title
stringlengths 1
290
| body
stringlengths 0
228k
⌀ | html_url
stringlengths 46
51
| comments
list | pull_request
dict | number
int64 1
5.59k
| is_pull_request
bool 2
classes |
|---|---|---|---|---|---|---|
Improving elasticsearch integration
|
- adding murmurhash signature to sample in index
- adding optional credentials for remote elasticsearch server
- enabling sample update in index
- upgrade the elasticsearch 7.10.1 python client
- adding ElasticsearchBulider to instantiate a dataset from an index and a filtering query
|
https://github.com/huggingface/datasets/pull/3042
|
[
"@lhoestq @albertvillanova Iwas trying to fix the failing tests in circleCI but is there a test elasticsearch instance somewhere? If not, can I launch a docker container to have one?"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3042",
"html_url": "https://github.com/huggingface/datasets/pull/3042",
"diff_url": "https://github.com/huggingface/datasets/pull/3042.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3042.patch",
"merged_at": null
}
| 3,042
| true
|
Load private data files + use glob on ZIP archives for json/csv/etc. module inference
|
As mentioned in https://github.com/huggingface/datasets/issues/3032 loading data files from private repository isn't working correctly because of the data files resolved.
#2986 did a refactor of the data files resolver. I added authentication to it.
I also improved it to glob inside ZIP archives to look for json/csv/etc. files and infer which dataset builder (json/csv/etc.) to use.
Fix https://github.com/huggingface/datasets/issues/3032
Note that #2986 needs to get merged first
|
https://github.com/huggingface/datasets/pull/3041
|
[
"I have an error on windows:\r\n```python\r\naiohttp.client_exceptions.ClientConnectorCertificateError: Cannot connect to host moon-staging.huggingface.co:443 ssl:True [SSLCertVerificationError: (1, '[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: certificate has expired (_ssl.c:1131)')]\r\n```\r\nat the `fsspec` call in `xglob`:\r\n```python\r\nfs, *_ = fsspec.get_fs_token_paths(urlpath, storage_options=storage_options)\r\n```\r\n\r\nLooks like the windows CI has an SSL issue... ",
"I can reproduce it on my windows machine. On linux it works fine though",
"I'm just skipping the windows test for now",
"The Windows CI failure seems unrelated to this PR\r\n```python\r\nERROR tests/test_arrow_dataset.py::test_dummy_dataset_serialize_s3\r\n```"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3041",
"html_url": "https://github.com/huggingface/datasets/pull/3041",
"diff_url": "https://github.com/huggingface/datasets/pull/3041.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3041.patch",
"merged_at": "2021-10-12T15:25:46"
}
| 3,041
| true
|
[save_to_disk] Using `select()` followed by `save_to_disk` saves complete dataset making it hard to create dummy dataset
|
## Describe the bug
When only keeping a dummy size of a dataset (say the first 100 samples), and then saving it to disk to upload it in the following to the hub for easy demo/use - not just the small dataset is saved but the whole dataset with an indices file. The problem with this is that the dataset is still very big.
## Steps to reproduce the bug
E.g. run the following:
```python
from datasets import load_dataset, save_to_disk
nlp = load_dataset("glue", "mnli", split="train")
nlp.save_to_disk("full")
nlp = nlp.select(range(100))
nlp.save_to_disk("dummy")
```
Now one can see that both `"dummy"` and `"full"` have the same size. This shouldn't be the case IMO.
## Expected results
IMO `"dummy"` should be much smaller so that one can easily play around with the dataset on the hub.
## Actual results
Specify the actual results or traceback.
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.2.dev0
- Platform: Linux-5.11.0-34-generic-x86_64-with-glibc2.10
- Python version: 3.8.5
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/3040
|
[
"Hi,\r\n\r\nthe `save_to_disk` docstring explains that `flatten_indices` has to be called on a dataset before saving it to save only the shard/slice of the dataset.",
"That works! Thansk!\r\n\r\nMight be worth doing that automatically actually in case the `save_to_disk` is called on a dataset that has an indices mapping :-)",
"I agree with @patrickvonplaten: this issue is reported recurrently, so better if we implement the `.flatten_indices()` automatically?",
"That would be great indeed - I don't really see a use case where one would not like to call `.flatten_indices()` before calling `save_to_disk`",
"+1 on this !"
] | null | 3,040
| false
|
Add sberquad dataset
| null |
https://github.com/huggingface/datasets/pull/3039
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3039",
"html_url": "https://github.com/huggingface/datasets/pull/3039",
"diff_url": "https://github.com/huggingface/datasets/pull/3039.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3039.patch",
"merged_at": "2021-10-13T10:16:04"
}
| 3,039
| true
|
add sberquad dataset
| null |
https://github.com/huggingface/datasets/pull/3038
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3038",
"html_url": "https://github.com/huggingface/datasets/pull/3038",
"diff_url": "https://github.com/huggingface/datasets/pull/3038.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3038.patch",
"merged_at": null
}
| 3,038
| true
|
SberQuad
| null |
https://github.com/huggingface/datasets/pull/3037
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3037",
"html_url": "https://github.com/huggingface/datasets/pull/3037",
"diff_url": "https://github.com/huggingface/datasets/pull/3037.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3037.patch",
"merged_at": null
}
| 3,037
| true
|
Protect master branch to force contributions via Pull Requests
|
In order to have a clearer Git history in the master branch, I propose to protect it so that all contributions must be done through a Pull Request and no direct commits to master are allowed.
- The Pull Request allows to give context, discuss any potential issues and improve the quality of the contribution
- The Pull Request will eventually be squashed and merged into master with a single commit that links to the Pull Request page (with all the context/discussions)
Note that we already implemented a protection in the master branch to avoid *merge* commits and ensure a linear history. This proposal goes one step further by avoiding all kind of direct commits and forcing contributions **only** through Pull Requests.
Please note that we can temporarily deactivate this protection if we need to make a direct commit, e.g. at each new version release.
The only way GitHub allows this kind or protection is by requiring a minimal number (at least one) of approvals of the Pull Request. The inconvenient is that the PR creator cannot approve their own PR: another person must approve it before it can be merged into master. To circumvent this, we could eventually disable this protection in the master branch when an urgent commit is needed (e.g. for a hotfix) and there is no other person available at that time to approve the PR.
|
https://github.com/huggingface/datasets/issues/3036
|
[
"It would be nice to protect the master from direct commits, but still having a way to merge our own PRs when no review is required (for example when updating a dataset_infos.json file, or minor bug fixes - things that happen quite often actually).\r\nDo you know if there's a way ?",
"you can if you're an admin of the repo",
"This is done. Now the master branch is protected:\r\n- [x] Require a pull request before merging: all commits must be made to a non-protected branch and submitted via a pull request\r\n - Required number of approvals before merging: 1 \r\n- [x] Require linear history: prevent merge commits from being pushed\r\n- [x] These requirements are not enforced for administrators\r\n- [x] Additionally, the master branch is also protected against deletion and force pushes\r\n\r\nCC: @lhoestq @julien-c @thomwolf "
] | null | 3,036
| false
|
`load_dataset` does not work with uploaded arrow file
|
## Describe the bug
I've preprocessed and uploaded a dataset here: https://huggingface.co/datasets/ami-wav2vec2/ami_headset_single_preprocessed . The dataset is in `.arrow` format.
The dataset can correctly be loaded when doing:
```bash
git lfs install
git clone https://huggingface.co/datasets/ami-wav2vec2/ami_headset_single_preprocessed
```
followed by
```python
from datasets import load_from_disk
ds = load_from_disk("./ami_headset_single_preprocessed")
```
However when I try to directly download the dataset as follows:
```python
from datasets import load_dataset
ds = load_dataset("ami-wav2vec2/ami_headset_single_preprocessed")
```
the following error occurs:
```bash
/usr/local/lib/python3.7/dist-packages/datasets/load.py in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, ignore_verifications, keep_in_memory, save_infos, script_version, use_auth_token, task, streaming, **config_kwargs)
1115 ignore_verifications=ignore_verifications,
1116 try_from_hf_gcs=try_from_hf_gcs,
-> 1117 use_auth_token=use_auth_token,
1118 )
1119
/usr/local/lib/python3.7/dist-packages/datasets/builder.py in download_and_prepare(self, download_config, download_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, **download_and_prepare_kwargs)
635 if not downloaded_from_gcs:
636 self._download_and_prepare(
--> 637 dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
638 )
639 # Sync info
/usr/local/lib/python3.7/dist-packages/datasets/builder.py in _download_and_prepare(self, dl_manager, verify_infos, **prepare_split_kwargs)
724 try:
725 # Prepare split will record examples associated to the split
--> 726 self._prepare_split(split_generator, **prepare_split_kwargs)
727 except OSError as e:
728 raise OSError(
/usr/local/lib/python3.7/dist-packages/datasets/builder.py in _prepare_split(self, split_generator)
1186 generator, unit=" tables", leave=False, disable=bool(logging.get_verbosity() == logging.NOTSET)
1187 ):
-> 1188 writer.write_table(table)
1189 num_examples, num_bytes = writer.finalize()
1190
/usr/local/lib/python3.7/dist-packages/datasets/arrow_writer.py in write_table(self, pa_table, writer_batch_size)
424 # reorder the arrays if necessary + cast to self._schema
425 # we can't simply use .cast here because we may need to change the order of the columns
--> 426 pa_table = pa.Table.from_arrays([pa_table[name] for name in self._schema.names], schema=self._schema)
427 batches: List[pa.RecordBatch] = pa_table.to_batches(max_chunksize=writer_batch_size)
428 self._num_bytes += sum(batch.nbytes for batch in batches)
/usr/local/lib/python3.7/dist-packages/pyarrow/table.pxi in pyarrow.lib.Table.from_arrays()
/usr/local/lib/python3.7/dist-packages/pyarrow/table.pxi in pyarrow.lib._sanitize_arrays()
/usr/local/lib/python3.7/dist-packages/pyarrow/array.pxi in pyarrow.lib.asarray()
/usr/local/lib/python3.7/dist-packages/pyarrow/table.pxi in pyarrow.lib.ChunkedArray.cast()
/usr/local/lib/python3.7/dist-packages/pyarrow/compute.py in cast(arr, target_type, safe)
279 else:
280 options = CastOptions.unsafe(target_type)
--> 281 return call_function("cast", [arr], options)
282
283
/usr/local/lib/python3.7/dist-packages/pyarrow/_compute.pyx in pyarrow._compute.call_function()
/usr/local/lib/python3.7/dist-packages/pyarrow/_compute.pyx in pyarrow._compute.Function.call()
/usr/local/lib/python3.7/dist-packages/pyarrow/error.pxi in pyarrow.lib.pyarrow_internal_check_status()
/usr/local/lib/python3.7/dist-packages/pyarrow/error.pxi in pyarrow.lib.check_status()
ArrowNotImplementedError: Unsupported cast from struct<train: struct<name: string, num_bytes: int64, num_examples: int64, dataset_name: string>, validation: struct<name: string, num_bytes: int64, num_examples: int64, dataset_name: string>, test: struct<name: string, num_bytes: int64, num_examples: int64, dataset_name: string>> to list using function cast_list
```
## Expected results
The dataset should be correctly loaded with `load_dataset` IMO.
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.2.dev0
- Platform: Linux-5.11.0-34-generic-x86_64-with-glibc2.10
- Python version: 3.8.5
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/3035
|
[
"Hi ! This is not a bug, this is simply not implemented.\r\n`save_to_disk` is for on-disk serialization and was not made compatible for the Hub.\r\nThat being said, I agree we actually should make it work with the Hub x)",
"cc @LysandreJik maybe we can solve this at the same time as adding `push_to_hub`"
] | null | 3,035
| false
|
Errors loading dataset using fs = a gcsfs.GCSFileSystem
|
## Describe the bug
Cannot load dataset using a `gcsfs.GCSFileSystem`. I'm not sure if this should be a bug in `gcsfs` or here...
Basically what seems to be happening is that since datasets saves datasets as folders and folders aren't "real objects" in gcs, gcsfs raises a 404 error. There are workarounds if you use gcsfs directly to download the file, but as is I can't get `load_from_disk` to work.
## Steps to reproduce the bug
```python
from datasets import load_dataset
# load some dataset
dataset = load_dataset("squad", split="train")
# save it to gcs
import gcsfs
fs = gcsfs.GCSFileSystem(project="my-gs-project")
dataset.save_to_disk("gs://my-bucket/squad", fs=fs)
# try to load it from gcs
from datasets import load_from_disk
dataset2 = load_from_disk("my-bucket/squad", fs=fs)
```
## Expected results
`dataset2` would be a copy of `dataset` but loaded from my bucket.
## Actual results
Long traceback but essentially it's a 404 error from gcsfs saying the object `my-bucket/squad` doesn't exist when this is called:
https://github.com/huggingface/datasets/blob/9c81b7d2e6d9feae69a084a3abda265a4ca07fb5/src/datasets/arrow_dataset.py#L977
This is because there is no actual object called `my-bucket/squad`, there are objects called `my-bucket/squad/dataset.arrow`, etc.
Note that *this* works fine, since it's explicitly saying "download all the objects with this prefix":
```python
fs.download(src_dataset_path + "/*", dataset_path.as_posix(), recursive=True)
```
For example, I can do a workaround this way:
```python
import tempfile
with tempfile.TemporaryDirectory() as temppath:
fs.download("gs://my-bucket/squad/*", temppath)
dataset2 = load_from_disk(temppath)
```
It's unclear to me if it's `gcsfs`'s responsibility to say "hey that's folder not a file, I should try to get objects inside of it not the object itself", or if that's `datasets`'s responsibility... I'm leaning towards the latter since you're never loading a dataset from one file using this function/method, only a dataset folder?
Another minor thing that should maybe should be rolled into this bug...
https://github.com/huggingface/datasets/blob/9c81b7d2e6d9feae69a084a3abda265a4ca07fb5/src/datasets/arrow_dataset.py#L968
These fail if you pass in a `gs://` path, e.g.
```python
dataset2 = load_from_disk("gs://my-bucket/squad", fs=fs)
```
Because at this point, `dataset_info_path` is `gs:/my-bucket/squad/dataset_info.json`, gcsfs throws a:
```
Invalid bucket name: 'gs:'
```
error
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.1
- Platform: macOS Big Sur 11.6
- Python version: 3.7.12
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/3034
|
[] | null | 3,034
| false
|
Actual "proper" install of ruamel.yaml in the windows CI
|
It was impossible to update the package directly with `pip`. Indeed it was installed with `distutils` which prevents `pip` or `conda` to uninstall it.
I had to `rm` a directory from the `site-packages` python directory, and then do `pip install ruamel.yaml`
It's not that "proper" but I couldn't find better solutions
|
https://github.com/huggingface/datasets/pull/3033
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3033",
"html_url": "https://github.com/huggingface/datasets/pull/3033",
"diff_url": "https://github.com/huggingface/datasets/pull/3033.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3033.patch",
"merged_at": "2021-10-05T17:54:56"
}
| 3,033
| true
|
Error when loading private dataset with "data_files" arg
|
## Describe the bug
A clear and concise description of what the bug is.
Private datasets with no loading script can't be loaded using `data_files` parameter.
## Steps to reproduce the bug
```python
from datasets import load_dataset
data_files = {"train": "**/train/*/*.jsonl", "valid": "**/valid/*/*.jsonl"}
dataset = load_dataset('dalle-mini/encoded', data_files=data_files, use_auth_token=True, streaming=True)
```
Same error happens in non-streaming mode.
## Expected results
Files should be loaded (whether in streaming or not).
