps8AAAABE0gLAw | 8,253 | Feature request: Add a streaming_shard operator for early sample-level sharding when file-level sharding is insufficient | ## Feature request: Add a streaming_shard operator for early sample-level sharding when file-level sharding is insufficient
### Proposed API
It would be useful to provide a first-class streaming sharding operator in `datasets`, for example:
```python
dataset = datasets.load_dataset(..., streaming=True, split="train"... | OPEN | 2026-06-09T03:14:46 | 2026-06-29T13:00:36 | null | https://github.com/huggingface/datasets/issues/8253 | {
"login": "muyihao"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"I feel like IterableDatasetShard is not well suited for this and shouldn't be used / should be improved. Instead there is `ds.shard()` or even `split_dataset_by_node()` which works when `num_physical_files < global_world_size`. WDYT ?",
"I checked current `main`, and the requested behavior seems to sit between t... |
I_kwDODunzps8AAAABEprXCA | 8,247 | VORTEXRAG benchmarked on 6 HuggingFace datasets — new open-source RAG framework | Hi HuggingFace Datasets team!
**VORTEXRAG** was evaluated on 6 datasets available on the HuggingFace Hub using datasets.load_dataset():
- NQ, TriviaQA, WebQ, PopQA, HotpotQA, 2WikiMultiHopQA
Results: EM 74.8, F1 82.6 averaged across all 6. Full evaluation code is in the repo.
- GitHub: https://github.com/vignesh202... | OPEN | 2026-06-07T11:14:50 | 2026-06-07T11:14:50 | null | https://github.com/huggingface/datasets/issues/8247 | {
"login": "vignesh2027"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABEpG_1w | 8,246 | RecursionError when loading GLB file with deeply nested JSON in Mesh feature | ## Description
Loading a GLB file with deeply nested JSON chunk causes a RecursionError, crashing the application.
## Step to Reproduce
```python
from datasets import Dataset, Mesh
import tempfile, os, struct
def make_malicious_glb(depth=5000):
json_chunk = (b'{"a":' * depth + b'1' + b'}' * depth)
magic = b... | OPEN | 2026-06-07T07:10:22 | 2026-08-12T06:46:48 | null | https://github.com/huggingface/datasets/issues/8246 | {
"login": "BeyondBirthday07"
} | {
"nodes": []
} | [
"#self-assign"
] |
I_kwDODunzps8AAAABEiAVMA | 8,242 | "eval_strategy": "no" perform evaluation | ### Describe the bug
Hi,
I'm trying to train a diffusion model and I getting stuck with
```
strace -p 967873
strace: Process 967873 attached
ioctl(9, _IOC(_IOC_READ|_IOC_WRITE, 0x46, 0x2a, 0x20), 0x7fff18579d90) = 0
ioctl(9, _IOC(_IOC_READ|_IOC_WRITE, 0x46, 0x2a, 0x20), 0x7fff18579d90) = 0
```
But my issue is that ... | OPEN | 2026-06-05T18:08:17 | 2026-06-20T14:48:43 | null | https://github.com/huggingface/datasets/issues/8242 | {
"login": "SamuelLarkin"
} | {
"nodes": []
} | [
"Wrong repo, this belongs in axolotl, not `datasets`. And `eval_strategy: \"no\"` was never going to stop it.\n\nWhat hangs you is the diffusion plugin's sample generation. `DiffusionGenerationCallback.on_step_end` fires on `global_step % generation_interval == 0` and ignores `eval_strategy`. When you have no eval ... |
I_kwDODunzps8AAAABEgwy_Q | 8,241 | load_dataset("json", ...) infers a different schema when the JSONL file starts with a UTF-8 BOM | ### Describe the bug
When a JSONL file starts with a UTF-8 BOM (`\xef\xbb\xbf`), `datasets.load_dataset("json", data_files=...)` infers a different schema than when the same file is loaded without the BOM. Specifically, the loader's "mixed-struct-types" pre-scan — which normally promotes columns with heterogeneous nes... | CLOSED | 2026-06-05T14:39:46 | 2026-06-09T15:48:27 | 2026-06-09T15:48:27 | https://github.com/huggingface/datasets/issues/8241 | {
"login": "f-salvetti"
} | {
"nodes": []
} | [
"#self-assign"
] |
I_kwDODunzps8AAAABDFJofw | 8,217 | Multi-GPU Training with Large Dataset — NCCL Timeouts / Slow Data Loading + Excessive Memory Usage with keep_in_memory=True | Hi, I'm training a large dataset using Lerobot (which relies on Hugging Face Datasets) and encountering severe performance issues when scaling to multi-GPU training.
### Problem Description
- Unstable/slow data loading + NCCL timeouts
- Step time fluctuates drastically: ~1s/step (fast) vs ~45s/step (extremely slow)
... | CLOSED | 2026-05-22T09:55:13 | 2026-06-03T02:12:43 | 2026-06-03T02:12:43 | https://github.com/huggingface/datasets/issues/8217 | {
"login": "D222097"
} | {
"nodes": []
} | [
"Hi @D222097,\n\nWe have audited the architectural footprint of your multi-GPU training setup. The behaviors you are experiencing are not bugs within the `datasets` library, but rather fundamental limitations of Python's process architecture interacting with hardware topologies. \n\nHere is the clinical breakdown o... |
I_kwDODunzps8AAAABCdZ3ew | 8,204 | Please enable Private Vulnerability Reporting | Hi datasets team,
I have a security finding in datasets that I would like to disclose privately. GitHub Private Vulnerability Reporting is currently disabled on this repo.
Could you turn it on? It's at Settings > Security > Code security and analysis > Private vulnerability reporting. Once enabled I will file the ful... | OPEN | 2026-05-16T13:28:01 | 2026-05-16T13:28:01 | null | https://github.com/huggingface/datasets/issues/8204 | {
"login": "0xBassia"
} | {
"nodes": []
} | [] |
I_kwDODunzps8AAAABCZ9hOA | 8,201 | Encountered an error when loading data in WebDataset format using `load_datasets` during multi-matchines training. | ### Describe the bug
Single-machine training works fine, but multi-machine training throws up all sorts of weird bugs. Help me!
### Steps to reproduce the bug
train_dataset = load_dataset(
"webdataset", data_files=args.train_dataset, split="train", streaming=True,cache_dir='/dev/shm/.cache'
)
args.train... | OPEN | 2026-05-15T19:21:24 | 2026-05-20T12:58:57 | null | https://github.com/huggingface/datasets/issues/8201 | {
"login": "aihao2000"
} | {
"nodes": []
} | [
"Hi @aihao2000 ! I was looking into this multi-machine training scenario and just opened a PR (#8203 ) that safely skips feature inference for empty shards during distributed streaming.\n\nplease! Check it out and see if it looks like it will resolve the weird bugs you were hitting!",
"it looks like there less da... |
I_kwDODunzps8AAAABCZXobQ | 8,199 | Replace AutoFeatureExtractor with AutoImageProcessor in image preprocessing docs | ## Summary
The `docs/source/use_dataset.mdx` documentation uses `AutoFeatureExtractor` from `transformers` in its image preprocessing example. For vision models, `AutoImageProcessor` is now the recommended API and better reflects current `transformers` conventions.