## Actual results
Error:
```
---------------------------------------------------------------------------
FileNotFoundError Traceback (most recent call last)
/usr/local/lib/python3.7/dist-packages/datasets/load.py in prepare_module(path, script_version, download_config, download_mode, dataset, force_local_path, dynamic_modules_path, return_resolved_file_path, return_associated_base_path, data_files, **download_kwargs)
539 try:
--> 540 local_path = cached_path(file_path, download_config=download_config)
541 except FileNotFoundError:
8 frames
FileNotFoundError: Couldn't find file at https://huggingface.co/datasets/dalle-mini/encoded/resolve/main/encoded.py
During handling of the above exception, another exception occurred:
HTTPError Traceback (most recent call last)
HTTPError: 404 Client Error: Not Found for url: https://huggingface.co/api/datasets/dalle-mini/encoded?full=true
During handling of the above exception, another exception occurred:
FileNotFoundError Traceback (most recent call last)
/usr/local/lib/python3.7/dist-packages/datasets/load.py in prepare_module(path, script_version, download_config, download_mode, dataset, force_local_path, dynamic_modules_path, return_resolved_file_path, return_associated_base_path, data_files, **download_kwargs)
547 except Exception:
548 raise FileNotFoundError(
--> 549 f"Couldn't find a directory or a {resource_type} named '{path}'. "
550 f"It doesn't exist locally at {expected_dir_for_combined_path_abs} or remotely on {hf_api.endpoint}/datasets"
551 )
FileNotFoundError: Couldn't find a directory or a dataset named 'dalle-mini/encoded'. It doesn't exist locally at /content/dalle-mini/encoded or remotely on https://huggingface.co/datasets
```
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.1
- Platform: Linux-5.4.104+-x86_64-with-Ubuntu-18.04-bionic
- Python version: 3.7.12
- PyArrow version: 3.0.0
@lhoestq
|
https://github.com/huggingface/datasets/issues/3032
|
[
"We'll do a release tomorrow or on wednesday to make the fix available :)\r\n\r\nThanks for reproting !"
] | null | 3,032
| false
|
Align tqdm control with cache control
|
Currently, once disabled with `disable_progress_bar`, progress bars cannot be re-enabled again. To overcome this limitation, this PR introduces the `set_progress_bar_enabled` function that accepts a boolean indicating whether to display progress bars. The goal is to provide a similar API to the existing cache control API. Following the Zen of Python (😄), there should be one and preferably only one obvious way to do it, so I'm also deprecating the aforementioned `disable_progress_bar` function. Additionally, I justify the deprecation with the fact that this function has never been in the docs.
Moreover, similar API changes have recently been introduced to [`tfds`](https://github.com/tensorflow/datasets/blob/a1e8b98f45b0214082b546cc967c67c43fffda55/tensorflow_datasets/core/utils/tqdm_utils.py#L98-L112).
Considering the popularity of the [comment](https://github.com/huggingface/datasets/issues/1627#issuecomment-751383559) I made a while ago, this API (`set_progress_bar_enabled` and `is_progress_bar_enabled`) should be mentioned in the docs, but I'm not sure where to put it exactly. Maybe we can replace the `logging_methods` page under `package_reference` with `utility_methods` and then introduce two subsections on that page: `Logging methods` and `tqdm control`.
Additionally, this PR:
* adds the `disable_tqdm` keyword arg of `Dataset._map_single` to the `ignore_kwargs` list to ignore it when computing the fingerprint (forgot to add it in #2696)
* deletes the unused components in `tqdm_utils.py`, which seem to be inherited from `tfds`
* disables the tqdm output in the test suite. As I see it, this output doesn't seem informative, but let me know if this is not a good idea
|
https://github.com/huggingface/datasets/pull/3031
|
[
"Could you add this function to the documentation please ?\r\n\r\nYou can add it in `main_classes.rst`, and maybe add a `Tip` section in the `map` section in the `process.rst`"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3031",
"html_url": "https://github.com/huggingface/datasets/pull/3031",
"diff_url": "https://github.com/huggingface/datasets/pull/3031.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3031.patch",
"merged_at": "2021-10-18T14:59:30"
}
| 3,031
| true
|
Add `remove_columns` to `IterableDataset`
|
Fixes #2944
WIP
* Not tested yet.
* We might want to allow batched remove for efficiency.
@lhoestq Do you think it should have `batched=` and `batch_size=`?
|
https://github.com/huggingface/datasets/pull/3030
|
[
"Thanks ! That looks all good :)\r\n\r\nI don't think that batching would help. Indeed we're dealing with python iterators that yield elements one by one, so batched `map` needs to accumulate a batch, apply the function, and then yield examples from the batch.\r\n\r\nThough once we have parallel processing in `map`, we can reconsider it\r\n\r\nAlso feel free to check the CI failure - apparently the import of `Union` is missing",
"Thanks for the review and explaining that! \r\nOn top of what you said, I think `remove_columns` is very unlikely to be a bottleneck, so it doesn't matter anyways.",
"Thank you for reviewing! @mariosasko \r\n\r\nI wonder how the checking would work. Is there any checking present in `IterableDataset ` now? What if `.remove_columns()` is applied after some arbitrary `.map()`?",
"> I wonder how the checking would work. Is there any checking present in IterableDataset now? What if .remove_columns() is applied after some arbitrary .map()?\r\n\r\nThat's the challenge here indeed ^^ In this case it's not trivial to know the names of the columns. Feel free to open an issue so we can discuss this"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3030",
"html_url": "https://github.com/huggingface/datasets/pull/3030",
"diff_url": "https://github.com/huggingface/datasets/pull/3030.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3030.patch",
"merged_at": "2021-10-08T15:31:53"
}
| 3,030
| true
|
Use standard open-domain validation split in nq_open
|
The nq_open dataset originally drew the validation set from this file:
https://github.com/google-research-datasets/natural-questions/blob/master/nq_open/NQ-open.efficientqa.dev.1.1.sample.jsonl
However, that's the dev set used specifically and only for the efficientqa competition, and it's not the same dev set as is used in every open-domain question answering paper (including the Lee et al paper that introduced the open-domain variant of NQ, cited at the top of the dataset file). This PR changes nq_open to use the standard validation split and bumps the version to 2.0.0 since this is a breaking change.
|
https://github.com/huggingface/datasets/pull/3029
|
[
"I had to run datasets-cli with --ignore_verifications the first time since it was complaining about a missing file, but now it runs without that flag fine. I moved dummy_data.zip to the new folder, but also had to modify the filename of the test file in the zip (should I not have done that?). Finally, I added the pretty name tag.",
"Great, thanks for the help."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3029",
"html_url": "https://github.com/huggingface/datasets/pull/3029",
"diff_url": "https://github.com/huggingface/datasets/pull/3029.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3029.patch",
"merged_at": "2021-10-05T14:56:45"
}
| 3,029
| true
|
Properly install ruamel-yaml for windows CI
| null |
https://github.com/huggingface/datasets/pull/3028
|
[
"@lhoestq I would say this does not \"properly\" install `ruamel-yaml`, but the contrary, you overwrite the previous version without desinstalling it first.\r\n\r\nAccording to `pip` docs:\r\n> This can break your system if the existing package is of a different version or was installed with a different package manager!\r\n\r\nNote that our case fulfills both conditions:\r\n- the installing version (`0.17.16`) is different from the existing one (`0.15.87`)\r\n- you are installing using `pip` (`setuptools`), whereas the exisitng version was installed using `distutils`\r\n\r\nThat is why I did not fix the issue this way, made a hotfix pinning `huggingface_hub` (#3025), while looking for a permanent solution for the issue.",
"Yea I did this because we need the latest version of `huggingface_hub` for #2986 and because I didn't want to ssh to the windows worker x)\r\nMaybe it can be fixed by installing it with conda - let me try",
"Oh yea it may not work since it was first installed with distutils"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3028",
"html_url": "https://github.com/huggingface/datasets/pull/3028",
"diff_url": "https://github.com/huggingface/datasets/pull/3028.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3028.patch",
"merged_at": "2021-10-05T11:51:22"
}
| 3,028
| true
|
Resolve data_files by split name
|
This issue is about discussing the default behavior when someone loads a dataset that consists in data files. For example:
```python
load_dataset("lhoestq/demo1")
```
should return two splits "train" and "test" since the dataset repostiory is like
```
data/
├── train.csv
└── test.csv
```
Currently it returns only one split "train" which contains the data of both files
I started playing with this idea on this branch btw: `resolve-data_files-by-split-name`
Basically the idea is that if you named you data files after split names then the default pattern is
```python
{
"train": ["*train*"],
"test": ["*test*"],
"validation": ["*dev*", "valid"],
}
```
otherwise it's
```python
{
"train": ["*"]
}
```
Let me know what you think !
cc @albertvillanova @LysandreJik @vblagoje
|
https://github.com/huggingface/datasets/issues/3027
|
[
"Awesome @lhoestq I like the proposal and it works great on my JSON community dataset. Here is the [log](https://gist.github.com/vblagoje/714babc325bcbdd5de579fd8e1648892). ",
"From my discussion with @borisdayma it would be more general the files match if their paths contains the split name - not only if the filename contains the split name. For example for a dataset like this:\r\n```\r\ntrain/\r\n└── data.csv\r\ntest/\r\n└── data.csv\r\n```\r\n\r\nBut IMO the default should be \r\n```\r\ndata/\r\n├── train.csv\r\n└── test.csv\r\n```\r\nbecause it allows people to have other directories if they have different subsets of their data (different configurations, not splits)",
"I just created a PR for this at https://github.com/huggingface/datasets/pull/3221, let me know what you think :)"
] | null | 3,027
| false
|
added arxiv paper inswiss_judgment_prediction dataset card
| null |
https://github.com/huggingface/datasets/pull/3026
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3026",
"html_url": "https://github.com/huggingface/datasets/pull/3026",
"diff_url": "https://github.com/huggingface/datasets/pull/3026.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3026.patch",
"merged_at": "2021-10-08T16:01:24"
}
| 3,026
| true
|
Fix Windows test suite
|
Try a hotfix to restore Windows test suite.
Fix #3024.
|
https://github.com/huggingface/datasets/pull/3025
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3025",
"html_url": "https://github.com/huggingface/datasets/pull/3025",
"diff_url": "https://github.com/huggingface/datasets/pull/3025.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3025.patch",
"merged_at": "2021-10-05T09:58:27"
}
| 3,025
| true
|
Windows test suite fails
|
## Describe the bug
There is an error during installation of tests dependencies for Windows: https://app.circleci.com/pipelines/github/huggingface/datasets/7981/workflows/9b6a0114-2b8e-4069-94e5-e844dbbdba4e/jobs/49206
```
ERROR: Cannot uninstall 'ruamel-yaml'. It is a distutils installed project and thus we cannot accurately determine which files belong to it which would lead to only a partial uninstall.
```
|
https://github.com/huggingface/datasets/issues/3024
|
[] | null | 3,024
| false
|
Fix typo
| null |
https://github.com/huggingface/datasets/pull/3023
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3023",
"html_url": "https://github.com/huggingface/datasets/pull/3023",
"diff_url": "https://github.com/huggingface/datasets/pull/3023.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3023.patch",
"merged_at": "2021-10-05T11:56:55"
}
| 3,023
| true
|
MeDAL dataset: Add further description and update download URL
|
Added more details in the following sections:
* Dataset Structure
* Data Instances
* Data Splits
* Source Data
* Annotations
* Discussions of Biases
* LIcensing Information
|
https://github.com/huggingface/datasets/pull/3022
|
[
"@lhoestq I'm a bit confused by the error message. I haven't touched the YAML code at all - do you have any insight on that?",
"I just added the missing `pretty_name` tag in the YAML - sorry about that ;)",
"Thanks! Seems like it did the trick since the tests are passing. Let me know if there's anything else I can do in this PR!",
"It's all good thank you :)\r\n\r\nmerging !"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3022",
"html_url": "https://github.com/huggingface/datasets/pull/3022",
"diff_url": "https://github.com/huggingface/datasets/pull/3022.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3022.patch",
"merged_at": "2021-10-13T09:03:09"
}
| 3,022
| true
|
Support loading dataset from multiple zipped CSV data files
|
Fix partially #3018.
CC: @lewtun
|
https://github.com/huggingface/datasets/pull/3021
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3021",
"html_url": "https://github.com/huggingface/datasets/pull/3021",
"diff_url": "https://github.com/huggingface/datasets/pull/3021.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3021.patch",
"merged_at": "2021-10-06T08:36:45"
}
| 3,021
| true
|
Add a metric for the MATH dataset (competition_math).
|
This metric computes accuracy for the MATH dataset (https://arxiv.org/abs/2103.03874) after canonicalizing the prediction and the reference (e.g., converting "1/2" to "\\\\frac{1}{2}").
|
https://github.com/huggingface/datasets/pull/3020
|
[
"I believe the only failed test related to this PR is tests/test_metric_common.py::LocalMetricTest::test_load_metric_competition_math. It gives the following error:\r\n\r\nImportError: To be able to use this dataset, you need to install the following dependencies['math_equivalence'] using 'pip install git+https://github.com/hendrycks/math.git' for instance'\r\n\r\nIt fails along with (these fail with ImportError as well):\r\ntest_load_metric_bertscore\r\ntest_load_metric_bleurt\r\ntest_load_metric_comet\r\ntest_load_metric_coval\r\n\r\nLet me know if there is anything I need to change.",
"Hi ! The script looks all good thanks :)\r\n\r\nTo fix the CI you just need to merge `master` into your branch\r\n```\r\ngit fetch upstream/master\r\ngit merge upstream/master\r\n```\r\n\r\nThen you also need to add `math_equivalence` to the list of git packages installed for the tests in `additional-tests-requirements.txt`\r\nhttps://github.com/huggingface/datasets/blob/ba831e4bcd175ae3d52afbf7d12c4f625bf541b0/additional-tests-requirements.txt#L1-L3",
"I ran:\r\n\r\ngit fetch upstream\r\ngit merge upstream/master\r\n\r\nAnd I also added math_equivalence to the list of git packages installed for the tests in additional-tests-requirements.txt\r\n\r\ntests/test_metric_common.py fails with the same errors as before. tests/test_dataset_cards.py also fails, but it doesn't look related to this PR (it's an issue datasets/ami/README.md).",
"@lhoestq Anything else I can do? I re-merged again and am getting the same test failures as described in the previous comment."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3020",
"html_url": "https://github.com/huggingface/datasets/pull/3020",
"diff_url": "https://github.com/huggingface/datasets/pull/3020.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3020.patch",
"merged_at": "2021-10-22T10:29:31"
}
| 3,020
| true
|
Fix filter leaking
|
If filter is called after using a first transform `shuffle`, `select`, `shard`, `train_test_split`, or `filter`, then it could not work as expected and return examples from before the first transform. This is because the indices mapping was not taken into account when saving the indices to keep when doing the filtering
Affected versions: 1.12.0 and 1.12.1
This should fix #3010
|
https://github.com/huggingface/datasets/pull/3019
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3019",
"html_url": "https://github.com/huggingface/datasets/pull/3019",
"diff_url": "https://github.com/huggingface/datasets/pull/3019.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3019.patch",
"merged_at": "2021-10-05T08:33:07"
}
| 3,019
| true
|
Support multiple zipped CSV data files
|
As requested by @lewtun, support loading multiple zipped CSV data files.