## Problem
In the [[Apply data augmentations](https... | CLOSED | 2026-05-15T17:34:53 | 2026-07-22T17:35:56 | 2026-07-22T17:35:56 | https://github.com/huggingface/datasets/issues/8199 | {
"login": "ajaystar8"
} | {
"nodes": []
} | [
"I'd like to work on this. I'll open a PR shortly."
] |
I_kwDODunzps8AAAABCV-Bqw | 8,197 | Feature Request / Discussion: Skip origin_metadata fetch when streaming=True? | Hi team!
I've been looking into the initialization process of the `DatasetBuilder` and noticed that the ETag/origin metadata caching path is executed unconditionally.
While fetching the `origin_metadata` and calculating the ETag is necessary for safely managing a local disk cache when using `as_dataset()`, it seems t... | OPEN | 2026-05-15T07:43:57 | 2026-06-17T03:39:14 | null | https://github.com/huggingface/datasets/issues/8197 | {
"login": "yuxin00j"
} | {
"nodes": []
} | [
"I think it's great that it checks that every file does exist by default. It's relatively cheap since it uses fsspec which caches file info at folder level.\n\nOn the other hand why not add skip_origin_metadata=False from @TheDerbiedOne by default to DatasetBuilder and document what it does (load_dataset passes kwa... |
I_kwDODunzps8AAAABBhmhxw | 8,178 | The full dataset viewer is not available (click to read why). Only showing a preview of the rows. Job manager crashed while running this job (missing heartbeats). Error code: JobManagerCrashedError | ### Describe the bug
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
Job manager crashed while running this job (missing heartbeats).
Error code: JobManagerCrashedError
see: https://huggingface.co/datasets/Genius-Society/tt100k/viewer
### Steps to reproduce the bug... | OPEN | 2026-05-07T08:33:48 | 2026-05-31T00:09:38 | null | https://github.com/huggingface/datasets/issues/8178 | {
"login": "kakamond"
} | {
"nodes": []
} | [
"Oh yeah, we hit something pretty similar when running a dataset viewer on a large collection (~3M rows) a few months back. The JobManagerCrashedError with missing heartbeats usually showed up for us when the server process ran out of memory or couldn’t keep up with the load spikes during dataset streaming. Turns o... |
I_kwDODunzps8AAAABBOHifA | 8,171 | Dataset UI shows 0:00/0:00 for audio after push_to_hub | I’m seeing an issue where audio previews in the dataset viewer show `0:00/0:00` after uploading with push_to_hub.
## What I did
I tried to avoid `torchcodec` because I want to keep the packages small using the following code:
```python
import io
import os
import shutil
import tempfile
from pathlib import Path
imp... | CLOSED | 2026-05-04T13:17:25 | 2026-05-20T10:46:58 | 2026-05-20T10:46:58 | https://github.com/huggingface/datasets/issues/8171 | {
"login": "MNIKIEMA"
} | {
"nodes": []
} | [
"Hello,\nI am trying to find my first issue to resolve. I can't reproduce your problem. When I go in the hub UI to your dataset. I can see the length of each audio. It takes some time to load \n\n## Dataset Card\n<img width=\"584\" height=\"644\" alt=\"Image\" src=\"https://github.com/user-attachments/assets/8b0e38... |
I_kwDODunzps8AAAABBEPO0g | 8,169 | Streaming dataset hangs consistently | ### Describe the bug
When loading a dataset with `streaming=True`, some background process prevents the script from ever returning (or at least for a very long time).
This did not happen with `huggingface-hub < 1`.
### Steps to reproduce the bug
from datasets import load_dataset
ds = load_dataset("IRIIS-RESEARCH/N... | CLOSED | 2026-05-01T20:27:58 | 2026-05-06T16:07:04 | 2026-05-06T16:06:08 | https://github.com/huggingface/datasets/issues/8169 | {
"login": "michaelpginn"
} | {
"nodes": []
} | [
"+1 also encountered",
"I pushed a workaround for the current version of PyArrow (24.0.0) and older versions at https://github.com/huggingface/datasets/pull/8176\n\nFor future versions it should be fixed directly in PyArrow"
] |
I_kwDODunzps8AAAABA62Vtw | 8,167 | [BUG] Abnormal progress bar in dataset.map when load_from_cache_file=False | ### Describe the bug
datasets 4.8.5
```python
from datasets import load_dataset, load_from_disk
data = load_from_disk('pretrain_cached_dataset/train')
def dummy_map(example):
return example
dataset = data.map(dummy_map)
dataset = data.map(dummy_map, load_from_cache_file=True) ... | CLOSED | 2026-04-30T08:47:53 | 2026-05-29T10:59:07 | 2026-05-29T10:59:07 | https://github.com/huggingface/datasets/issues/8167 | {
"login": "Jintao-Huang"
} | {
"nodes": []
} | [
"https://github.com/huggingface/datasets/blob/4e3bf52ff436d841e61211dfa096b5c47197ae36/src/datasets/arrow_dataset.py#L3593-L3598",
"Hi, can you provide more details on the issue ?",
"<img width=\"340\" height=\"26\" alt=\"Image\" src=\"https://github.com/user-attachments/assets/0dca6f47-4a8d-4871-a908-f58df856e... |
I_kwDODunzps8AAAABA5ItqQ | 8,165 | IterableDataset: corrupted EXIF image silently terminates streaming iterator instead of skipping the sample | ## Summary
When iterating over a streaming `IterableDataset` that contains images with corrupted EXIF metadata (e.g., a TIFF rational tag with denominator=0), PIL raises `ZeroDivisionError` inside `exif_transpose`. This exception propagates through the HuggingFace `datasets` pipeline and **terminates the streaming ite... | OPEN | 2026-04-30T02:20:54 | 2026-07-13T15:57:08 | null | https://github.com/huggingface/datasets/issues/8165 | {
"login": "LIUYellowBlack"
} | {
"nodes": []
} | [
"Hi ! Would it be acceptable to return None instead of skipping ? This way we can provide the same API for non-streaming (since the length of the dataset is known in advance and fixed). Then a subsequent .filter call can take care of skipping the samples. Something like this maybe ?\n\n```python\nds = load_dataset(... |
I_kwDODunzps8AAAABA5FcPQ | 8,164 | Lance: HF auth token dropped for private repos in streaming mode (broken `storage_options` lookup) | ### Describe the bug
When `Lance._split_generators` receives `hf://` URIs - i.e. anywhere a `StreamingDownloadManager` is in play (`load_dataset(..., streaming=True)`, `get_dataset_split_names`, and the dataset-viewer worker) - the HF auth token is silently dropped before reaching `lance.dataset(...)`. Lance then atte... | CLOSED | 2026-04-30T02:03:33 | 2026-04-30T15:38:43 | 2026-04-30T15:38:43 | https://github.com/huggingface/datasets/issues/8164 | {
"login": "ericjaebeom"
} | {
"nodes": []
} | [
"I can see the fix. Changing `split(\"://\", 0)` to `split(\"://\", 1)` and removing the `+ \"://\"` suffix. The key lookup should be `\"hf\"` not `\"hf://\"`. \nI'd like to submit a PR for this. Is this still open?",
"Well, I am already writing a PR for this, and it is just about to be submitted."
] |
I_kwDODunzps8AAAABAwiLsw | 8,159 | tests/conftest.py, tests/_test_patching.py, tests/fixtures/fsspec.py and other files are missing from the PYPI distribution | ### Describe the bug
Version: 4.8.5
### Steps to reproduce the bug
n/a
### Expected behavior
n/a
### Environment info
n/a | OPEN | 2026-04-28T20:01:05 | 2026-04-30T14:40:23 | null | https://github.com/huggingface/datasets/issues/8159 | {
"login": "yurivict"
} | {
"nodes": []
} | [
"Hey, ran into something similar a while back when we were packaging a custom dataset loader for a client at v4.7.x. Turns out, PyPI distributions sometimes skip test files and fixtures like `conftest.py` since they're not critical for runtime—honestly, it's a packaging config thing in `setup.py` or `pyproject.toml... |
I_kwDODunzps8AAAABAUM1HQ | 8,149 | [Optimization] Prevent per-thread instantiation of Cloud Storage FileSystem during Data loading initialization | ### Feature request
Modify the dataset loading initialization, so that fsspec filesystem instances (like GCSFileSystem or S3FileSystem) are instantiated once in the main thread and explicitly passed down to the background threads.
By pre-instantiating the fs object in the main thread (where the directory cache from g... | OPEN | 2026-04-23T13:09:19 | 2026-04-24T14:14:59 | null | https://github.com/huggingface/datasets/issues/8149 | {
"login": "ankitaluthra1"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"Is it open for contribution?",
"I think arbitrary filesystems would be recreated in every thread in any case, unlike HfFileSystem which has custom code to handle this."
] |
I_kwDODunzps8AAAABAK-nPw | 8,145 | `.map()` on a streaming IterableDataset silently skips examples after load_state_dict | ### Describe the bug
After `ds.map(fn)` on a streaming IterableDataset, saving state_dict() and loading it into a fresh dataset does not resume at the correct position. It jumps forward by up to one Arrow read-chunk (~1000 rows).
### Steps to reproduce the bug
```python
from datasets import Dataset
ds1 = Dataset.fr... | CLOSED | 2026-04-22T03:08:39 | 2026-04-27T14:03:28 | 2026-04-27T14:03:28 | https://github.com/huggingface/datasets/issues/8145 | {
"login": "MonliH"
} | {
"nodes": []
} | [
"I think this line here is causing issues: \nhttps://github.com/huggingface/datasets/blob/2724a653e7582077be647d3fdfe8cd1fac27b85f/src/datasets/iterable_dataset.py#L2225\nIt ignored force_convert_to_python, so it consumed a full Arrow chunk.\nI think it can be fixed if you add `self.force_convert_to_python` to the ... |
I_kwDODunzps79a_wm | 8,131 | datasets.to_list: Behaviour change in version 4.7.0 with the addition of JSON type | ### Describe the bug
There is a change in the approach to decoding JSON types, added in https://github.com/huggingface/datasets/commit/d560b58e1f1ec96591a95a1780ee118f5eea6c74. Since 4.7.0, to_list() returns raw JSON strings for columns stored as Json(), while direct access via __getitem__ returns dicts.
In versions... | CLOSED | 2026-04-13T06:06:43 | 2026-04-27T15:32:30 | 2026-04-27T15:32:30 | https://github.com/huggingface/datasets/issues/8131 | {
"login": "ItsTania"
} | {
"nodes": []
} | [
"I think we can go for the same approach as in https://github.com/huggingface/datasets/pull/8122 which decodes the JSON data (and supports JSON nested in other objects)\n\n(Or decide to decode all the features types, which I think could make sense, but this needs more careful handling so this should be discussed in... |
I_kwDODunzps7800er | 8,129 | `load_dataset()` hangs when `hf_xet` is enabled | ### Describe the bug
Hi! I am getting a peculiar download issue on the latest versions of `datasets` and `hfh` where datasets hang indefinitely:
```py
from datasets import load_dataset
load_dataset("openai/gsm8k", "main", split="test")
```
Note that:
- streaming=True works.
- Direct Hub access works.
- Th... | CLOSED | 2026-04-10T21:00:24 | 2026-04-10T21:07:56 | 2026-04-10T21:07:56 | https://github.com/huggingface/datasets/issues/8129 | {
"login": "lewtun"
} | {
"nodes": []
} | [
"Apologies, seems to have been a stale cache issue. Nuking the datasets cache fixed the issue"
] |
I_kwDODunzps77t3eK | 8,125 | Feature request: being able to disable chmod for flat permission filesystems | ### Feature request
For flat permission file systems, `os.chmod` simply fails when trying to cache datasets - in some places such as: https://github.com/huggingface/datasets/blob/main/src/datasets/arrow_dataset.py#L4043-L4048 for arrow datasets.
This seems to be optional to obtained desired behavior as simply comment... | OPEN | 2026-04-08T08:17:14 | 2026-07-08T16:35:43 | null | https://github.com/huggingface/datasets/issues/8125 | {
"login": "younesbelkada"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"Context: the operation fails on GCS FUSE mounts, and likely will on similar s3-backed file systems.\n\n File \"/home/user/.venv/lib/python3.12/site-packages/datasets/arrow_dataset.py\", line 3747, in _map_single ... |
I_kwDODunzps76aTCF | 8,120 | Patch fields truncated — missing trailing newlines | Description: Multiple entries in the dataset have patches (test.patch and fix.patch) and f2p_patch fields with truncated content. The trailing blank lines that are part of valid unified diff format are stripped, causing git apply to fail with "corrupt patch" errors.
Affected instances (confirmed): atlanhq_atlan-python... | CLOSED | 2026-04-03T16:07:47 | 2026-04-30T15:45:52 | 2026-04-30T15:45:52 | https://github.com/huggingface/datasets/issues/8120 | {
"login": "hetao-git"
} | {
"nodes": []
} | [
"Hi, I'd like to work on this issue. I'll try to reproduce this issue and see where the newline is being stripped.",
"Hi @hetao-git\n\nI was able to reproduce this issue, I used the instance \"atlanhq_atlan-python_pr37\" from the AweAI-Team/Scale-SWE dataset. I observed that applying the GitHub patch (pull/37.dif... |
I_kwDODunzps76LAf0 | 8,116 | `to_json()` serializes `Json()` feature as escaped string instead of native JSON object | ### Describe the bug
When a dataset column uses the `Json()` feature type, calling `to_json()` writes the value as an escaped JSON string (e.g., `"{\"cat_a\":1,\"cat_b\":2}"`) instead of a native JSON object (e.g., `{"cat_a":1,"cat_b":2}`).