```python
from datasets import load_dataset
url = "https://domain.org/filename.zip"
data_files = {"train": "train_filename.csv", "test": "test_filename.csv"}
dataset = load_dataset("csv", data_dir=url, data_files=data_files)
```
|
https://github.com/huggingface/datasets/issues/3018
|
[
"@lhoestq I would like to draw your attention to the proposed API by @lewtun, using `data_dir` to pass the ZIP URL.\r\n\r\nI'm not totally convinced with this... What do you think?\r\n\r\nMaybe we could discuss other approaches...\r\n\r\nOne brainstorming idea: what about using URL chaining with the hop operator in `data_files`?",
"`data_dir` is currently exclusively used for manually downloaded data.\r\n\r\nMaybe we can have an API that only uses data_files as you are suggesting, using URL chaining ?\r\n```python\r\nfrom datasets import load_dataset\r\n\r\nurl = \"https://domain.org/filename.zip\"\r\ndata_files = {\"train\": \"zip://train_filename.csv::\" + url, \"test\": \"zip://test_filename.csv::\" + url}\r\ndataset = load_dataset(\"csv\", data_files=data_files)\r\n```\r\n\r\nURL chaining is used by `fsspec` to get access to files in nested filesystems of any kind. Since `fsspec` is being used by `pandas`, `dask` and also extensively by `datasets` I think it would be nice to use it here too",
"URL chaining sounds super nice to me! And it's also a nice way to leverage the same concepts we currently have in the docs around `fsspec` :)"
] | null | 3,018
| false
|
Remove unused parameter in xdirname
|
Minor fix to remove unused args `*p` in `xdirname`.
|
https://github.com/huggingface/datasets/pull/3017
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3017",
"html_url": "https://github.com/huggingface/datasets/pull/3017",
"diff_url": "https://github.com/huggingface/datasets/pull/3017.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3017.patch",
"merged_at": "2021-10-05T11:37:00"
}
| 3,017
| true
|
Fix Windows paths in LJ Speech dataset
|
Minor fix in LJ Speech dataset for Windows pathname component separator.
Related to #1878.
|
https://github.com/huggingface/datasets/pull/3016
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3016",
"html_url": "https://github.com/huggingface/datasets/pull/3016",
"diff_url": "https://github.com/huggingface/datasets/pull/3016.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3016.patch",
"merged_at": "2021-10-04T15:23:04"
}
| 3,016
| true
|
Extend support for streaming datasets that use glob.glob
|
This PR extends the support in streaming mode for datasets that use `glob`, by patching the function `glob.glob`.
Related to #2880, #2876, #2874
|
https://github.com/huggingface/datasets/pull/3015
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3015",
"html_url": "https://github.com/huggingface/datasets/pull/3015",
"diff_url": "https://github.com/huggingface/datasets/pull/3015.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3015.patch",
"merged_at": "2021-10-05T13:46:38"
}
| 3,015
| true
|
Fix Windows path in MATH dataset
|
Minor fix in MATH dataset for Windows pathname component separator.
Related to #2982.
|
https://github.com/huggingface/datasets/pull/3014
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3014",
"html_url": "https://github.com/huggingface/datasets/pull/3014",
"diff_url": "https://github.com/huggingface/datasets/pull/3014.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3014.patch",
"merged_at": "2021-10-04T12:46:44"
}
| 3,014
| true
|
Improve `get_dataset_infos`?
|
Using the dedicated function `get_dataset_infos` on a dataset that has no dataset-info.json file returns an empty info:
```
>>> from datasets import get_dataset_infos
>>> get_dataset_infos('wit')
{}
```
While it's totally possible to get it (regenerate it) with:
```
>>> from datasets import load_dataset_builder
>>> builder = load_dataset_builder('wit')
>>> builder.info
DatasetInfo(description='Wikipedia-based Image Text (WIT) Dataset is a large multimodal multilingual dataset. WIT is composed of a curated set\n of 37.6 million entity rich image-text examples with 11.5 million unique images across 108 Wikipedia languages. Its\n size enables WIT to be used as a pretraining dataset for multimodal machine learning models.\n', citation='@article{srinivasan2021wit,\n title={WIT: Wikipedia-based Image Text Dataset for Multimodal Multilingual Machine Learning},\n author={Srinivasan, Krishna and Raman, Karthik and Chen, Jiecao and Bendersky, Michael and Najork, Marc},\n journal={arXiv preprint arXiv:2103.01913},\n year={2021}\n}\n', homepage='https://github.com/google-research-datasets/wit', license='', features={'b64_bytes': Value(dtype='string', id=None), 'embedding': Sequence(feature=Value(dtype='float64', id=None), length=-1, id=None), 'image_url': Value(dtype='string', id=None), 'metadata_url': Value(dtype='string', id=None), 'original_height': Value(dtype='int32', id=None), 'original_width': Value(dtype='int32', id=None), 'mime_type': Value(dtype='string', id=None), 'caption_attribution_description': Value(dtype='string', id=None), 'wit_features': Sequence(feature={'language': Value(dtype='string', id=None), 'page_url': Value(dtype='string', id=None), 'attribution_passes_lang_id': Value(dtype='string', id=None), 'caption_alt_text_description': Value(dtype='string', id=None), 'caption_reference_description': Value(dtype='string', id=None), 'caption_title_and_reference_description': Value(dtype='string', id=None), 'context_page_description': Value(dtype='string', id=None), 'context_section_description': Value(dtype='string', id=None), 'hierarchical_section_title': Value(dtype='string', id=None), 'is_main_image': Value(dtype='string', id=None), 'page_changed_recently': Value(dtype='string', id=None), 'page_title': Value(dtype='string', id=None), 'section_title': Value(dtype='string', id=None)}, length=-1, id=None)}, post_processed=None, supervised_keys=None, task_templates=None, builder_name='wit', config_name='default', version=0.0.0, splits=None, download_checksums=None, download_size=None, post_processing_size=None, dataset_size=None, size_in_bytes=None)
```
Should we test if info is empty, and in that case regenerate it? Or always generate it?
|
https://github.com/huggingface/datasets/issues/3013
|
[
"To keeps things simple maybe we should use `load_dataset_builder` in `get_dataset_infos`.\r\n`load_dataset_builder` instantiates a builder and runs the _infos() method in order to give you the most up-to-date infos, even if the dataset_infos.json is outdated or missing."
] | null | 3,013
| false
|
Replace item with float in metrics
|
As pointed out by @mariosasko in #3001, calling `float()` instad of `.item()` is faster.
Moreover, it might avoid potential issues if any of the third-party functions eventually returns a `float` instead of an `np.float64`.
Related to #3001.
|
https://github.com/huggingface/datasets/pull/3012
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3012",
"html_url": "https://github.com/huggingface/datasets/pull/3012",
"diff_url": "https://github.com/huggingface/datasets/pull/3012.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3012.patch",
"merged_at": "2021-10-04T11:30:33"
}
| 3,012
| true
|
load_dataset_builder should error if "name" does not exist?
|
```
import datasets as ds
builder = ds.load_dataset_builder('sent_comp', name="doesnotexist")
builder.info.config_name
```
returns
```
'doesnotexist'
```
Shouldn't it raise an error instead?
For this dataset, the only valid values for `name` should be: `"default"` or `None` (ie. argument not passed)
|
https://github.com/huggingface/datasets/issues/3011
|
[
"Yes I think it should raise an error. Currently it looks like it instantiates a custom configuration with the name given by the user:\r\nhttps://github.com/huggingface/datasets/blob/ba27ce33bf568374cf23a07669fdd875b5718bc2/src/datasets/builder.py#L391-L397"
] | null | 3,011
| false
|
Chain filtering is leaking
|
## Describe the bug
As there's no support for lists within dataset fields, I convert my lists to json-string format. However, the bug described is occurring even when the data format is 'string'.
These samples show that filtering behavior diverges from what's expected when chaining filterings.
On sample 2 the second filtering leads to "leaking" of data that should've been filtered on the first filtering into the results.
## Steps to reproduce the bug
Sample 1:
```python
import datasets
import json
items = [[1, 2], [3], [4]]
jsoned_items = map(json.dumps, [[1, 2], [3], [4]])
ds = datasets.Dataset.from_dict({'a': jsoned_items})
print(list(ds))
# > Prints: [{'a': '[1, 2]'}, {'a': '[3]'}, {'a': '[4]'}] as expected
filtered = ds
# get all lists that are shorter than 2
filtered = filtered.filter(lambda x: len(json.loads(x['a'])) < 2, load_from_cache_file=False)
print(list(filtered))
# > Prints: [{'a': '[3]'}, {'a': '[4]'}] as expected
# get all lists, which have a value bigger than 3 on its zero index
filtered = filtered.filter(lambda x: json.loads(x['a'])[0] > 3, load_from_cache_file=False)
print(list(filtered))
# > Should be: [{'a': [4]}]
# > Prints: [{'a': [3]}]
```
Sample 2:
```python
import datasets
import json
items = [[1, 2], [3], [4]]
jsoned_items = map(json.dumps, [[1, 2], [3], [4]])
ds = datasets.Dataset.from_dict({'a': jsoned_items})
print(list(ds))
# > Prints: [{'a': '[1, 2]'}, {'a': '[3]'}, {'a': '[4]'}]
filtered = ds
# get all lists, which have a value bigger than 3 on its zero index
filtered = filtered.filter(lambda x: json.loads(x['a'])[0] > 3, load_from_cache_file=False)
print(list(filtered))
# > Prints: [{'a': '[4]'}] as expected
# get all lists that are shorter than 2
filtered = filtered.filter(lambda x: len(json.loads(x['a'])) < 2, load_from_cache_file=False)
print(list(filtered))
# > Prints: [{'a': '[1, 2]'}]
# > Should be: [{'a': '[4]'}] (remain intact)
```
## Expected results
Expected and actual results are attached to the code snippets.
## Actual results
Expected and actual results are attached to the code snippets.
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.1
- Platform: Windows-10-10.0.19042-SP0
- Python version: 3.9.7
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/3010
|
[
"### Update:\r\nI wrote a bit cleaner code snippet (without transforming to json) that can expose leaking.\r\n```python\r\nimport datasets\r\nimport json\r\n\r\nitems = ['ab', 'c', 'df']\r\n\r\nds = datasets.Dataset.from_dict({'col': items})\r\nprint(list(ds))\r\n# > Prints: [{'col': 'ab'}, {'col': 'c'}, {'col': 'df'}]\r\n\r\nfiltered = ds\r\n\r\n# get all items that are starting with a character with ascii code bigger than 'a'\r\nfiltered = filtered.filter(lambda x: x['col'][0] > 'a', load_from_cache_file=False)\r\nprint(list(filtered))\r\n# > Prints: [{'col': 'c'}, {'col': 'df'}] as expected\r\n\r\n# get all items that are shorter than 2\r\nfiltered = filtered.filter(lambda x: len(x['col']) < 2, load_from_cache_file=False)\r\nprint(list(filtered))\r\n# > Prints: [{'col': 'ab'}] -> this is a leaked item from the first filter\r\n# > Should be: [{'col': 'c'}]\r\n```",
"Thanks for reporting. I'm looking into it",
"I just pushed a fix ! We'll do a new release soon.\r\nIn the meantime feel free to install `datasets` from source to play with it",
"Thanks, I'm already using it from your branch!"
] | null | 3,010
| false
|
Fix Windows paths in SUPERB benchmark datasets
|
Minor fix in SUPERB benchmark datasets for Windows pathname component separator.
Related to #2884, #2783 and #2619.
|
https://github.com/huggingface/datasets/pull/3009
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3009",
"html_url": "https://github.com/huggingface/datasets/pull/3009",
"diff_url": "https://github.com/huggingface/datasets/pull/3009.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3009.patch",
"merged_at": "2021-10-04T13:43:24"
}
| 3,009
| true
|
Fix precision/recall metrics with None average
|
Related to issue #2979 and PR #2992.
|
https://github.com/huggingface/datasets/pull/3008
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3008",
"html_url": "https://github.com/huggingface/datasets/pull/3008",
"diff_url": "https://github.com/huggingface/datasets/pull/3008.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3008.patch",
"merged_at": "2021-10-04T09:29:36"
}
| 3,008
| true
|
Correct a typo
| null |
https://github.com/huggingface/datasets/pull/3007
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3007",
"html_url": "https://github.com/huggingface/datasets/pull/3007",
"diff_url": "https://github.com/huggingface/datasets/pull/3007.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3007.patch",
"merged_at": "2021-10-04T09:27:57"
}
| 3,007
| true
|
Fix Windows paths in CommonLanguage dataset
|
Minor fix in CommonLanguage dataset for Windows pathname component separator.
Related to #2989.
|
https://github.com/huggingface/datasets/pull/3006
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3006",
"html_url": "https://github.com/huggingface/datasets/pull/3006",
"diff_url": "https://github.com/huggingface/datasets/pull/3006.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3006.patch",
"merged_at": "2021-10-04T09:07:58"
}
| 3,006
| true
|
DatasetDict.filter and Dataset.filter crashes with any "fn_kwargs" argument
|
## Describe the bug
The ".filter" method of DatasetDict or Dataset objects fails when passing any "fn_kwargs" argument
## Steps to reproduce the bug
```python
import datasets
example_dataset = datasets.Dataset.from_dict({"a": {1, 2, 3, 4}})
def filter_value(example, value):
return example['a'] == value
filtered = example_dataset.filter(filter_value, fn_kwargs={'value': 3})
```
## Expected results
`filtered` is a dataset containing {"a": {3}}
## Actual results
> Traceback (most recent call last):
> File "C:\Users\qsemi\Documents\git\nlp_experiments\gpt_celebrity\src\test_faulty_filter.py", line 8, in <module>
> filtered = example_dataset.filter(filter_value, fn_kwargs={'value': 3})
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\arrow_dataset.py", line 185, in wrapper
> out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\fingerprint.py", line 398, in wrapper
> out = func(self, *args, **kwargs)
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\arrow_dataset.py", line 2169, in filter
> indices = self.map(
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\arrow_dataset.py", line 1686, in map
> return self._map_single(
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\arrow_dataset.py", line 185, in wrapper
> out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\fingerprint.py", line 398, in wrapper
> out = func(self, *args, **kwargs)
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\arrow_dataset.py", line 2048, in _map_single
> batch = apply_function_on_filtered_inputs(
> File "C:\Users\qsemi\miniconda3\envs\main\lib\site-packages\datasets\arrow_dataset.py", line 1939, in apply_function_on_filtered_inputs
> function(*fn_args, effective_indices, **fn_kwargs) if with_indices else function(*fn_args, **fn_kwargs)
> TypeError: get_indices_from_mask_function() got an unexpected keyword argument 'value'
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.1
- Platform: Windows-10-10.0.19042-SP0
- Python version: 3.9.7
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/3005
|
[
"Hi @DrMatters, thanks for reporting.\r\n\r\nThis issue was fixed 14 days ago: #2950.\r\n\r\nCurrently, the fix is only in the master branch and will be made available in our next library release.\r\n\r\nIn the meantime, you can incorporate the fix by installing datasets from the master branch:\r\n```shell\r\npip install -U git+ssh://git@github.com/huggingface/datasets.git@master#egg=datasest\r\n```\r\nor\r\n```shell\r\npip install -U git+https://github.com/huggingface/datasets.git@master#egg=datasets\r\n```",
"Thanks, sorry for bothering"
] | null | 3,005
| false
|
LexGLUE: A Benchmark Dataset for Legal Language Understanding in English.
|
Inspired by the recent widespread use of the GLUE multi-task benchmark NLP dataset (Wang et al., 2018), the subsequent more difficult SuperGLUE (Wang et al., 2019), other previous multi-task NLP benchmarks (Conneau and Kiela, 2018; McCann et al., 2018), and similar initiatives in other domains (Peng et al., 2019), we introduce the Legal General Language Understanding Evaluation (LexGLUE) benchmark, a benchmark dataset to evaluate the performance of NLP methods in legal tasks. LexGLUE is based on seven existing legal NLP datasets, selected using criteria largely from SuperGLUE.
As in GLUE and SuperGLUE (Wang et al., 2019b,a), one of our goals is to push towards generic (or ‘foundation’) models that can cope with multiple NLP tasks, in our case legal NLP tasks possibly with limited task-specific fine-tuning. Another goal is to provide a convenient and informative entry point for NLP researchers and practitioners wishing to explore or develop methods for legalNLP. Having these goals in mind, the datasets we include in LexGLUE and the tasks they address have been simplified in several ways to make it easier for newcomers and generic models to address all tasks.
LexGLUE benchmark is accompanied by experimental infrastructure that relies on Hugging Face Transformers library and resides at: https://github.com/coastalcph/lex-glue.
|
https://github.com/huggingface/datasets/pull/3004
|
[
"Please wait until Tuesday. Arxiv pre-print is pending. 🤗 ",
"Hi @lhoestq, I updated the README with the Arxiv publication info and now the tests are not passing.\r\n\r\nIt seems that the error is completely irrelevant to my code:\r\n\r\n```\r\n Attempting uninstall: ruamel.yaml\r\n Found existing installation: ruamel-yaml 0.15.87\r\nERROR: Cannot uninstall 'ruamel-yaml'. It is a distutils installed project and thus we cannot accurately determine which files belong to it which would lead to only a partial uninstall.\r\n```",
"Hi ! Thanks for adding this one. Amazing work :o \r\n\r\nYea you can ignore the `ruamel-yaml` error, it's unrelated to your PR and fixed on `master`",
"Hi @lhoestq, \r\n\r\n- I fetched and merged the upstream master to get the `ruamel-yaml` fix.\r\n- I updated the README.md file including more information for the inputs and labels, while I also turned all tables in HTML format hoping that they will render nicely in the dataset card in the HF website.\r\n- I also simplified the CaseHOLD input, excl. the unused `question` field and the `context` replicas, as suggested.\r\n"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3004",
"html_url": "https://github.com/huggingface/datasets/pull/3004",
"diff_url": "https://github.com/huggingface/datasets/pull/3004.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3004.patch",
"merged_at": "2021-10-13T13:37:01"
}
| 3,004
| true
|
common_language: Fix license in README.md
|
...it's correct elsewhere
|
https://github.com/huggingface/datasets/pull/3003
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3003",
"html_url": "https://github.com/huggingface/datasets/pull/3003",
"diff_url": "https://github.com/huggingface/datasets/pull/3003.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3003.patch",
"merged_at": "2021-10-04T09:27:01"
}
| 3,003
| true
|
Remove a reference to the open Arrow file when deleting a TF dataset created with to_tf_dataset
|
This [comment](https://github.com/huggingface/datasets/issues/2934#issuecomment-922970919) explains the issue. This PR fixes that with a `weakref` callback, and additionally:
* renames `TensorflowDatasetMixIn` to `TensorflowDatasetMixin` for consistency
* correctly indents `TensorflowDatasetMixin`'s docstring
* replaces `tf.data.AUTOTUNE` with `tf.data.experimental.AUTOTUNE` (we support TF>=2.2 according to the [setup.py](https://github.com/huggingface/datasets/blob/fc46bba66ba4f432cc10501c16a677112e13984c/setup.py#L188) and `AUTOTUNE` has been moved to the experimental part of `tf.data` in 1.X if I'm not mistaken)
Fixes #2934
|
https://github.com/huggingface/datasets/pull/3002
|
[
"@lhoestq The test passes even without the try/except block!",
"Hey, I'm a little late because I was caught up in the course work, but I double-checked this and it looks great. Thanks for fixing!"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3002",
"html_url": "https://github.com/huggingface/datasets/pull/3002",
"diff_url": "https://github.com/huggingface/datasets/pull/3002.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3002.patch",
"merged_at": "2021-10-13T09:03:23"
}
| 3,002
| true
|
Fix cast to Python scalar in Matthews Correlation metric
|
This PR is motivated by issue #2964.
The Matthews Correlation metric relies on sklearn's `matthews_corrcoef` function to compute the result. This function returns either `float` or `np.float64` (see the [source](https://github.com/scikit-learn/scikit-learn/blob/844b4be24d20fc42cc13b957374c718956a0db39/sklearn/metrics/_classification.py#L906-L909)). Obviously, calling `.item()` on the float value will fail, so I'm fixing this with the built-in `float()` function, which covers both cases. Surprisingly, on my machine, casting `np.float64` to a Python scalar with `float()` is even faster than with the `.item()` method.
|
https://github.com/huggingface/datasets/pull/3001
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3001",
"html_url": "https://github.com/huggingface/datasets/pull/3001",
"diff_url": "https://github.com/huggingface/datasets/pull/3001.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3001.patch",
"merged_at": "2021-10-04T09:26:12"
}
| 3,001
| true
|
Fix json loader when conversion not implemented
|
Sometimes the arrow json parser fails if the `block_size` is too small and returns an `ArrowNotImplementedError: JSON conversion to struct...` error.
By increasing the block size it makes it work again.
Hopefully it should help with https://github.com/huggingface/datasets/issues/2799
I tried with the file mentioned in the issue and it worked for me
cc @lewtun can you try again from this branch ?
|
https://github.com/huggingface/datasets/pull/3000
|
[
"And we're already at PR number 3,000 ! ^^",
"Thank you so much for fixing this @lhoestq 😍 ! I just tested the branch out and it works like a charm!"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/3000",
"html_url": "https://github.com/huggingface/datasets/pull/3000",
"diff_url": "https://github.com/huggingface/datasets/pull/3000.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/3000.patch",
"merged_at": "2021-10-01T17:54:23"
}
| 3,000
| true
|
Set trivia_qa writer batch size
|
Save some RAM when generating trivia_qa
|
https://github.com/huggingface/datasets/pull/2999
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2999",
"html_url": "https://github.com/huggingface/datasets/pull/2999",
"diff_url": "https://github.com/huggingface/datasets/pull/2999.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2999.patch",
"merged_at": "2021-10-01T16:34:55"
}
| 2,999
| true
|
cannot shuffle dataset loaded from disk
|
## Describe the bug
dataset loaded from disk cannot be shuffled.
## Steps to reproduce the bug
```
my_dataset = load_from_disk('s3://my_file/validate', fs=s3)
sample = my_dataset.select(range(100)).shuffle(seed=1234)
```
## Actual results
```
sample = my_dataset .select(range(100)).shuffle(seed=1234)
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 185, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/site-packages/datasets/fingerprint.py", line 398, in wrapper
out = func(self, *args, **kwargs)
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 2494, in shuffle
new_fingerprint=new_fingerprint,
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 185, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/site-packages/datasets/fingerprint.py", line 398, in wrapper
out = func(self, *args, **kwargs)
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 2303, in select
tmp_file = tempfile.NamedTemporaryFile("wb", dir=os.path.dirname(indices_cache_file_name), delete=False)
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/tempfile.py", line 547, in NamedTemporaryFile
(fd, name) = _mkstemp_inner(dir, prefix, suffix, flags, output_type)
File "/home/ubuntu/anaconda3/envs/pytorch_p37/lib/python3.7/tempfile.py", line 258, in _mkstemp_inner
fd = _os.open(file, flags, 0o600)
FileNotFoundError: [Errno 2] No such file or directory: '/tmp/tmpnnu5uhnx/my_file/validate/tmpy76d70g4'
```
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.1
- Python version: 3.7
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/2998
|
[] | null | 2,998
| false
|
Dataset has incorrect labels
|
The dataset https://huggingface.co/datasets/turkish_product_reviews has incorrect labels - all reviews are labelled with "1" (positive sentiment). None of the reviews is labelled with "0". See screenshot attached:
|
https://github.com/huggingface/datasets/issues/2997
|
[
"Hi @marshmellow77, thanks for reporting.\r\n\r\nThat issue is fixed since `datasets` version 1.9.0 (see 16bc665f2753677c765011ef79c84e55486d4347).\r\n\r\nPlease, update `datasets` with: `pip install -U datasets`",
"Thanks. Please note that the dataset explorer (https://huggingface.co/datasets/viewer/?dataset=turkish_product_reviews) still shows the incorrect state. The sentiment for the first few customer reviews is actually negative and should be labelled with \"0\", see screenshot:\r\n\r\n\r\n\r\n\r\n",
"Thanks @marshmellow77, good catch! I'm transferring this issue to https://github.com/huggingface/datasets-viewer. "
] | null | 2,997
| false
|
Remove all query parameters when extracting protocol
|
Fix `_get_extraction_protocol` to remove all query parameters, like `?raw=true`, `?dl=1`,...
|
https://github.com/huggingface/datasets/pull/2996
|
[
"Beware of cases like: `http://ufal.ms.mff.cuni.cz/umc/005-en-ur/download.php?f=umc005-corpus.zip` or `gzip://bg-cs.xml::https://opus.nlpl.eu/download.php?f=Europarl/v8/xml/bg-cs.xml.gz`. I see these URLs in the errors (https://observablehq.com/@huggingface/quality-assessment-of-datasets-loading?collection=@huggingface/datasets), but not in the `Extraction protocol for file at xxx is not implemented yet` error, so I'm not sure if they would break now or not.\r\n\r\nMaybe: first try to find an extension, and if none, try to remove the `?...` part and retry to find the extension.\r\n\r\nBy the way, here is the list of URLs for errors of this type, with a '?' in the URL:\r\n\r\n```\r\nhttps://dl.orangedox.com/WyaCpL?dl=1\r\nhttps://drive.google.com/u/0/uc?id=0Bz8a_Dbh9QhbaW12WVVZS2drcnM&export=download\r\nhttps://drive.google.com/u/0/uc?id=1-CaP3xHgZxOGjQ3pXC5tr9YnIajmel-t&export=download\r\nhttps://drive.google.com/u/0/uc?id=11EBGHMAswT5JDO60xh7gnZfYjpMQs7h7&export=download\r\nhttps://drive.google.com/u/0/uc?id=13JCCr-IjZK7uhbLXeufptr_AxvsKinVl&export=download\r\nhttps://drive.google.com/u/0/uc?id=13ZyFc2qepAYSg9WIFaeJ9y402gblsl2e&export=download\r\nhttps://drive.google.com/u/0/uc?id=15auwrFAlq52JJ61u7eSfnhT9rZtI5sjk&export=download\r\nhttps://drive.google.com/u/0/uc?id=16OgJ_OrfzUF_i3ftLjFn9kpcyoi7UJeO&export=download\r\nhttps://drive.google.com/u/0/uc?id=1BFYF05rx-DK9Eb5hgoIgd6EcB8zOI-zu&export=download\r\nhttps://drive.google.com/u/0/uc?id=1Cz1Un9p8Xn9IpEMMrg2kXSDt0dnjxc4z&export=download\r\nhttps://drive.google.com/u/0/uc?id=1H7FphKVVCYoH49sUXl79CuztEfJLaKoF&export=download\r\nhttps://drive.google.com/u/0/uc?id=1NAeuWLgYBzLwU5jCdkrtj4_PRUocuvlb&export=download\r\nhttps://drive.google.com/u/0/uc?id=1OletxmPYNkz2ltOr9pyT0b0iBtUWxslh&export=download\r\nhttps://drive.google.com/u/0/uc?id=1OletxmPYNkz2ltOr9pyT0b0iBtUWxslh&export=download/\r\nhttps://drive.google.com/u/0/uc?id=1R1jR4DcH2UEaM1ZwDSRHdfTGvkCNu6NW&export=download\r\nhttps://drive.google.com/u/0/uc?id=1hDHeoFIfQzJec1NgZNXh3CTNbchiIvuG&export=download\r\nhttps://drive.google.com/u/0/uc?id=1wxwqnWGRzwvc_-ugRoFX8BPgpO3Q7sch&export=download\r\nhttps://drive.google.com/u/0/uc?id=1ydsOTvBZXKqcRvXawOuePrJ99slOEbkk&export=download\r\nhttps://drive.google.com/uc?export=download&id=0BwmD_VLjROrfTHk4NFg2SndKcjQ\r\nhttps://drive.google.com/uc?export=download&id=0Bz8a_Dbh9QhbQ2Vic1kxMmZZQ1k\r\nhttps://drive.google.com/uc?export=download&id=0Bz8a_Dbh9QhbZlU4dXhHTFhZQU0\r\nhttps://drive.google.com/uc?export=download&id=0Bz8a_Dbh9Qhbd2JNdDBsQUdocVU\r\nhttps://drive.google.com/uc?export=download&id=1-w-0uqaC6hnRn1F_3XqJEvi09zlcTIhX\r\nhttps://drive.google.com/uc?export=download&id=11wMGqNVSwwk6zUnDaJEgm3qT71kAHeff\r\nhttps://drive.google.com/uc?export=download&id=17FGi8KI9N9SuGe7elM8qU8_3fx4sfgTr\r\nhttps://drive.google.com/uc?export=download&id=1AHUm1-_V9GCtGuDcc8XrMUCJE8B-HHoL\r\nhttps://drive.google.com/uc?export=download&id=1CBrh-9OrSpKmPQBxTK_ji6mq6WTN_U9U\r\nhttps://drive.google.com/uc?export=download&id=1Ev4RqWcPsLI9rgOGAKh-_dFKqcEZ1u-G\r\nhttps://drive.google.com/uc?export=download&id=1GTHUJxxmjLmG2lnF9dwRgIDRFZaOY3-F\r\nhttps://drive.google.com/uc?export=download&id=1GcUN6mytEcOMBBOvjJOQzBmEkc-LdgQg\r\nhttps://drive.google.com/uc?export=download&id=1J3mucMFTWrgAYa3LuBZoLRR3CzzYD3fa\r\nhttps://drive.google.com/uc?export=download&id=1Jjhbal535VVz2ap4v4r_rN1UEHTdLK5P\r\nhttps://drive.google.com/uc?export=download&id=1L7aoUXzHPzyzQ0ns4ApBbYepsjFOtXil\r\nhttps://drive.google.com/uc?export=download&id=1M1M5yIOyjKWGprc3LUeVVwxgKXxgpqxm\r\nhttps://drive.google.com/uc?export=download&id=1Nug7-Sri50mkJL4GrWw6C2ZIbfeU-6Am\r\nhttps://drive.google.com/uc?export=download&id=1PGa8j1_IqxiGTc3SU6NMB38sAzxCPS34\r\nhttps://drive.google.com/uc?export=download&id=1QsV8C5EPJrQl37mwva_5-IJOrCaOi2tH\r\nhttps://drive.google.com/uc?export=download&id=1RsGLINVce-0GsDkCLDuLZmoLuzfmoCuQ\r\nhttps://drive.google.com/uc?export=download&id=1TuWH7uwu6V90QWmZn25qhou1rm97Egmn\r\nhttps://drive.google.com/uc?export=download&id=1U7WdBpd9kJ85S7BbBhWUSiy9NnXrKdO6\r\nhttps://drive.google.com/uc?export=download&id=1USoQ8lJgN8kAWnUnRrupMGrPMLlDVqlV\r\nhttps://drive.google.com/uc?export=download&id=1Uit4Og1pk-br_0UJIO5sdhApyhTuHzqo\r\nhttps://drive.google.com/uc?export=download&id=1Z2ty5hU0tIGRZRDlFQZLO7b5vijRfvo0\r\nhttps://drive.google.com/uc?export=download&id=1ZyFGufe4puX3vjGPbp4xg9Hca3Gwq22g\r\nhttps://drive.google.com/uc?export=download&id=1ZzlIQvw1KNBG97QQCfdatvVrrbeLaM1u\r\nhttps://drive.google.com/uc?export=download&id=1_AckYkinAnhqmRQtGsQgUKAnTHxxX5J0\r\nhttps://drive.google.com/uc?export=download&id=1__EjA6oZsgXQpggPm-h54jZu3kP6Y6zu\r\nhttps://drive.google.com/uc?export=download&id=1aHPVfC5TrlnUjehtagVZoDfq4VccgaNT\r\nhttps://drive.google.com/uc?export=download&id=1cqu_YAgvlyVSzzjcUyP1Cz7q0k8Pw7vN\r\nhttps://drive.google.com/uc?export=download&id=1dUIqVwvoZAtbX_-z5axCoe97XNcFo1No\r\nhttps://drive.google.com/uc?export=download&id=1eTtRs5cUlBP5dXsx-FTAlmXuB6JQi2qj\r\nhttps://drive.google.com/