This causes two problems:
1. The output JSON file is not interoperable with ... | CLOSED | 2026-04-02T21:59:15 | 2026-04-10T13:12:51 | 2026-04-10T13:12:51 | https://github.com/huggingface/datasets/issues/8116 | {
"login": "ain-soph"
} | {
"nodes": []
} | [
"Hi! I ran into this issue as well, I think the root cause was that under `_batch_json()` of `JsonDatasetWriter` it was directly exporting the encoded string without decoding it first. I will try to add a PR with a small fix. ",
"#self-assign"
] |
I_kwDODunzps749rc2 | 8,110 | `set_format` resets features | ### Describe the bug
Calling `dataset.set_format('...')` resets the informative features such as `Array2D` etc. back into `List(List(Value))` and so on.
### Steps to reproduce the bug
```py
>>> dataset.features
{'lld': Array2D(shape=(None, 26), dtype='float32')}
>>> dataset.set_format('torch')
>>> dataset.features
... | CLOSED | 2026-03-31T07:59:32 | 2026-05-11T07:15:16 | 2026-05-11T07:15:16 | https://github.com/huggingface/datasets/issues/8110 | {
"login": "plutonium-239"
} | {
"nodes": []
} | [
"couldn't replicate with: \n\n``` bash\nPlatform: macos\npython: 3.10.9\ndatasets: 4.8.4\nhuggingface_hub: 0.36.2\npyarrow: 23.0.1\npandas: 2.3.3\nfsspec: 2026.2.0\n\n```\n\n\n``` python\nimport datasets\nfrom datasets import Array2D, Dataset\nimport numpy as np\n\n# Create a dataset with Array2D feature\ndata = {\... |
I_kwDODunzps71k8L4 | 8,085 | ImportError: cannot import name 'VideoReader' from 'torchvision.io' when using datasets with torch format | ### Describe the bug
For PyTorch 2.11 + torchvision 0.26.0 when using the datasets library and setting the format to "torch", an ImportError occurs because the datasets library attempts to import VideoReader from torchvision.io, but this object is not found in the installed torchvision version.
**Error message**:
> ... | CLOSED | 2026-03-23T11:16:20 | 2026-03-23T14:23:25 | 2026-03-23T14:00:57 | https://github.com/huggingface/datasets/issues/8085 | {
"login": "DamJanusz"
} | {
"nodes": []
} | [
"`datasets` 4.8.4 is out and includes a fix :)"
] |
I_kwDODunzps70VYi- | 8,079 | StepExamplesIterable._iter_arrow drops samples for non-zero ranks for single-sharded | ### Describe the bug
As titled
### Steps to reproduce the bug
Running following Python script:
```
"""
Reproducer for HuggingFace datasets 4.x bug:
StepExamplesIterable._iter_arrow drops samples for non-zero ranks
when Arrow batches have size 1 (e.g. single-shard streaming WebDataset).
Bug: split_dataset_by_node w... | CLOSED | 2026-03-19T04:38:51 | 2026-03-19T17:29:40 | 2026-03-19T17:29:40 | https://github.com/huggingface/datasets/issues/8079 | {
"login": "wwwjn"
} | {
"nodes": []
} | [] |
I_kwDODunzps7zk_cT | 8,078 | Inconsistent loading of LFS-hosted files in Dangindev/viet-cultural-vqa dataset | ### Describe the bug
Describe the bug
Some files in the Dangindev/viet-cultural-vqa dataset fail to load via the Hugging Face datasets library.
xet-hosted files load fine
LFS-hosted files sometimes fail
Example:
Fails: https://huggingface.co/datasets/Dangindev/viet-cultural-vqa/blob/main/images/am_thuc/banh_chung_T... | CLOSED | 2026-03-17T06:44:01 | 2026-03-30T04:14:29 | 2026-03-30T04:14:29 | https://github.com/huggingface/datasets/issues/8078 | {
"login": "ngayngo9x"
} | {
"nodes": []
} | [
"I also have been stuck in same problem ",
"This should be fixed, thanks for reporting ! let us know if you ever encounter other issues in the future",
"It's okay. \n\nGood job, thanks your support! "
] |
I_kwDODunzps7zhZjc | 8,075 | `.batch()` error on formatted datasets | The `.batch()` method currently assumes the input (batch) is always a dictionary, which causes errors when it isn't. This can happen with formatted datasets, since formats like `"pyarrow"`, `"pandas"` (only affects `IterableDataset`), and `"polars"` return tables/dataframes instead of dictionaries.
For example:
```pyt... | CLOSED | 2026-03-17T01:08:13 | 2026-04-10T13:14:03 | 2026-04-10T13:14:03 | https://github.com/huggingface/datasets/issues/8075 | {
"login": "mariosasko"
} | {
"nodes": []
} | [
"Hi ! Good catch :) Since table-formatted iterable datasets have an Arrow path (i.e. they have `.iter_arrow()`) I guess a simple fix would be to make `.batch()` make the dataset arrow-formatted and use a `batch_arrow_fn` in that case (instead of the current `_batch_fn` that expects dictionaries). It would also be b... |
I_kwDODunzps7yzk24 | 8,065 | ARM audio support broken | ### Describe the bug
`datasets` 4.0 moved to using `torchcodec` for audio datasets, which doesn't support ARM hardware under Linux and Windows
### Steps to reproduce the bug
try:
* Try to install `torchcodec` under Linux ARM64
or:
* Try to install `torchcodec` under Windows ARM64
or, on either of those platforms... | OPEN | 2026-03-13T22:06:36 | 2026-03-23T14:16:47 | null | https://github.com/huggingface/datasets/issues/8065 | {
"login": "leondz"
} | {
"nodes": []
} | [
"See also https://github.com/NVIDIA/garak/issues/1622#issuecomment-3937720381",
"Hi @leondz , thanks for the detailed bug report — the description and reproduction steps are really clear.\n\nI’d love to take a shot at investigating and potentially fixing this issue. Before I start digging into the codebase, I jus... |
I_kwDODunzps7w65yv | 8,052 | [Security] Zip/Rar/7z Archive Extraction Path Traversal (Incomplete CVE-2007-4559 Mitigation) | ## Security Vulnerability Report
**Reporter:** Conner Webber (conner.webber000@gmail.com)
**Severity:** HIGH (CVSS 3.1: 8.2)
**CWE:** CWE-22 (Improper Limitation of a Pathname to a Restricted Directory)
## Summary
The `ZipExtractor`, `RarExtractor`, and `SevenZipExtractor` classes in `datasets/utils/extract.py` call... | CLOSED | 2026-03-08T19:09:32 | 2026-03-10T23:27:45 | 2026-03-10T23:27:45 | https://github.com/huggingface/datasets/issues/8052 | {
"login": "spartan8806"
} | {
"nodes": []
} | [
"Hi! I'm currently exploring the codebase and came across this issue. I'd be interested in investigating it further and trying to reproduce the behavior locally.\n\nIf no one is actively working on it, I'd be happy to take a look and attempt a fix.\n",
"The steps to reproduce don't cause any file extraction outsi... |
I_kwDODunzps7wTYiu | 8,048 | Support 3D Datasets | ### Feature request
In `datasets` library we have options to load `image` and `audio` datasets which can be viewed in the Dataset Viewer.