uc?export=download&id=1fUR3MqJ8jTMka6owA0S-Fe6aHmiophc_\r\nhttps://drive.google.com/uc?export=download&id=1ffWfITKFMJeqjT8loC8aiCLRNJpc_XnF\r\nhttps://drive.google.com/uc?export=download&id=1g89WgFHMRbr4QrvA0ngh26PY081Nv3lx\r\nhttps://drive.google.com/uc?export=download&id=1meSNZHxd_0TZLKCRCYGN-Ke3IA5c1qOE\r\nhttps://drive.google.com/uc?export=download&id=1okwGJiOZmTpNRNgJLCnjFF4Q0H1z4l6_\r\nhttps://drive.google.com/uc?export=download&id=1phryJg4FjCFkn0mSCqIOP2-FscAeKGV0\r\nhttps://drive.google.com/uc?export=download&id=1s8NSFT4Kz0caKZ4VybPNzt88F8ZanprY\r\nhttps://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C\r\nhttps://drive.google.com/uc?export=download&id=1ytVZ4AhubFDOEL7o7XrIRIyhU8g9wvKA\r\nhttps://drive.google.com/uc?id=12Uz59TYg_NtxOy7SXraYeXPMRT7oaO7X\r\nhttps://drive.google.com/uc?id=1PGH5H_oW7wUvMw_5xaXvbEN7DFll-wDX\r\nhttps://github.com/MaazAmjad/Datasets-for-Urdu-news/blob/master/Urdu%20Fake%20News%20Dataset.zip?raw=true\r\nhttps://github.com/TevenLeScao/glucose/blob/master/GLUCOSE_training_data.zip?raw=true\r\nhttps://github.com/TevenLeScao/what-time-is-it/blob/master/gutenberg_time_phrases.zip?raw=true\r\nhttps://github.com/aviaefrat/cryptonite/blob/main/data/cryptonite-official-split.zip?raw=true\r\nhttps://github.com/facebookresearch/Imppres/blob/master/dataset/IMPPRES.zip?raw=true\r\nhttps://github.com/ljos/navnkjenner/blob/master/data/bokmaal/no_bokmaal-ud-train.bioes?raw=true\r\nhttps://github.com/ljos/navnkjenner/blob/master/data/nynorsk/no_nynorsk-ud-train.bioes?raw=true\r\nhttps://github.com/ljos/navnkjenner/blob/master/data/samnorsk/no_samnorsk-ud-train.bioes?raw=true\r\nhttps://github.com/mirfan899/Urdu/blob/master/sentiment/imdb_urdu_reviews.csv.tar.gz?raw=true\r\nhttps://github.com/omilab/Neural-Sentiment-Analyzer-for-Modern-Hebrew/blob/master/data/morph_train.tsv?raw=true\r\nhttps://github.com/omilab/Neural-Sentiment-Analyzer-for-Modern-Hebrew/blob/master/data/token_train.tsv?raw=true\r\nhttps://lindat.mff.cuni.cz/repository/xmlui/bitstream/handle/11858/00-097C-0000-0023-625F-0/hindencorp05.plaintext.gz?sequence=3&isAllowed=y\r\nhttps://repo.sadilar.org/bitstream/handle/20.500.12185/299/nchlt_afrikaans_named_entity_annotated_corpus.zip?sequence=3&isAllowed=y\r\nhttps://repo.sadilar.org/bitstream/handle/20.500.12185/312/nchlt_isixhosa_named_entity_annotated_corpus.zip?sequence=3&isAllowed=y\r\nhttps://repo.sadilar.org/bitstream/handle/20.500.12185/319/nchlt_isizulu_named_entity_annotated_corpus.zip?sequence=3&isAllowed=y\r\nhttps://repo.sadilar.org/bitstream/handle/20.500.12185/328/nchlt_sepedi_named_entity_annotated_corpus.zip?sequence=3&isAllowed=y\r\nhttps://repo.sadilar.org/bitstream/handle/20.500.12185/334/nchlt_sesotho_named_entity_annotated_corpus.zip?sequence=3&isAllowed=y\r\nhttps://repo.sadilar.org/bitstream/handle/20.500.12185/341/nchlt_setswana_named_entity_annotated_corpus.zip?sequence=3&isAllowed=y\r\nhttps://repo.sadilar.org/bitstream/handle/20.500.12185/346/nchlt_siswati_named_entity_annotated_corpus.zip?sequence=3&isAllowed=y\r\nhttps://www.dropbox.com/s/tohrsllcfy7rch4/SimpleQuestions_v2.tgz?dl=1\r\nhttps://zenodo.org/record/1043504/files/corpus-webis-tldr-17.zip?download=1\r\nhttps://zenodo.org/record/1489920/files/articles-training-byarticle-20181122.zip?download=1\r\nhttps://zenodo.org/record/1489920/files/articles-training-bypublisher-20181122.zip?download=1\r\nhttps://zenodo.org/record/2787612/files/SICK.zip?download=1\r\nhttps://zenodo.org/record/3553423/files/Swahili%20data.zip?download=1\r\nhttps://zenodo.org/record/3707949/files/tapaco_v1.0.zip?download=1\r\nhttps://zenodo.org/record/4300294/files/train.csv?download=1\r\n```\r\n\r\n",
"Hi @severo, I just saw your comment. Thank you.\r\n\r\nFinally I just swapped the 2 parsings: first I extract extension and then I remove query parameters. 😉 ",
"OK :) Maybe we should add some unit tests to ensure we improve the detection without regressions (it's Friday afternoon, I trust the unit tests more than my analysis of the code)",
"Great! For the tests, I think we should also add some URLs in the form: `http://ufal.ms.mff.cuni.cz/umc/005-en-ur/download.php?f=umc005-corpus.zip` to be sure they are still correctly detected."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2996",
"html_url": "https://github.com/huggingface/datasets/pull/2996",
"diff_url": "https://github.com/huggingface/datasets/pull/2996.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2996.patch",
"merged_at": "2021-10-04T08:48:13"
}
| 2,996
| true
|
Fix trivia_qa unfiltered
|
Fix https://github.com/huggingface/datasets/issues/2993
|
https://github.com/huggingface/datasets/pull/2995
|
[
"CI fails due to missing tags, but they will be added in https://github.com/huggingface/datasets/pull/2949"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2995",
"html_url": "https://github.com/huggingface/datasets/pull/2995",
"diff_url": "https://github.com/huggingface/datasets/pull/2995.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2995.patch",
"merged_at": "2021-10-01T10:04:10"
}
| 2,995
| true
|
Fix loading compressed CSV without streaming
|
When implementing support to stream CSV files (https://github.com/huggingface/datasets/commit/ad489d4597381fc2d12c77841642cbeaecf7a2e0#diff-6f60f8d0552b75be8b3bfd09994480fd60dcd4e7eb08d02f721218c3acdd2782), a regression was introduced preventing loading compressed CSV files in non-streaming mode.
This PR fixes it, allowing loading compressed/uncompressed CSV files in streaming/non-streaming mode.
Fix #2977.
|
https://github.com/huggingface/datasets/pull/2994
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2994",
"html_url": "https://github.com/huggingface/datasets/pull/2994",
"diff_url": "https://github.com/huggingface/datasets/pull/2994.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2994.patch",
"merged_at": "2021-10-01T15:53:15"
}
| 2,994
| true
|
Can't download `trivia_qa/unfiltered`
|
## Describe the bug
For some reason, I can't download `trivia_qa/unfilted`. A file seems to be missing... I am able to see it fine though the viewer tough...
## Steps to reproduce the bug
```python
>>> from datasets import load_dataset
>>> load_dataset("trivia_qa", "unfiltered")
Downloading and preparing dataset trivia_qa/unfiltered (download: 3.07 GiB, generated: 27.23 GiB, post-processed: Unknown size, total: 30.30 GiB) to /gpfsscratch/rech/six/commun/datasets/trivia_qa/unfiltered/1.1.0/9977a5d6f72acfd92f587de052403e8138b43bb0d1ce595016c3baf7e14deba6...
Traceback (most recent call last):
File "/gpfswork/rech/six/commun/modules/datasets_modules/datasets/trivia_qa/9977a5d6f72acfd92f587de052403e8138b43bb0d1ce595016c3baf7e14deba6/trivia_qa.py", line 251, in _add_context
with open(os.path.join(file_dir, fname), encoding="utf-8") as f:
FileNotFoundError: [Errno 2] No such file or directory: '/gpfsscratch/rech/six/commun/datasets/downloads/extracted/9fcb7eddc6afd46fd074af3c5128931dfe4b548f933c925a23847faf4c1995ad/evidence/wikipedia/Peanuts.txt'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/load.py", line 852, in load_dataset
use_auth_token=use_auth_token,
File "/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/builder.py", line 616, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/builder.py", line 693, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/builder.py", line 1107, in _prepare_split
disable=bool(logging.get_verbosity() == logging.NOTSET),
File "/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/tqdm/std.py", line 1133, in __iter__
for obj in iterable:
File "/gpfswork/rech/six/commun/modules/datasets_modules/datasets/trivia_qa/9977a5d6f72acfd92f587de052403e8138b43bb0d1ce595016c3baf7e14deba6/trivia_qa.py", line 303, in _generate_examples
example = parse_example(article)
File "/gpfswork/rech/six/commun/modules/datasets_modules/datasets/trivia_qa/9977a5d6f72acfd92f587de052403e8138b43bb0d1ce595016c3baf7e14deba6/trivia_qa.py", line 274, in parse_example
_add_context(article.get("EntityPages", []), "WikiContext", wiki_dir),
File "/gpfswork/rech/six/commun/modules/datasets_modules/datasets/trivia_qa/9977a5d6f72acfd92f587de052403e8138b43bb0d1ce595016c3baf7e14deba6/trivia_qa.py", line 253, in _add_context
except (IOError, datasets.Value("errors").NotFoundError):
File "<string>", line 5, in __init__
File "/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/features.py", line 265, in __post_init__
self.pa_type = string_to_arrow(self.dtype)
File "/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/features.py", line 134, in string_to_arrow
f"Neither {datasets_dtype} nor {datasets_dtype + '_'} seems to be a pyarrow data type. "
ValueError: Neither errors nor errors_ seems to be a pyarrow data type. Please make sure to use a correct data type, see: https://arrow.apache.org/docs/python/api/datatypes.html#factory-functions
```
## Expected results
I am able to load another subset (`rc`), but unable to load.
I am not sure why the try/except doesn't catch it...
https://github.com/huggingface/datasets/blob/9675a5a1e7b99a86f9c250f6ea5fa5d1e6d5cc7d/datasets/trivia_qa/trivia_qa.py#L253
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.11.0
- Platform: Linux-4.18.0-147.51.2.el8_1.x86_64-x86_64-with-redhat-8.1-Ootpa
- Python version: 3.7.10
- PyArrow version: 3.0.0
|
https://github.com/huggingface/datasets/issues/2993
|
[
"wooo that was fast! thank you @lhoestq !\r\nit is able to process now, though it's ignoring all files and ending up with 0 examples now haha :/\r\n\r\nFor subset \"unfiltered\":\r\n```python\r\n>>> load_dataset(\"trivia_qa\", \"unfiltered\")\r\nDownloading and preparing dataset trivia_qa/unfiltered (download: 3.07 GiB, generated: 27.23 GiB, post-processed: Unknown size, total: 30.30 GiB) to /gpfsscratch/rech/six/commun/datasets/trivia_qa/unfiltered/1.1.0/910043a609bb2bdf62b4874f68e0c24fb648cf81e40a358f4bd54c919d72c9ab...\r\n100%|███████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 1354.53it/s]\r\n100%|█████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 40.60it/s]\r\nTraceback (most recent call last):\r\n File \"<stdin>\", line 1, in <module>\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/load.py\", line 1198, in load_dataset\r\n use_auth_token=use_auth_token,\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/builder.py\", line 647, in download_and_prepare\r\n dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/builder.py\", line 748, in _download_and_prepare\r\n verify_splits(self.info.splits, split_dict)\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/utils/info_utils.py\", line 74, in verify_splits\r\n raise NonMatchingSplitsSizesError(str(bad_splits))\r\ndatasets.utils.info_utils.NonMatchingSplitsSizesError: [{'expected': SplitInfo(name='test', num_bytes=2906575347, num_examples=10832, dataset_name='trivia_qa'), 'recorded': SplitInfo(name='test', num_bytes=0, num_examples=0, dataset_name='trivia_qa')}, {'expected': SplitInfo(name='validation', num_bytes=3038966234, num_examples=11313, dataset_name='trivia_qa'), 'recorded': SplitInfo(name='validation', num_bytes=0, num_examples=0, dataset_name='trivia_qa')}]\r\n```\r\nFor subset \"rc\":\r\n```python\r\n>>> load_dataset(\"trivia_qa\", \"rc\")\r\nDownloading and preparing dataset trivia_qa/rc (download: 2.48 GiB, generated: 14.92 GiB, post-processed: Unknown size, total: 17.40 GiB) to /gpfsscratch/rech/six/commun/datasets/trivia_qa/rc/1.1.0/910043a609bb2bdf62b4874f68e0c24fb648cf81e40a358f4bd54c919d72c9ab...\r\n100%|███████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 3806.08it/s]\r\n100%|█████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 51.57it/s]\r\nTraceback (most recent call last):\r\n File \"<stdin>\", line 1, in <module>\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/load.py\", line 1198, in load_dataset\r\n use_auth_token=use_auth_token,\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/builder.py\", line 647, in download_and_prepare\r\n dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/builder.py\", line 748, in _download_and_prepare\r\n verify_splits(self.info.splits, split_dict)\r\n File \"/gpfswork/rech/six/commun/conda/victor/lib/python3.7/site-packages/datasets/utils/info_utils.py\", line 74, in verify_splits\r\n raise NonMatchingSplitsSizesError(str(bad_splits))\r\ndatasets.utils.info_utils.NonMatchingSplitsSizesError: [{'expected': SplitInfo(name='test', num_bytes=1577814583, num_examples=17210, dataset_name='trivia_qa'), 'recorded': SplitInfo(name='test', num_bytes=0, num_examples=0, dataset_name='trivia_qa')}, {'expected': SplitInfo(name='train', num_bytes=12750976012, num_examples=138384, dataset_name='trivia_qa'), 'recorded': SplitInfo(name='train', num_bytes=0, num_examples=0, dataset_name='trivia_qa')}, {'expected': SplitInfo(name='validation', num_bytes=1688535379, num_examples=18669, dataset_name='trivia_qa'), 'recorded': SplitInfo(name='validation', num_bytes=0, num_examples=0, dataset_name='trivia_qa')}]\r\n```\r\n\r\nCould you look into that when you get a chance?\r\nI wonder if it's not something they changed on the file to download? i couldn't find any information",
"@VictorSanh have you tried passing `download_mode=\"force_redownload\"`?\r\n```python\r\nds = load_dataset(\"trivia_qa\", \"unfiltered\", download_mode=\"force_redownload\")\r\n```",
"I aggressively rmed caches, especially rming the `datasets/downloads/extracted/c3d265fa20d99a147a76e4f5e...` solved the issue.\r\nthank you both!\r\n"
] | null | 2,993
| false
|
Fix f1 metric with None average
|
Fix #2979.
|
https://github.com/huggingface/datasets/pull/2992
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2992",
"html_url": "https://github.com/huggingface/datasets/pull/2992",
"diff_url": "https://github.com/huggingface/datasets/pull/2992.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2992.patch",
"merged_at": "2021-10-01T14:17:38"
}
| 2,992
| true
|
add docmentation for the `Unix style pattern` matching feature that can be leverage for `data_files` into `load_dataset`
|
Unless I'm mistaken, it seems that in the new documentation it is no longer mentioned that you can use Unix style pattern matching in the `data_files` argument of the `load_dataset` method.