LIKE:
```bash
from datasets import load_dataset, Image
from datasets import load_dataset, Audio
```
> https://huggingface.co/docs/datasets/en/image_load & https://huggingface.co/do... | CLOSED | 2026-03-06T01:23:08 | 2026-06-02T17:40:12 | 2026-06-02T17:38:55 | https://github.com/huggingface/datasets/issues/8048 | {
"login": "Vinay-Umrethe"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"Closing, as this got implemented in:\n\n[feat: add 3D mesh support and MeshFolder builder #8055](https://github.com/huggingface/datasets/pull/8055)\n\n[fix: embed_external_files=True for mesh support #8224](https://github.com/huggingface/datasets/pull/8224)"
] |
I_kwDODunzps7wOjwo | 8,046 | `Dataset.map()` with `num_proc > 1` fails when some shards produce `List(Value('null'))` (empty lists) | ### Describe the bug
When using `Dataset.map()` with `num_proc > 1`, if the map function produces a list column where some shards contain only empty lists `[]` and other shards contain non-empty lists of structs, the feature type inference is inconsistent across shards:
- Shards with only empty lists → `List(Value('n... | CLOSED | 2026-03-05T20:12:13 | 2026-03-09T17:52:13 | 2026-03-09T17:52:13 | https://github.com/huggingface/datasets/issues/8046 | {
"login": "ain-soph"
} | {
"nodes": []
} | [
"\n### Root cause\n\nIn `src/datasets/features/features.py`, both `_check_if_features_can_be_aligned` and `_align_features` use:\n\n```python\nisinstance(v, Value) and v.dtype == \"null\"\n```\n\nto check for null compatibility. However, `List(Value('null'))` is **not** an instance of `Value` — it's an instance of ... |
I_kwDODunzps7vnyRo | 8,038 | Typo in iterable_dataset.py | ### Describe the bug
`streaming=True**kwargs,` should be `streaming=True, **kwargs,`
https://github.com/huggingface/datasets/blob/81027be09d5cd9f06a6d64ef1a8a3e9ebd0f86fd/src/datasets/iterable_dataset.py#L2950
### Steps to reproduce the bug
```
from Datasets import Dataset
IterableDataset.from_csv("file.csv")
```
#... | CLOSED | 2026-03-04T05:37:15 | 2026-03-09T17:12:54 | 2026-03-09T17:12:54 | https://github.com/huggingface/datasets/issues/8038 | {
"login": "sybaik1"
} | {
"nodes": []
} | [] |
I_kwDODunzps7u82YW | 8,037 | IterableDataset.filter() chaining fails due to features being lost when is_typed=True | ### Describe the bug
Chaining multiple .filter() calls on an IterableDataset raises a TypeError: 'NoneType' object is not a mapping on the second .filter() call.
In `IterableDataset.filter()`, the internal `ex_iterable` is wrapped with FormattedExamplesIterable:
```python
ex_iterable = FormattedExamplesIterable(
... | CLOSED | 2026-03-02T02:30:24 | 2026-03-09T17:44:34 | 2026-03-09T17:44:34 | https://github.com/huggingface/datasets/issues/8037 | {
"login": "sh0416"
} | {
"nodes": []
} | [] |
I_kwDODunzps7uqxFr | 8,034 | [BUG] load_datasets() cannot use the generated Arrow cache correctly | ### Describe the bug
The datasets library cannot use the generated Arrow cache correctly, seemingly due to a flaw in the internal hash symbol calculation logic.
The following code provides verification.
I am trying to locate the specific code position, and if there are further developments, I will update here.
### S... | OPEN | 2026-02-28T08:11:23 | 2026-03-04T13:50:08 | null | https://github.com/huggingface/datasets/issues/8034 | {
"login": "Nexround"
} | {
"nodes": []
} | [
"PYTHONHASHSEED=1 uv run python -c\n \"\nfrom datasets.load import LocalDatasetModuleFactory\nmod = LocalDatasetModuleFactory('/cache/datasets/imagenet-1k')\ndm = mod.get_module()\nprint('Hash:', dm.hash)\n\"\nResolving data files: 100%|█████████| 294/294 [00:00<00:00, 20584.68it/s]\nResolving data files: 100%|████... |
I_kwDODunzps7uhb53 | 8,029 | `cache_dir` option in `download_config` in `load_dataset` is not respected | ### Describe the bug
Downloaded files still go to `~/.cache/huggingface/hub/` even if I specified `cache_dir` option in `download_config` in `load_dataset`.
### Steps to reproduce the bug
Run my freshly written script and found that downloaded files did not go where I want.
```python
'''
下载OpenWebText数据集，允许使用代理
'''
... | OPEN | 2026-02-27T16:05:30 | 2026-02-27T23:34:28 | null | https://github.com/huggingface/datasets/issues/8029 | {
"login": "TsXor"
} | {
"nodes": []
} | [
"Actually, I used debugger to step into the code of `datasets`, and found `load_dataset` calls into `cached_path`, where it calls into `hf_hub_download` method for `hf://` links:\nhttps://github.com/huggingface/datasets/blob/6e8eaeed4d202b72eff44ba1a9ec5b7d81d2d3e6/src/datasets/utils/file_utils.py#L182-L195\n\nSome... |
I_kwDODunzps7s4rqs | 8,015 | Iterating a streaming dataset correctly in random order | ### Describe the bug
For training to work properly it is considered good practice in machine learning to sample the dataset randomly (uniformly). I am not sure how the user is expected to do it correctly when the creator of the dataset did a bad job.
https://huggingface.co/docs/hub/datasets-webdataset
```
**Shuffle*... | OPEN | 2026-02-22T11:34:06 | 2026-05-11T15:08:20 | null | https://github.com/huggingface/datasets/issues/8015 | {
"login": "unrealwill"
} | {
"nodes": []
} | [
"Hi, reshuffling is actually pretty easy:\n\n```python\n\"\"\" shuffle.py \"\"\"\nfrom datasets import load_dataset\n\nnum_proc = 8 # tweak it here depending on your machine\nseed = 42\nds = load_dataset(\"jackyhate/text-to-image-2M\", num_proc=num_proc)\nds = ds.shuffle(seed=seed)\nds.push_to_hub(\"my-username/te... |
I_kwDODunzps7sQ6QD | 8,012 | HC3 dataset fails to load with datasets>=3 due to legacy script (HC3.py) | ### Describe the bug
It looks like the current documentation suggests loading the dataset like this:
`load_dataset("Hello-SimpleAI/HC3", split="train")
`
But with datasets>=3.x, this raises:
RuntimeError: Dataset scripts are no longer supported, but found HC3.py
The dataset loads correctly if we specify the parqu... | CLOSED | 2026-02-19T16:27:51 | 2026-02-20T11:58:20 | 2026-02-20T11:58:20 | https://github.com/huggingface/datasets/issues/8012 | {
"login": "navan0"
} | {
"nodes": []
} | [
"This is not a bug. This is intended behaviour. The only ways to resolve this are\n1. use the old `datasets` versions\n2. load the dataset and upload it to your own repo without a script\n3. raise an issue in the dataset and wait for the owner to update it."
] |
I_kwDODunzps7sPWzu | 8,010 | audiofolder / folder-based loader rejects metadata.csv when file_name column is inferred as large_string | ### Describe the bug
When loading a dataset with load_dataset("audiofolder", data_dir=...) and a metadata.csv (with a file_name column), loading can fail with:
`ValueError: file_name or *_file_name must be present as dictionary key (with type string) in metadata files`
even when the CSV clearly has a file_name colum... | CLOSED | 2026-02-19T15:06:09 | 2026-02-19T15:29:41 | 2026-02-19T15:29:41 | https://github.com/huggingface/datasets/issues/8010 | {
"login": "Koosh0610"
} | {
"nodes": []
} | [
"Proposing a PR for this [here](https://github.com/huggingface/datasets/pull/8011)\n\nPS: I didn't find a contributing documentation. Please recommend any changes in issue or PR if any. Thank you!"
] |
I_kwDODunzps7rPcqh | 8,007 | Add option for loading audio with video | ### Describe the bug
Currently, `torchcodec` don't allow extracting `Audio` from `Video` https://github.com/meta-pytorch/torchcodec/issues/1158, but when I upload videos with audio to hub using `videofolder`, then this is not possible to retrieve audio from it. Probably `VideoDecoder` can be extended with `audio` para... | OPEN | 2026-02-16T09:11:02 | 2026-06-17T15:25:31 | null | https://github.com/huggingface/datasets/issues/8007 | {
"login": "Samoed"
} | {
"nodes": []
} | [
"Hi,\nI’m interested in working on this issue.\n\nFrom what I understand, audio decoding would involve an AudioDecoder, and previous discussions have suggested handling it separately. I’d like to explore whether integrating audio extraction in relation to the current video decoding workflow would be appropriate, or... |
I_kwDODunzps7rGq1a | 8,006 | Regression: from_generator crashes if one generator call returns no results | ### Describe the bug
Dataset.from_generator splits any list kwarg to the generator and makes a separate call to the generator for each member of the list even in the num_proc=1 case.
I have a generator that processes a number of files, filtering them and producing examples. It used to be the case that things worked fi... | CLOSED | 2026-02-15T16:16:48 | 2026-02-24T13:58:12 | 2026-02-24T13:58:12 | https://github.com/huggingface/datasets/issues/8006 | {
"login": "hartmans"
} | {
"nodes": []
} | [] |
I_kwDODunzps7q9L9J | 8,005 | Multi-channel audio is automatically cast to mono, num_channels is ignored | ### Describe the bug
The `num_channels` parameter in `datasets.Audio()` is documented to preserve stereo channels when set to `None` (preserve original) or `2` (explicit stereo), but it currently downmixes all audio to mono regardless of this setting.
### Steps to reproduce the bug
```python
import numpy as np
impor... | OPEN | 2026-02-14T17:28:03 | 2026-02-25T15:17:18 | null | https://github.com/huggingface/datasets/issues/8005 | {
"login": "ZackHodari"
} | {
"nodes": []
} | [
"**Workaround**\nDirectly load audio using torchcodec, this is what datasets does under the hood (but doesn't maintain multi-channel)\n\n```python\nimport torchcodec\n\ndecoder = torchcodec.decoders.AudioDecoder(audio[\"bytes\"])\naudio_samples = decoder.get_all_samples()\n\naudio = audio_samples.data.numpy()\nsamp... |
I_kwDODunzps7qqhA2 | 8,002 | The `Sequence` class in features do not have "dtype" | ### Describe the bug
**I'm not sure if this is a bug.**
I see that a `FeatureType` object contains an attribute called `self.dtype` that is not covered when this feature is a `Sequence` or a `List`.
When I try to run a multilabel classification with this example script from the transformers library:
https://github.... | OPEN | 2026-02-13T12:45:19 | 2026-02-13T12:52:36 | null | https://github.com/huggingface/datasets/issues/8002 | {
"login": "gonzalo-santamaria-iic"
} | {
"nodes": []
} | [] |
I_kwDODunzps7pX8Ta | 7,999 | Too many dataloader workers: 4 (max is dataset.num_shards=3). Stopping 1 dataloader workers. | Hi !
I’m working on training with a large-scale dataset (100+ Parquet files) using lazy loading, and I’m struggling to understand/optimize the num_shards setting— in the lerobot repo: streaming_datasets.py:
```
from datasets import load_dataset
self.hf_dataset: datasets.IterableDataset = load_dataset(
self.... | OPEN | 2026-02-09T09:26:37 | 2026-02-15T14:25:56 | null | https://github.com/huggingface/datasets/issues/7999 | {
"login": "D222097"
} | {
"nodes": []
} | [
"Hi, thanks for the clear question and code snippet!\n\nFrom my understanding, **hf_dataset.num_shards** represents the number of actual iterable partitions that the streaming dataset can be read from in parallel. This is not the same as the number of underlying Parquet files—many files can be grouped into a much s... |
I_kwDODunzps7pNehu | 7,998 | [doc] Inconsistant ENV VAR Name for Progress Bar Toggle | Code uses env var name `HF_DATASETS_DISABLE_PROGRESS_BARS`.
https://github.com/huggingface/datasets/blob/025593f2f0722f31fc136e0ae45da4ff44d4416a/src/datasets/config.py#L221-L226
Docstrings and warnings report env var name `HF_DATASETS_DISABLE_PROGRESS_BAR` without the ending `S`.
https://github.com/huggingface/data... | CLOSED | 2026-02-08T12:16:44 | 2026-02-16T15:29:50 | 2026-02-16T15:29:50 | https://github.com/huggingface/datasets/issues/7998 | {
"login": "Moenupa"
} | {
"nodes": []
} | [] |
I_kwDODunzps7o1eC2 | 7,994 | Bump fsspec upper bound constraint | Would it be possible to bump fsspec upper bound to the latest (2026.2.0)?
I saw you had some API compat issues in the past (https://github.com/huggingface/datasets/issues/7326) and I understand the need for an upper bound.