This feature was mentioned [here](https://huggingface.co/docs/datasets/loading_datasets.html#from-a-community-dataset-on-the-hugging-face-hub) in the previous documentation.
I'd love to hear your opinion @lhoestq , @albertvillanova and @stevhliu
|
https://github.com/huggingface/datasets/issues/2991
|
[] | null | 2,991
| false
|
Make Dataset.map accept list of np.array
|
Fix #2987.
|
https://github.com/huggingface/datasets/pull/2990
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2990",
"html_url": "https://github.com/huggingface/datasets/pull/2990",
"diff_url": "https://github.com/huggingface/datasets/pull/2990.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2990.patch",
"merged_at": "2021-10-01T13:57:45"
}
| 2,990
| true
|
Add CommonLanguage
|
This PR adds the Common Language dataset (https://zenodo.org/record/5036977)
The dataset is intended for language-identification speech classifiers and is already used by models on the Hub:
* https://huggingface.co/speechbrain/lang-id-commonlanguage_ecapa
* https://huggingface.co/anton-l/wav2vec2-base-langid
cc @patrickvonplaten
|
https://github.com/huggingface/datasets/pull/2989
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2989",
"html_url": "https://github.com/huggingface/datasets/pull/2989",
"diff_url": "https://github.com/huggingface/datasets/pull/2989.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2989.patch",
"merged_at": "2021-10-01T17:00:03"
}
| 2,989
| true
|
IndexError: Invalid key: 14 is out of bounds for size 0
|
## Describe the bug
A clear and concise description of what the bug is.
Hi. I am trying to implement stochastic weighted averaging optimizer with transformer library as described here https://pytorch.org/blog/pytorch-1.6-now-includes-stochastic-weight-averaging/ , for this I am using a run_clm.py codes which is working fine before adding SWA optimizer, the moment I modify the model with `swa_model = AveragedModel(model)` in this script, I am getting the below error, since I am NOT touching the dataloader part, I am confused why this is occurring, I very much appreciate your opinion on this @lhoestq
## Steps to reproduce the bug
```
Traceback (most recent call last):
File "run_clm.py", line 723, in <module>
main()
File "run_clm.py", line 669, in main
train_result = trainer.train(resume_from_checkpoint=checkpoint)
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/transformers/trainer.py", line 1258, in train
for step, inputs in enumerate(epoch_iterator):
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 435, in __next__
data = self._next_data()
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 475, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in <listcomp>
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 1530, in __getitem__
format_kwargs=self._format_kwargs,
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 1517, in _getitem
pa_subtable = query_table(self._data, key, indices=self._indices if self._indices is not None else None)
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/datasets/formatting/formatting.py", line 368, in query_table
_check_valid_index_key(key, size)
File "/user/dara/libs/anaconda3/envs/success/lib/python3.7/site-packages/datasets/formatting/formatting.py", line 311, in _check_valid_index_key
raise IndexError(f"Invalid key: {key} is out of bounds for size {size}")
IndexError: Invalid key: 14 is out of bounds for size 0
```
## Expected results
not getting the index error
## Actual results
Please see the above
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: datasets 1.12.1
- Platform: linux
- Python version: 3.7.11
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/2988
|
[
"Hi ! Could you check the length of the `self.dataset` object (i.e. the Dataset object passed to the data loader) ? It looks like the dataset is empty.\r\nNot sure why the SWA optimizer would cause this though.",
"Any updates on this? \r\nThe same error occurred to me too when running `cardiffnlp/twitter-roberta-base-sentiment` on a custom dataset. This happened when I tried to do `model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3])` without using sagemaker distribution. \r\nPython: 3.6.13\r\ndatasets: 1.6.2",
"Hi @ruisi-su, do you have this issue while using SWA as well, or just data parallel ?\r\n\r\nIf you have a code example to reproduce this issue it would also be helpful",
"@lhoestq I had this issue without SWA. I followed [this](https://github.com/huggingface/notebooks/blob/master/sagemaker/03_distributed_training_data_parallelism/sagemaker-notebook.ipynb) notebook to utilize multiple gpus on the [roberta-base](https://huggingface.co/cardiffnlp/twitter-roberta-base-sentiment) model. This tutorial could only work if I am on `ml.p3.16xlarge`, which I don't have access to. So I tried using just `model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3]` before calling `trainer.fit()`. But maybe this is not the right way to do distributed training. I can provide a code example if that will be more helpful.",
"It might be an issue with old versions of `datasets`, can you try updating `datasets` ?",
"FYI I encountered the exact same error using the latest versions of `datasets`, `transformers` and `pyarrow`, without using any kind of SWA or dataparallel: \r\n\r\n```\r\n# packages in environment at C:\\Users\\zhang\\mambaforge:\r\n#\r\n# Name Version Build Channel\r\ncudatoolkit 11.0.3 h3f58a73_9 https://mirrors.ustc.edu.cn/anaconda/cloud/conda-forge\r\ndatasets 1.17.0 pypi_0 pypi\r\npyarrow 6.0.1 pypi_0 pypi\r\npytorch 1.7.1 py3.9_cuda110_cudnn8_0 pytorch\r\ntornado 6.1 py39hb82d6ee_2 https://mirrors.ustc.edu.cn/anaconda/cloud/conda-forge\r\n```\r\n\r\n```\r\n> python --version\r\n> 3.9.7\r\n```",
"Same error here! Datasets version `1.18.3` freshly updated.\r\n\r\n`IndexError: Invalid key: 90 is out of bounds for size 0`\r\n\r\nMy task is finetuning the model for token classification.\r\n\r\n**Solved**: I make a mistake while updating the dataset during the map, you should check that you return the correct values.\r\n",
"cc @sgugger This probably comes from the `Trainer` removing all the columns of a dataset, do you think we can improve the error message in this case ?",
"The `Trainer` clearly logs when it removes columns in the dataset. I'm not too sure of where the bug appears as I haven't seen a clear reproducer. Happy to display a more helpful error message, but I'd need a reproducer to see what the exact problem is to design the right test and warning :-) ",
"Well, if I can try to suggest how to reproduce, please try by do not returning any updated content in the map function used to tokenize input (e.g., in TokenClassification). I can leave here my wrong version for reference:\r\n\r\n```python\r\ndef preprocess_function(examples):\r\n\r\n text = examples[\"text\"]\r\n \r\n inputs = tokenizer(\r\n text,\r\n max_length=512,\r\n truncation=\"only_second\",\r\n return_offsets_mapping=True,\r\n padding=\"max_length\",\r\n )\r\n\r\n offset_mapping = inputs.pop(\"offset_mapping\")\r\n # ... processing code\r\n\r\n inputs[\"labels\"] = label_ids\r\n #return inputs\r\n \r\ntrain_ds = train_ds.map(preprocess_function, batched=False)\r\ntest_ds = test_ds.map(preprocess_function, batched=False)\r\neval_ds = eval_ds.map(preprocess_function, batched=False)\r\n```\r\n\r\nOf course, returning inputs solved the problem. As suggestion, a possible error message could display \"IndexError: the `key` required by trainer are not found in the dataset\" (just an hypothesis, I think there could be something better). \r\n\r\nPlease tell me if you need more details to reproduce, glad to help!",
"That's the thing though. The `Trainer` has no idea which inputs are required or not since all models can have different kinds of inputs, and it can work for models outside of the Transformers library. I can add a clear error message if I get an empty batch, as this is easy to detect, but that's pretty much it.",
"I think that it could be enough to ease the identification of the problem.",
"Done in [this commit](https://github.com/huggingface/transformers/commit/c87cfd653c4de3d4743a9ae09d749282d94d5829)"
] | null | 2,988
| false
|
ArrowInvalid: Can only convert 1-dimensional array values
|
## Describe the bug
For the ViT and LayoutLMv2 demo notebooks in my [Transformers-Tutorials repo](https://github.com/NielsRogge/Transformers-Tutorials), people reported an ArrowInvalid issue after applying the following function to a Dataset:
```
def preprocess_data(examples):
images = [Image.open(path).convert("RGB") for path in examples['image_path']]
words = examples['words']
boxes = examples['bboxes']
word_labels = examples['ner_tags']
encoded_inputs = processor(images, words, boxes=boxes, word_labels=word_labels,
padding="max_length", truncation=True)
return encoded_inputs
```
```
Full trace:
---------------------------------------------------------------------------
ArrowInvalid Traceback (most recent call last)
<ipython-input-8-0fc3efc6f0c2> in <module>()
27
28 train_dataset = datasets['train'].map(preprocess_data, batched=True, remove_columns=datasets['train'].column_names,
---> 29 features=features)
30 test_dataset = datasets['test'].map(preprocess_data, batched=True, remove_columns=datasets['test'].column_names,
31 features=features)
13 frames
/usr/local/lib/python3.7/dist-packages/datasets/arrow_dataset.py in map(self, function, with_indices, input_columns, batched, batch_size, drop_last_batch, remove_columns, keep_in_memory, load_from_cache_file, cache_file_name, writer_batch_size, features, disable_nullable, fn_kwargs, num_proc, suffix_template, new_fingerprint, desc)
1701 new_fingerprint=new_fingerprint,
1702 disable_tqdm=disable_tqdm,
-> 1703 desc=desc,
1704 )
1705 else:
/usr/local/lib/python3.7/dist-packages/datasets/arrow_dataset.py in wrapper(*args, **kwargs)
183 }
184 # apply actual function
--> 185 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
186 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
187 # re-apply format to the output
/usr/local/lib/python3.7/dist-packages/datasets/fingerprint.py in wrapper(*args, **kwargs)
396 # Call actual function
397
--> 398 out = func(self, *args, **kwargs)
399
400 # Update fingerprint of in-place transforms + update in-place history of transforms
/usr/local/lib/python3.7/dist-packages/datasets/arrow_dataset.py in _map_single(self, function, with_indices, input_columns, batched, batch_size, drop_last_batch, remove_columns, keep_in_memory, load_from_cache_file, cache_file_name, writer_batch_size, features, disable_nullable, fn_kwargs, new_fingerprint, rank, offset, disable_tqdm, desc, cache_only)
2063 writer.write_table(batch)
2064 else:
-> 2065 writer.write_batch(batch)
2066 if update_data and writer is not None:
2067 writer.finalize() # close_stream=bool(buf_writer is None)) # We only close if we are writing in a file
/usr/local/lib/python3.7/dist-packages/datasets/arrow_writer.py in write_batch(self, batch_examples, writer_batch_size)
409 typed_sequence = OptimizedTypedSequence(batch_examples[col], type=col_type, try_type=col_try_type, col=col)
410 typed_sequence_examples[col] = typed_sequence
--> 411 pa_table = pa.Table.from_pydict(typed_sequence_examples)
412 self.write_table(pa_table, writer_batch_size)
413
/usr/local/lib/python3.7/dist-packages/pyarrow/table.pxi in pyarrow.lib.Table.from_pydict()
/usr/local/lib/python3.7/dist-packages/pyarrow/array.pxi in pyarrow.lib.asarray()
/usr/local/lib/python3.7/dist-packages/pyarrow/array.pxi in pyarrow.lib.array()
/usr/local/lib/python3.7/dist-packages/pyarrow/array.pxi in pyarrow.lib._handle_arrow_array_protocol()
/usr/local/lib/python3.7/dist-packages/datasets/arrow_writer.py in __arrow_array__(self, type)
106 storage = numpy_to_pyarrow_listarray(self.data, type=type.value_type)
107 else:
--> 108 storage = pa.array(self.data, type.storage_dtype)
109 out = pa.ExtensionArray.from_storage(type, storage)
110 elif isinstance(self.data, np.ndarray):
/usr/local/lib/python3.7/dist-packages/pyarrow/array.pxi in pyarrow.lib.array()
/usr/local/lib/python3.7/dist-packages/pyarrow/array.pxi in pyarrow.lib._sequence_to_array()
/usr/local/lib/python3.7/dist-packages/pyarrow/error.pxi in pyarrow.lib.pyarrow_internal_check_status()
/usr/local/lib/python3.7/dist-packages/pyarrow/error.pxi in pyarrow.lib.check_status()
ArrowInvalid: Can only convert 1-dimensional array values
```
It can be fixed by adding the following line:
```diff
def preprocess_data(examples):
images = [Image.open(path).convert("RGB") for path in examples['image_path']]
words = examples['words']
boxes = examples['bboxes']
word_labels = examples['ner_tags']
encoded_inputs = processor(images, words, boxes=boxes, word_labels=word_labels,
padding="max_length", truncation=True)
+ encoded_inputs["image"] = np.array(encoded_inputs["image"])
return encoded_inputs
```
However, would be great if this can be fixed within Datasets itself.
|
https://github.com/huggingface/datasets/issues/2987
|
[
"Hi @NielsRogge, thanks for reporting!\r\n\r\nIn `datasets`, we were handling N-dimensional arrays only when passed as an instance of `np.array`, not when passed as a list of `np.array`s.\r\n\r\nI'm fixing it."
] | null | 2,987
| false
|
Refac module factory + avoid etag requests for hub datasets
|
## Refactor the module factory
When trying to extend the `data_files` logic to avoid doing unnecessary ETag requests, I noticed that the module preparation mechanism needed a refactor:
- the function was 600 lines long
- it was not readable
- it contained many different cases that made it complex to maintain
- it was hard to properly test it
- it was hard to extend without breaking anything
The module preparation mechanism is in charge of taking the name of a dataset or a metric given by the user (ex: "squad", "accuracy", "lhoestq/test", "path/to/my/script.py", "path/to/my/data/directory", "json", "csv") and return a module (possibly downloaded from the Hub) that contains the dataset builder or the metric class to use.
### Implementation details
I decided to separate all these use cases into different dataset/metric module factories.
First, the metric module factories:
- **CanonicalMetricModuleFactory**: "accuracy", "rouge", ...
- **LocalMetricModuleFactory**: "path/to/my/metric.py"
Then, the dataset module factories:
- **CanonicalDatasetModuleFactory**: "squad", "glue", ...
- **CommunityDatasetModuleFactoryWithScript**: "lhoestq/test"
- **CommunityDatasetModuleFactoryWithoutScript**: "lhoestq/demo1"
- **PackagedDatasetModuleFactory**: "json", "csv", ...
- **LocalDatasetModuleFactoryWithScript**: "path/to/my/script.py"
- **LocalDatasetModuleFactoryWithoutScript**: "path/to/my/data/directory"
And finally, additional factories when users have no internet:
- **CachedDatasetModuleFactory**
- **CachedMetricModuleFactory**
### Breaking changes
One thing is that I still don't know at what extent we want to keep backward compatibility for `prepare_module`. For now I just kept it (except I removed two parameters) just in case, but it's not used anywhere anymore.
## Avoid etag requests for hub datasets
To do this I added a class `DataFilesDict` that can be hashed to define the cache directory of the dataset.
It contains the usual data files formatted as `{"train": ["train.txt"]}` for example.
But each list of file is a `DataFilesList` that also has a `origin_metadata` attribute that contains metadata about the origin of each file:
- for URLs: it stores the ETags of the files
- for local files: it stores the last modification data
- for files from a Hugging Face repository on the Hub: it stores the pattern (`*`, `*.csv`, "train.txt", etc.) and the commit sha of the repository (so there're no ETag requests !)