But I wonder if you think your CI and tests are a good proxy to catch fsspec API breakage? If t... | CLOSED | 2026-02-06T11:37:54 | 2026-02-16T15:29:00 | 2026-02-16T15:29:00 | https://github.com/huggingface/datasets/issues/7994 | {
"login": "hadim"
} | {
"nodes": []
} | [] |
I_kwDODunzps7oRb0h | 7,991 | list(api.list_datasets()) giving jsondecode error | i am using the python api wrapper to list all datasets available on hugging face. This is for research, and i need all datasets to determine which % have language tags and other related questions requiring the total list. However, the following code that worked a few months ago:
```
from huggingface_hub import HfApi
a... | CLOSED | 2026-02-04T14:39:46 | 2026-02-05T10:30:09 | 2026-02-05T10:30:09 | https://github.com/huggingface/datasets/issues/7991 | {
"login": "Moll-j"
} | {
"nodes": []
} | [] |
I_kwDODunzps7oNkVq | 7,990 | Dataset.map crashes when first examples return None and later examples return dict — writer not initialized | ### Describe the bug
I detected a serious [bug from datasets/arrow_dataset.py](https://github.com/huggingface/datasets/blob/main/src/datasets/arrow_dataset.py#L3676)
---
**Description of the bug**
`Dataset.map` crashes with `writer is None` when the map function returns `None` for the first few examples and a diction... | OPEN | 2026-02-04T10:43:20 | 2026-05-26T19:09:11 | null | https://github.com/huggingface/datasets/issues/7990 | {
"login": "meta-program"
} | {
"nodes": []
} | [
"map() expects a function that always return dict or that always return none",
"Hi, I'd like to work on this if the team is open to it.\n\nI understand from @lhoestq's comment that `map()` currently expects\na consistent return type. However, the crash with `AttributeError: \n'NoneType' object has no attribute 'w... |
I_kwDODunzps7oLmBb | 7,988 | `Dataset.map()` breaks when `function` calls `import polars as pl` and `num_proc`>1: "UnboundLocalError: cannot access local variable 'pl' where it is not associated with a value" | ### Describe the bug
# Repro
These two conditions seem to consistently reproduce the issue:
- function passed to `Dataset.map()` explicitly or implicitly calls `import polars as pl`
- `num_proc` > 1
# Trace
```
RemoteTraceback Traceback (most recent call last)
RemoteTraceback:
"""
Traceback... | CLOSED | 2026-02-04T08:42:23 | 2026-02-24T14:08:16 | 2026-02-24T14:08:16 | https://github.com/huggingface/datasets/issues/7988 | {
"login": "ligz08"
} | {
"nodes": []
} | [] |
I_kwDODunzps7oBw7L | 7,986 | `Dataset.map()` causes cache miss/fingerprint change when closure captures self containing non-deterministic state. | ### Describe the bug
When using `.map()` with a function defined inside a **class (of which has any non-deterministic states)** method (a closure), if that function captures `self` to access a configuration variable (e.g., self.foo), the fingerprint mechanism serializes the entire class instance state.
If the class i... | OPEN | 2026-02-03T19:16:49 | 2026-02-06T08:38:13 | null | https://github.com/huggingface/datasets/issues/7986 | {
"login": "Cloud0310"
} | {
"nodes": []
} | [
"I suggest metion this in docs specifically for attention with use, tell users explicitly to pass arguments with `fn_kwargs` param or using `functools.partial` to create a pure funcion."
] |
I_kwDODunzps7n8obB | 7,984 | Data | OPEN | 2026-02-03T14:01:48 | 2026-02-03T14:01:48 | null | https://github.com/huggingface/datasets/issues/7984 | {
"login": "iLenceJhay"
} | {
"nodes": []
} | [] | |
I_kwDODunzps7mpvMC | 7,970 | cast_column(..., Audio) fails with load_dataset("csv",) | ### Describe the bug
Attempt to load a dataset from a csv with a single `audio` column with a single row with a path to an audio file fails when casting the column to Audio, but the exact same dataset created from a dictionary succeeds.
### Steps to reproduce the bug
1. Have any valid audio file `audio.wav`
2. Have ... | OPEN | 2026-01-29T09:33:35 | 2026-04-04T12:45:45 | null | https://github.com/huggingface/datasets/issues/7970 | {
"login": "jstangroome"
} | {
"nodes": []
} | [
"The following code *does* work:\n```py\nfrom datasets import load_dataset,Audio,Features\n\ndataset = load_dataset(\"csv\",data_files=\"audio.csv\",features=Features({\"audio\": Audio()}))\nprint(dataset[\"train\"][0][\"audio\"])\n```",
"Thanks for reporing ! Are you using pandas v3 by any chance ? The CSV loade... |
I_kwDODunzps7mXwrD | 7,968 | Potential conflicting type checks and dead code in `/src/datasets/table.py` | When statically analyzing and manually reviewing the code, I noticed a potential logic conflicting in `/src/datasets/table.py` as follows:
```
def to_blocks(table: Union[pa.Table, Table]) -> list[list[TableBlock]]:
if isinstance(table, pa.Table):
return [[InMemoryTable(table)]]
elif isinstance(ta... | CLOSED | 2026-01-28T11:34:53 | 2026-05-11T11:04:04 | 2026-05-11T11:04:04 | https://github.com/huggingface/datasets/issues/7968 | {
"login": "rc4typecheck"
} | {
"nodes": []
} | [
"ConcatenationTable is a subclass of datasets.table.Table but not pa.Table, so it should be fine",
"Hi! Triaging stale issues — I don't think this one is actually dead code; the premise is off.\n\nThe claim is that `isinstance(table, ConcatenationTable)` at line 4 is unreachable because the line-2 `isinstance(tab... |
I_kwDODunzps7l-8ld | 7,965 | `huggingface_hub.errors.HfHubHTTPError: 404 Client Error: Not Found for url` when fetching a dataset with `datasets.load_dataset` | Not a bug but a question. We started getting the following error:
https://github.com/mlflow/mlflow/actions/runs/21368603305/job/61506951617
```
ests/data/test_huggingface_dataset_and_source.py::test_from_huggingface_dataset_constructs_expected_dataset_with_revision - huggingface_hub.errors.HfHubHTTPError: 404 Client ... | CLOSED | 2026-01-27T02:20:31 | 2026-01-28T15:14:50 | 2026-01-28T15:14:50 | https://github.com/huggingface/datasets/issues/7965 | {
"login": "harupy"
} | {
"nodes": []
} | [
"Hi ! Yes you should use `cornell-movie-review-data/rotten_tomatoes` instead of `rotten_tomatoes`, which is the legacy name. Those datasets have been moved under their actual owners accounts some time ago (but we were keeping the old names as aliases)\n\nSome other impacted names are:\n- `imdb` -> `stanfordnlp/imdb... |
I_kwDODunzps7lT2AI | 7,958 | [CUDA Tensors Not working in ~v4.5.0] set_format(type="torch", device="cuda") returns cpu | ### Describe the bug
The problem is that when calling:
```ds.set_format(type="torch", columns = ["input", "labels"], device="cuda")```
The device type of the individual datapoints is now: `cpu` as opposed to `cuda:0`.
With `v4.0.0` it still works. With `v4.5.0` it doesn't work anymore.
Related Issue:
https://github... | CLOSED | 2026-01-23T12:06:48 | 2026-03-04T13:17:28 | 2026-03-04T13:17:28 | https://github.com/huggingface/datasets/issues/7958 | {
"login": "ai-nikolai"
} | {
"nodes": []
} | [
"@lhoestq tagging you here as you were on the previous issue, hope that's fine. ",
"Thanks for reporting, let me take a look",
"I reverted this change which caused the issue #7961 , I'll do a new release soon but in the meantime feel free to install `datasets` from source",
"@lhoestq thanks a lot. I am actual... |
I_kwDODunzps7k08AC | 7,956 | Is the 10k files / folder limit a hard limit for a dataset repo? | ### Feature request
Can the hard limit of 10k files per folder be extended with acceptable loss in performance?
### Motivation
I'm uploading a lance dataset to huggingface hub and have a folder inside lance internals (`data/*.lance/_transactions`) that has > 20k atomic transaction records and my commits are being re... | OPEN | 2026-01-21T16:37:38 | 2026-01-22T09:38:38 | null | https://github.com/huggingface/datasets/issues/7956 | {
"login": "pavanramkumar"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"Yes, that's a hard limit. Can you split your files into different folders? Or we'll probably have a new repo type in the near to mid future that will relax this limit a bit. ",
"Thanks! Working around this with a different sharding parameter to have fewer overall fragments (and therefore fewer files in `*.lance/... |
I_kwDODunzps7ktEe5 | 7,954 | all_exhausted_without_replacement working same as first_exhausted | ### Describe the bug
>>> from datasets import Dataset, interleave_datasets
>>> d1 = Dataset.from_dict({"a": [0, 1, 2]})
>>> d2 = Dataset.from_dict({"a": [10, 11, 12, 13]})
>>> d3 = Dataset.from_dict({"a": [20, 21, 22]})
>>> dataset = interleave_datasets([d1, d2, d3], stopping_strategy="all_exhausted_without_replacemen... | CLOSED | 2026-01-21T07:50:31 | 2026-03-23T14:03:13 | 2026-03-23T14:03:13 | https://github.com/huggingface/datasets/issues/7954 | {
"login": "prathamk-tw"
} | {
"nodes": []
} | [
"https://github.com/huggingface/datasets/pull/7955",
"it was fixed by https://github.com/huggingface/datasets/pull/7955"
] |
I_kwDODunzps7j4xPb | 7,947 | MMLU get_dataset_config_names provides different lists of subsets in online and offline modes | ### Describe the bug
When getting the config names of `cais/mmlu` in online mode, it provides the different subjects and `all` but in offline mode is provides `default` even with cached version of the dataset.
### Steps to reproduce the bug
1. First download dataset in online mode so that it is cached:
```
$ HF_DATA... | OPEN | 2026-01-16T19:20:08 | 2026-01-16T19:20:08 | null | https://github.com/huggingface/datasets/issues/7947 | {
"login": "rikrd"
} | {
"nodes": []
} | [] |
I_kwDODunzps7jjSZ2 | 7,946 | Question: Is there a faster way to push_to_hub for large image datasets? | # Question: Is there a faster way to `push_to_hub` for large image datasets? Or could this approach be integrated?
Hi! I frequently work with large image datasets (100k-300k+ samples) and found that `dataset.push_to_hub()` can be quite slow.
cc @lhoestq - would love your thoughts on this!
I experimented with an alte... | OPEN | 2026-01-15T13:54:37 | 2026-01-17T13:27:20 | null | https://github.com/huggingface/datasets/issues/7946 | {
"login": "adithya-s-k"
} | {
"nodes": []
} | [
"This is a really interesting approach, especially combining parallel parquet\nconversion with upload_large_folder and hf_xet.\n\nOne question / observation:\nThe shard calculation currently uses max_shard_size_mb as a proxy for\nsamples_per_shard. Since sample size can vary a lot across datasets,\nwould it make se... |
I_kwDODunzps7i6IeO | 7,939 | datasets.load_from_disk progress bar optional manual control | ### Feature request
This is tangentially related to [https://github.com/huggingface/datasets/issues/7918](https://github.com/huggingface/datasets/issues/7918).
When loading a dataset with > 16 files a progress bar is shown (unless stdout is redirected or [https://github.com/huggingface/datasets/pull/7919](https://git... | OPEN | 2026-01-13T03:19:13 | 2026-06-28T23:30:45 | null | https://github.com/huggingface/datasets/issues/7939 | {
"login": "Tommigun1980"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [] |
I_kwDODunzps7iPo1_ | 7,936 | _add_retries_to_file_obj_read_method makes file_obj invalid for pyarrow | ### Describe the bug
I'm trying to use `load_dataset` to construct a dataset that read parquet data on HDFS streamingly, like
```python
ds = load_dataset(
"parquet",
data_files={
"train": "hdfs://xxx/train*.parquet",
"test": "hdfs://xxx/test*.parquet"
},
streaming=True,
)
```
I encounte... | CLOSED | 2026-01-09T07:05:25 | 2026-01-14T13:47:50 | 2026-01-13T13:38:03 | https://github.com/huggingface/datasets/issues/7936 | {
"login": "li-yi-dong"
} | {
"nodes": []
} | [
"hmm not sure how to fix this, I believe `file_obj.__getattr__ = lambda _, attr: getattr(orig_file_obj, attr)` would make all the methods point to the original file_obj",
"> hmm not sure how to fix this, I believe `file_obj.__getattr__ = lambda _, attr: getattr(orig_file_obj, attr)` would make all the methods poi... |
I_kwDODunzps7iDyGN | 7,934 | xPath cannot handle hdfs:///xxxx properly | ### Describe the bug
_as_str('hdfs:///xxxx') would return hdfs://xxxx. Removing one / and making the path invalid.
For the use case like
```
ds = load_dataset(
"parquet",
data_files={
"train": "hdfs:///user/path/to/data/train*.parquet",
},
streaming=True,
storage_options={
"host": ... | OPEN | 2026-01-08T12:14:11 | 2026-01-08T12:14:11 | null | https://github.com/huggingface/datasets/issues/7934 | {
"login": "li-yi-dong"
} | {
"nodes": []
} | [] |
I_kwDODunzps7hKo9P | 7,931 | Enable CORS + HTTP Range support for browser partial reads on cas-bridge.xethub.hf.co (Parquet row-group access) | ### Feature request
## Summary
Browser-based data tools need Range requests to read Parquet efficiently (footer + selected row groups). Downloads from the Hub redirect to cas-bridge.xethub.hf.co (Xet bridge). The redirected host fails CORS preflight for Range/HEAD workflows, blocking partial reads. ([Hugging Face](htt... | CLOSED | 2026-01-03T04:23:54 | 2026-02-24T21:31:37 | 2026-02-24T16:35:44 | https://github.com/huggingface/datasets/issues/7931 | {
"login": "cornhundred"
} | {
"nodes": [
{
"name": "enhancement"
}
]
} | [
"Cc @assafvayner maybe ? or @cfahlgren1 @severo if you've already encountered this ?",
"OK, reproduced with hyparquet on https://huggingface.co/spaces/hyperparam/hyperparam, see https://huggingface.co/spaces/hyperparam/hyperparam?url=https%3A%2F%2Fhuggingface.co%2Fdatasets%2Ffacebook%2Fresearch-plan-gen%2Fblob%2F... |
I_kwDODunzps7hDseJ | 7,928 | `load_dataset` `revision` param not respected when fetching from cache | ### Describe the bug
`datasets.load_dataset` `revision` semantics are a bit inconsistent when the dataset is not found on the huggingface hub. When fetching the latest cached version of the dataset, the `revision` argument is ignored, so long as any cached versions of the dataset already exist in the HF cache.
### St... | OPEN | 2026-01-02T08:20:47 | 2026-04-17T16:01:16 | null | https://github.com/huggingface/datasets/issues/7928 | {
"login": "Scott-Simmons"
} | {
"nodes": []
} | [
"This might be better placed as a feature request not a bug, since the logging `Using the latest cached version of the dataset since sentientfutures/ahb couldn't be found on the Hugging Face Hub` is clear.",
"https://github.com/huggingface/datasets/pull/7929 This only solves the case of invalid revisions. Fetchin... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.