This way if any file changes, the hash of the `DataFilesDict` changes too !
You can instantiate a `DataFilesDict` by using patterns for local/remote files or files in a HF repository:
- for local/remote files: `DataFilesDict.from_local_or_remote(patterns)`
- for files in a HF repository: `DataFilesDict.from_hf_repo(patterns, dataset_info)`
Fix #2859
## TODO
Fix the latest test:
- [x] fix the call to dataset_info in offline mode (related to https://github.com/huggingface/huggingface_hub/issues/372)
Add some more tests:
- [x] test all the factories
- [x] test the new data files logic
Other:
- [x] docstrings
- [x] comments
|
https://github.com/huggingface/datasets/pull/2986
|
[
"> One thing is that I still don't know at what extent we want to keep backward compatibility for prepare_module. For now I just kept it (except I removed two parameters) just in case, but it's not used anywhere anymore.\r\n\r\nFYI, various other projects currently use it, thus clearly a major version would be required:\r\n\r\nhttps://github.com/search?q=org%3Ahuggingface+prepare_module&type=code",
"Yea so I kept `prepare_module` and changed it to use all the factories I added, so all the use cases in the link you shared are still working. The only two parameters I removed are minor IMO and were a bit hacky anyway (return_resolved_file_path and return_associated_base_path). I think they were only used internally in `datasets` but let me know if you're aware of a use case I didn't think of.",
"I think I'm done with the tests :) I'll do the comments/docs and then we just wait for https://github.com/huggingface/huggingface_hub/pull/373 to get merged",
"When there's a new release of `huggingface_hub` (probably on monday), it will fix the CI.\r\n\r\nThe PR is ready for review. Let me know if I need to clarify some parts",
"One additional change I did: the tests won't affect the number of downloads on the website anymore. And users can choose to not update the number of downloads with `HF_UPDATE_DOWNLOAD_COUNTS=0`",
"CI failures are simply due to RAM issues with circleci workers.\r\nAnd on windows there is an issue with installing `ruamel.yaml` from the bump of `huggingface_hub` (fixed on master)"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2986",
"html_url": "https://github.com/huggingface/datasets/pull/2986",
"diff_url": "https://github.com/huggingface/datasets/pull/2986.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2986.patch",
"merged_at": "2021-10-11T11:05:51"
}
| 2,986
| true
|
add new dataset kan_hope
|
## Adding a Dataset
- **Name:** *KanHope*
- **Description:** *A code-mixed English-Kannada dataset for Hope speech detection*
- **Task:** *Binary Text Classification*
- **Paper:** *https://arxiv.org/abs/2108.04616*
- **Data:** *https://github.com/adeepH/kan_hope/tree/main/dataset*
- **Motivation:** *The dataset is amongst the very few resources available for code-mixed low-resourced Dravidian languages of India*
|
https://github.com/huggingface/datasets/pull/2985
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2985",
"html_url": "https://github.com/huggingface/datasets/pull/2985",
"diff_url": "https://github.com/huggingface/datasets/pull/2985.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2985.patch",
"merged_at": "2021-10-01T16:55:19"
}
| 2,985
| true
|
Exceeded maximum rows when reading large files
|
## Describe the bug
A clear and concise description of what the bug is.
When using `load_dataset` with json files, if the files are too large, there will be "Exceeded maximum rows" error.
## Steps to reproduce the bug
```python
dataset = load_dataset('json', data_files=data_files) # data files have 3M rows in a single file
```
## Expected results
No error
## Actual results
```
~/anaconda3/envs/python/lib/python3.9/site-packages/datasets/packaged_modules/json/json.py in _generate_tables(self, files)
134 with open(file, encoding="utf-8") as f:
--> 135 dataset = json.load(f)
136 except json.JSONDecodeError:
~/anaconda3/envs/python/lib/python3.9/json/__init__.py in load(fp, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw)
292 """
--> 293 return loads(fp.read(),
294 cls=cls, object_hook=object_hook,
~/anaconda3/envs/python/lib/python3.9/json/__init__.py in loads(s, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw)
345 parse_constant is None and object_pairs_hook is None and not kw):
--> 346 return _default_decoder.decode(s)
347 if cls is None:
~/anaconda3/envs/python/lib/python3.9/json/decoder.py in decode(self, s, _w)
339 if end != len(s):
--> 340 raise JSONDecodeError("Extra data", s, end)
341 return obj
JSONDecodeError: Extra data: line 2 column 1 (char 20321)
During handling of the above exception, another exception occurred:
ArrowInvalid Traceback (most recent call last)
<ipython-input-20-ab3718a6482f> in <module>
----> 1 dataset = load_dataset('json', data_files=data_files)
~/anaconda3/envs/python/lib/python3.9/site-packages/datasets/load.py in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, ignore_verifications, keep_in_memory, save_infos, script_version, use_auth_token, task, streaming, **config_kwargs)
841
842 # Download and prepare data
--> 843 builder_instance.download_and_prepare(
844 download_config=download_config,
845 download_mode=download_mode,
~/anaconda3/envs/python/lib/python3.9/site-packages/datasets/builder.py in download_and_prepare(self, download_config, download_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, **download_and_prepare_kwargs)
606 logger.warning("HF google storage unreachable. Downloading and preparing it from source")
607 if not downloaded_from_gcs:
--> 608 self._download_and_prepare(
609 dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
610 )
~/anaconda3/envs/python/lib/python3.9/site-packages/datasets/builder.py in _download_and_prepare(self, dl_manager, verify_infos, **prepare_split_kwargs)
684 try:
685 # Prepare split will record examples associated to the split
--> 686 self._prepare_split(split_generator, **prepare_split_kwargs)
687 except OSError as e:
688 raise OSError(
~/anaconda3/envs/python/lib/python3.9/site-packages/datasets/builder.py in _prepare_split(self, split_generator)
1153 generator = self._generate_tables(**split_generator.gen_kwargs)
1154 with ArrowWriter(features=self.info.features, path=fpath) as writer:
-> 1155 for key, table in utils.tqdm(
1156 generator, unit=" tables", leave=False, disable=bool(logging.get_verbosity() == logging.NOTSET)
1157 ):
~/anaconda3/envs/python/lib/python3.9/site-packages/datasets/packaged_modules/json/json.py in _generate_tables(self, files)
135 dataset = json.load(f)
136 except json.JSONDecodeError:
--> 137 raise e
138 raise ValueError(
139 f"Not able to read records in the JSON file at {file}. "
~/anaconda3/envs/python/lib/python3.9/site-packages/datasets/packaged_modules/json/json.py in _generate_tables(self, files)
114 while True:
115 try:
--> 116 pa_table = paj.read_json(
117 BytesIO(batch), read_options=paj.ReadOptions(block_size=block_size)
118 )
~/anaconda3/envs/python/lib/python3.9/site-packages/pyarrow/_json.pyx in pyarrow._json.read_json()
~/anaconda3/envs/python/lib/python3.9/site-packages/pyarrow/error.pxi in pyarrow.lib.pyarrow_internal_check_status()
~/anaconda3/envs/python/lib/python3.9/site-packages/pyarrow/error.pxi in pyarrow.lib.check_status()
ArrowInvalid: Exceeded maximum rows
```
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version:
- Platform: Linux
- Python version: 3.9
- PyArrow version: 4.0.1
|
https://github.com/huggingface/datasets/issues/2984
|
[
"Hi @zijwang, thanks for reporting this issue.\r\n\r\nYou did not mention which `datasets` version you are using, but looking at the code in the stack trace, it seems you are using an old version.\r\n\r\nCould you please update `datasets` (`pip install -U datasets`) and check if the problem persists?"
] | null | 2,984
| false
|
added SwissJudgmentPrediction dataset
| null |
https://github.com/huggingface/datasets/pull/2983
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2983",
"html_url": "https://github.com/huggingface/datasets/pull/2983",
"diff_url": "https://github.com/huggingface/datasets/pull/2983.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2983.patch",
"merged_at": "2021-10-01T16:03:05"
}
| 2,983
| true
|
Add the Math Aptitude Test of Heuristics dataset.
| null |
https://github.com/huggingface/datasets/pull/2982
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2982",
"html_url": "https://github.com/huggingface/datasets/pull/2982",
"diff_url": "https://github.com/huggingface/datasets/pull/2982.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2982.patch",
"merged_at": "2021-10-01T12:21:00"
}
| 2,982
| true
|
add wit dataset
|
Resolves #2902 based on conversation there - would also close #2810. Open to suggestions/help 😀
CC @hassiahk @lhoestq @yjernite
|
https://github.com/huggingface/datasets/pull/2981
|
[
"Opening this up - Here's whats left to make it really shine:\r\n- [ ] Update dataset card (this is being worked on by the author)\r\n- [ ] Add `dataset_info.json`. Requires downloading the entire dataset. I believe @lhoestq mentioned he may have a machine he's using for this sort of thing.\r\n\r\nI think both of these _could_ be done in separate PRs, but I'm not picky either way.\r\n\r\nAlso, we may find that users (and/or the authors) would like additional configurations for languages, image only, embeddings only, etc. I think we can wait and see what the community says on this and address in different PR.",
"Finally got around to this PR, sorry for the delay!\r\n\r\nI added some minimal info to the dataset card so we can at least have it on the hub and work more on it later.\r\n\r\nUnfortunately, this re-triggered the tests and it looks like we need a rebase? cc @nateraw ",
"Yea we need to rebase. Also I think it would be best to update the features to all be strings to match the original datasets format. I ran into a lot of issues with incorrect data types deep into the Dataset info Json process because it couldn't cast correctly. \n\nI mentioned this to @mariosasko - he might pick up this PR if I don't have bandwidth to finish soon. ",
"> I think this is a clear occasion to start supporting \"canonical\" datasets:\r\n> - without dataset_infos.json file\r\n> - without dummy data\r\n\r\nI agree we can support datasets without them. However the dataset_infos.json file can at least be **highly encouraged** since it has very nice info about the size of the dataset in MB and in number of examples, and it also helps guaranteeing the integrity of the dataset when it's downloaded (verification of the file checksums and the split sizes).",
"This dataset now lives here: https://huggingface.co/datasets/wikimedia/wit_base. @nateraw @yjernite Thanks for the initial work."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2981",
"html_url": "https://github.com/huggingface/datasets/pull/2981",
"diff_url": "https://github.com/huggingface/datasets/pull/2981.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2981.patch",
"merged_at": null
}
| 2,981
| true
|
OpenSLR 25: ASR data for Amharic, Swahili and Wolof
|
## Adding a Dataset
- **Name:** *SLR25*
- **Description:** *Subset 25 from OpenSLR. Other subsets have been added to https://huggingface.co/datasets/openslr, 25 covers Amharic, Swahili and Wolof data*
- **Paper:** *https://www.openslr.org/25/ has citations for each of the three subsubsets. *
- **Data:** *Currently the three links to the .tar.bz2 files can be found a thttps://www.openslr.org/25/*
- **Motivation:** *Increase ASR data for underrepresented African languages. Also, other subsets of OpenSLR speech recognition have been uploaded, so this would be easy.*
https://github.com/huggingface/datasets/blob/master/datasets/openslr/openslr.py already has been created for various other OpenSLR subsets, this should be relatively straightforward to do.
|
https://github.com/huggingface/datasets/issues/2980
|
[
"Whoever handles this just needs to: \r\n\r\n- [ ] fork the HuggingFace Datasets repo\r\n- [ ] update the [existing dataset script](https://github.com/huggingface/datasets/blob/master/datasets/openslr/openslr.py) to add SLR25. Lots of copypasting from other sections of the script should make that easy. \r\nAmharic URL: https://www.openslr.org/resources/25/data_readspeech_am.tar.bz2. \r\nSwahili URL: https://www.openslr.org/resources/25/data_broadcastnews_sw.tar.bz2, \r\nWolof URL: https://www.openslr.org/resources/25/data_readspeech_wo.tar.bz2\r\n- [ ] update the [data card](https://github.com/huggingface/datasets/blob/master/datasets/openslr/README.md) to include information about SLR25. There's lots of other examples to draw from. \r\n- [ ] add the appropriate language tags to the data card as well. https://www.w3.org/International/questions/qa-choosing-language-tags, or just use `sw`, `am`, and `wo` for consistency. \r\n- [ ] make a pull request to merge your changes back into HuggingFace's repo",
"... also the example in \"use in datasets library\" should be updated. It currently says \r\n\r\nBut you actually have to specify a subset, e.g. \r\n```python\r\ndataset = load_dataset(\"openslr\", \"SLR32\")\r\n```",
"\r\n\r\n"
] | null | 2,980
| false
|
ValueError when computing f1 metric with average None
|
## Describe the bug
When I try to compute the f1 score for each class in a multiclass classification problem, I get a ValueError. The same happens with recall and precision. I traced the error to the `.item()` in these scripts, which is probably there for the other averages. E.g. from f1.py:
```python
return {
"f1": f1_score(
references,
predictions,
labels=labels,
pos_label=pos_label,
average=average,
sample_weight=sample_weight,
).item(),
}
```
Since the result is an array with more than one item, the `.item()` throws the error. I didn't submit a PR because this might be needed for the other averages, I'm not very familiar with the library
## Steps to reproduce the bug
```python
from datasets import load_metric
metric = load_metric("f1")
metric.add_batch(predictions=[2,34,1,34,1,2,3], references=[23,52,1,3,523,5,8])
metric.compute(average=None)
```
## Expected results
`array([0.66666667, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. ])`
## Actual results
ValueError: can only convert an array of size 1 to a Python scalar
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.1
- Platform: Windows-10-10.0.19041-SP0
- Python version: 3.9.5
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/2979
|
[
"Hi @asofiaoliveira, thanks for reporting.\r\n\r\nI'm fixing it."
] | null | 2,979
| false
|
Run CI tests against non-production server
|
Currently, the CI test suite performs requests to the HF production server.
As discussed with @elishowk, we should refactor our tests to use the HF staging server instead, like `huggingface_hub` and `transformers`.
|
https://github.com/huggingface/datasets/issues/2978
|
[
"Hey @albertvillanova could you provide more context, including extracts from the discussion we had ?\r\n\r\nLet's ping @Pierrci @julien-c and @n1t0 for their opinion about that",
"@julien-c increased the huggingface.co production workers in order to see if it solve [the 502 you had this morning](https://app.circleci.com/pipelines/github/huggingface/datasets/7843/workflows/fc83fa32-18f5-4dc3-9e2f-ba277ae1af74)\r\n\r\nFor the decision process: be aware that moon-staging does not have persistent repos (they are deleted regularly). as a consequence, **if the moon-staging solution is validated**, you should consider a way to keep the repository that are loaded in tests. These are the ones I found : https://github.com/huggingface/datasets/blob/d488db2f64f312f88f72bbc57a09b7eddb329182/tests/test_load.py and https://github.com/huggingface/datasets/blob/40773111c3e7db8a992fa1c48af32d900a1018d6/tests/test_streaming_download_manager."
] | null | 2,978
| false
|
Impossible to load compressed csv
|
## Describe the bug
It is not possible to load from a compressed csv anymore.
## Steps to reproduce the bug
```python
load_dataset('csv', data_files=['/path/to/csv.bz2'])
```
## Problem and possible solution
This used to work, but the commit that broke it is [this one](https://github.com/huggingface/datasets/commit/ad489d4597381fc2d12c77841642cbeaecf7a2e0#diff-6f60f8d0552b75be8b3bfd09994480fd60dcd4e7eb08d02f721218c3acdd2782).
`pandas` usually gets the compression information from the filename itself (which was previously directly passed). Now, since it gets a file descriptor, it might be good to auto-infer the compression or let the user pass the `compression` kwarg to `load_dataset` (or maybe warn the user if the file ends with a commonly known compression scheme?).
## Environment info
- `datasets` version: 1.10.0 (and over)
- Platform: Linux-5.8.0-45-generic-x86_64-with-glibc2.17
- Python version: 3.8.10
- PyArrow version: 3.0.0
|
https://github.com/huggingface/datasets/issues/2977
|
[
"Hi @Valahaar, thanks for reporting and for your investigation about the source cause.\r\n\r\nYou are right and that commit prevents `pandas` from inferring the compression. On the other hand, @lhoestq did that change to support loading that dataset in streaming mode. \r\n\r\nI'm fixing it."
] | null | 2,977
| false
|
Can't load dataset
|
I'm trying to load a wikitext dataset
```
from datasets import load_dataset
raw_datasets = load_dataset("wikitext")
```
ValueError: Config name is missing.
Please pick one among the available configs: ['wikitext-103-raw-v1', 'wikitext-2-raw-v1', 'wikitext-103-v1', 'wikitext-2-v1']
Example of usage:
`load_dataset('wikitext', 'wikitext-103-raw-v1')`.
If I try
```
from datasets import load_dataset
raw_datasets = load_dataset("wikitext-2-v1")
```
FileNotFoundError: Couldn't find file at https://raw.githubusercontent.com/huggingface/datasets/1.12.1/datasets/wikitext-2-v1/wikitext-2-v1.py
#### Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.12.1
- Platform: Linux-5.4.104+-x86_64-with-Ubuntu-18.04-bionic (colab)
- Python version: 3.7.12
- PyArrow version: 3.0.0
|
https://github.com/huggingface/datasets/issues/2976
|
[
"Hi @mskovalova, \r\n\r\nSome datasets have multiple configurations. Therefore, in order to load them, you have to specify both the *dataset name* and the *configuration name*.\r\n\r\nIn the error message you got, you have a usage example:\r\n- To load the 'wikitext-103-raw-v1' configuration of the 'wikitext' dataset, you should use: \r\n ```python\r\n load_dataset('wikitext', 'wikitext-103-raw-v1')\r\n ```\r\n\r\nIn your case, if you would like to load the 'wikitext-2-v1' configuration of the 'wikitext' dataset, please use:\r\n```python\r\nraw_datasets = load_dataset(\"wikitext\", \"wikitext-2-v1\")\r\n```",
"Hi, if I want to load the dataset from local file, then how to specify the configuration name?"
] | null | 2,976
| false
|
ignore dummy folder and dataset_infos.json
|
Fixes #2877
Added the `dataset_infos.json` to the ignored files list and also added check to ignore files which have parent directory as `dummy`.
Let me know if it is correct. Thanks :)
|
https://github.com/huggingface/datasets/pull/2975
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2975",
"html_url": "https://github.com/huggingface/datasets/pull/2975",
"diff_url": "https://github.com/huggingface/datasets/pull/2975.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2975.patch",
"merged_at": "2021-09-29T09:05:38"
}
| 2,975
| true
|
Actually disable dummy labels by default
|
So I might have just changed the docstring instead of the actual default argument value and not realized.
@lhoestq I'm sorry >.>
|
https://github.com/huggingface/datasets/pull/2974
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2974",
"html_url": "https://github.com/huggingface/datasets/pull/2974",
"diff_url": "https://github.com/huggingface/datasets/pull/2974.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2974.patch",
"merged_at": "2021-09-29T09:04:41"
}
| 2,974
| true
|
Fix JSON metadata of masakhaner dataset
|
Fix #2971.
|
https://github.com/huggingface/datasets/pull/2973
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2973",
"html_url": "https://github.com/huggingface/datasets/pull/2973",
"diff_url": "https://github.com/huggingface/datasets/pull/2973.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2973.patch",
"merged_at": "2021-09-27T12:59:58"
}
| 2,973
| true
|
OSError: Not enough disk space.
|
## Describe the bug
I'm trying to download `natural_questions` dataset from the Internet, and I've specified the cache_dir which locates in a mounted disk and has enough disk space. However, even though the space is enough, the disk space checking function still reports the space of root `/` disk having no enough space.
The file system structure is like below. The root `/` has `115G` disk space available, and the `sda1` is mounted to `/mnt`, which has `1.2T` disk space available:
```
/
/mnt/sda1/path/to/args.dataset_cache_dir
```
## Steps to reproduce the bug
```python
dataset_config = DownloadConfig(
cache_dir=os.path.abspath(args.dataset_cache_dir),
resume_download=True,
)
dataset = load_dataset("natural_questions", download_config=dataset_config)
```
## Expected results
Can download the dataset without an error.
## Actual results
The following error raised:
```
OSError: Not enough disk space. Needed: 134.92 GiB (download: 41.97 GiB, generated: 92.95 GiB, post-processed: Unknown size)
```
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.9.0
- Platform: Ubuntu 18.04
- Python version: 3.8.10
- PyArrow version:
|
https://github.com/huggingface/datasets/issues/2972
|
[
"Maybe we can change the disk space calculating API from `shutil.disk_usage` to `os.statvfs` in UNIX-like system, which can provide correct results.\r\n```\r\nstatvfs = os.statvfs('path')\r\navail_space_bytes = statvfs.f_frsize * statvfs.f_bavail\r\n```",
"Hi @qqaatw, thanks for reporting.\r\n\r\nCould you please try:\r\n```python\r\ndataset = load_dataset(\"natural_questions\", cache_dir=os.path.abspath(args.dataset_cache_dir))\r\n```",
"@albertvillanova it works! Thanks for your suggestion. Is that a bug of `DownloadConfig`?",
"`DownloadConfig` only sets the location to download the files. On the other hand, `cache_dir` sets the location for both downloading and caching the data. You can find more information here: https://huggingface.co/docs/datasets/loading_datasets.html#cache-directory",
"I had encountered the same error when running a command `ds = load_dataset('food101')` in a docker container. The error I got: `OSError: Not enough disk space. Needed: 9.43 GiB (download: 4.65 GiB, generated: 4.77 GiB, post-processed: Unknown size)`\r\n\r\nIn case anyone encountered the same issue, this was my fix:\r\n\r\n```sh\r\n# starting the container (mount project directory onto /app, so that the code and data in my project directory are available in the container)\r\ndocker run -it --rm -v $(pwd):/app my-demo:latest bash\r\n```\r\n\r\n\r\n```python\r\n# other code ...\r\nds = load_dataset('food101', cache_dir=\"/app/data\") # set cache_dir to the absolute path of a directory (e.g. /app/data) that's mounted from the host (MacOS in my case) into the docker container\r\n\r\n# this assumes ./data directory exists in your project folder. If not, create it or point it to any other existing directory where you want to store the cache\r\n```\r\n\r\nThanks @albertvillanova for posting the fix above :-) "
] | null | 2,972
| false
|
masakhaner dataset load problem
|
## Describe the bug
Masakhaner dataset is not loading
## Steps to reproduce the bug
```
from datasets import load_dataset
dataset = load_dataset("masakhaner",'amh')
```
## Expected results
Expected the return of a dataset
## Actual results
```
NonMatchingSplitsSizesError Traceback (most recent call last)
<ipython-input-3-a6abc1161d4c> in <module>()
1 from datasets import load_dataset
2
----> 3 dataset = load_dataset("masakhaner",'amh')
3 frames
/usr/local/lib/python3.7/dist-packages/datasets/utils/info_utils.py
in verify_splits(expected_splits, recorded_splits)
72 ]
73 if len(bad_splits) > 0:
---> 74 raise NonMatchingSplitsSizesError(str(bad_splits))
75 logger.info("All the splits matched successfully.")
76
NonMatchingSplitsSizesError: [{'expected': SplitInfo(name='train', num_bytes=639927, num_examples=1751, dataset_name='masakhaner'), 'recorded': SplitInfo(name='train', num_bytes=639911, num_examples=1750, dataset_name='masakhaner')}, {'expected': SplitInfo(name='validation', num_bytes=92768, num_examples=251, dataset_name='masakhaner'), 'recorded': SplitInfo(name='validation', num_bytes=92753, num_examples=250, dataset_name='masakhaner')}, {'expected': SplitInfo(name='test', num_bytes=184286, num_examples=501, dataset_name='masakhaner'), 'recorded': SplitInfo(name='test', num_bytes=184271, num_examples=500, dataset_name='masakhaner')}]
```
## Environment info
Google Colab
|
https://github.com/huggingface/datasets/issues/2971
|
[
"Thanks for reporting, @ontocord. We are fixing the wrong metadata."
] | null | 2,971
| false
|
Magnet’s
|
## Adding a Dataset
- **Name:** *name of the dataset*
- **Description:** *short description of the dataset (or link to social media or blog post)*
- **Paper:** *link to the dataset paper if available*
- **Data:** *link to the Github repository or current dataset location*
- **Motivation:** *what are some good reasons to have this dataset*
Instructions to add a new dataset can be found [here](https://github.com/huggingface/datasets/blob/master/ADD_NEW_DATASET.md).
|
https://github.com/huggingface/datasets/issues/2970
|
[] | null | 2,970
| false
|
medical-dialog error
|
## Describe the bug
A clear and concise description of what the bug is.
When I attempt to download the huggingface datatset medical_dialog it errors out midway through
## Steps to reproduce the bug
```python
raw_datasets = load_dataset("medical_dialog", "en", split="train", download_mode="force_redownload", data_dir="./Medical-Dialogue-Dataset-English")
```
## Expected results
A clear and concise description of the expected results.
No error
## Actual results
```
3 frames
/usr/local/lib/python3.7/dist-packages/datasets/utils/info_utils.py in verify_splits(expected_splits, recorded_splits)
72 ]
73 if len(bad_splits) > 0:
---> 74 raise NonMatchingSplitsSizesError(str(bad_splits))
75 logger.info("All the splits matched successfully.")
76
NonMatchingSplitsSizesError: [{'expected': SplitInfo(name='train', num_bytes=0, num_examples=0, dataset_name='medical_dialog'), 'recorded': SplitInfo(name='train', num_bytes=295097913, num_examples=229674, dataset_name='medical_dialog')}]
```
Specify the actual results or traceback.
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version: 1.21.1
- Platform: colab
- Python version: colab 3.7
- PyArrow version: N/A
|
https://github.com/huggingface/datasets/issues/2969
|
[
"Hi @smeyerhot, thanks for reporting.\r\n\r\nYou are right: there is an issue with the dataset metadata. I'm fixing it.\r\n\r\nIn the meantime, you can circumvent the issue by passing `ignore_verifications=True`:\r\n```python\r\nraw_datasets = load_dataset(\"medical_dialog\", \"en\", split=\"train\", download_mode=\"force_redownload\", data_dir=\"./Medical-Dialogue-Dataset-English\", ignore_verifications=True)\r\n```"
] | null | 2,969
| false
|
`DatasetDict` cannot be exported to parquet if the splits have different features
|
## Describe the bug
I'm trying to use parquet as a means of serialization for both `Dataset` and `DatasetDict` objects. Using `to_parquet` alongside `from_parquet` or `load_dataset` for a `Dataset` works perfectly.
For `DatasetDict`, I use `to_parquet` on each split to save the parquet files in individual folders representing individual splits. This works too, as long as the splits have identical features. If a split has different features to neighboring splits, then loading the dataset will fail: a single schema is used to load both splits, resulting in a failure to load the second parquet file.
## Steps to reproduce the bug
The following works as expected:
```python
from datasets import load_dataset
ds = load_dataset("lhoestq/custom_squad")
ds['train'].to_parquet("./ds/train/split.parquet")
ds['validation'].to_parquet("./ds/validation/split.parquet")
brand_new_dataset = load_dataset("ds")
```
Modifying a single split to add a new feature ends up in a crash:
```python
from datasets import load_dataset
ds = load_dataset("lhoestq/custom_squad")
def identical_answers(e):
e['identical_answers'] = len(set(e['answers']['text'])) == 1
return e
ds['validation'] = ds['validation'].map(identical_answers)
ds['train'].to_parquet("./ds/train/split.parquet")
ds['validation'].to_parquet("./ds/validation/split.parquet")
brand_new_dataset = load_dataset("ds")
```
```
File "/home/lysandre/.config/JetBrains/PyCharm2021.2/scratches/datasets/upload_dataset.py", line 26, in <module>
brand_new_dataset = load_dataset("ds")
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/load.py", line 1151, in load_dataset
builder_instance.download_and_prepare(
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py", line 642, in download_and_prepare
self._download_and_prepare(
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py", line 732, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py", line 1194, in _prepare_split
writer.write_table(table)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_writer.py", line 428, in write_table
pa_table = pa.Table.from_arrays([pa_table[name] for name in self._schema.names], schema=self._schema)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_writer.py", line 428, in <listcomp>
pa_table = pa.Table.from_arrays([pa_table[name] for name in self._schema.names], schema=self._schema)
File "pyarrow/table.pxi", line 1257, in pyarrow.lib.Table.__getitem__
File "pyarrow/table.pxi", line 1833, in pyarrow.lib.Table.column
File "pyarrow/table.pxi", line 1808, in pyarrow.lib.Table._ensure_integer_index
KeyError: 'Field "identical_answers" does not exist in table schema'
```
It does work, however, to use the `save_to_disk` and `load_from_disk` methods:
```py
from datasets import load_from_disk
ds = load_dataset("lhoestq/custom_squad")
def identical_answers(e):
e['identical_answers'] = len(set(e['answers']['text'])) == 1
return e
ds['validation'] = ds['validation'].map(identical_answers)
ds.save_to_disk("local_path")
brand_new_dataset = load_from_disk("local_path")
```
## Expected results
The saving works correctly - but the loading fails. I would expect either an error when saving or an error-less instantiation of the dataset through the parquet files.
If it's helpful, I've traced a possible patch to the `write_table` method here:
https://github.com/huggingface/datasets/blob/26ff41aa3a642e46489db9e95be1e9a8c4e64bea/src/datasets/arrow_writer.py#L424-L425
The writer is built only if the parquet writer is `None`, but I expect we would want to build a new writer as the table schema has changed. Furthermore, it relies on having the property `update_features` set to `True` in order to update the features:
https://github.com/huggingface/datasets/blob/26ff41aa3a642e46489db9e95be1e9a8c4e64bea/src/datasets/arrow_writer.py#L254-L255
but the `ArrowWriter` is instantiated without that option in the `_prepare_split` method of the `ArrowBasedBuilder`:
https://github.com/huggingface/datasets/blob/26ff41aa3a642e46489db9e95be1e9a8c4e64bea/src/datasets/builder.py#L1190
Updating these two parts to recreate a schema on each split results in an error that is, unfortunately, out of my expertise:
```
File "/home/lysandre/.config/JetBrains/PyCharm2021.2/scratches/datasets/upload_dataset.py", line 27, in <module>
brand_new_dataset = load_dataset("ds")
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/load.py", line 1163, in load_dataset
ds = builder_instance.as_dataset(split=split, ignore_verifications=ignore_verifications, in_memory=keep_in_memory)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py", line 819, in as_dataset
datasets = utils.map_nested(
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/utils/py_utils.py", line 207, in map_nested
mapped = [
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/utils/py_utils.py", line 208, in <listcomp>
_single_map_nested((function, obj, types, None, True))
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/utils/py_utils.py", line 143, in _single_map_nested
return function(data_struct)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py", line 850, in _build_single_dataset
ds = self._as_dataset(
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py", line 920, in _as_dataset
dataset_kwargs = ArrowReader(self._cache_dir, self.info).read(
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py", line 217, in read
return self.read_files(files=files, original_instructions=instructions, in_memory=in_memory)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py", line 238, in read_files
pa_table = self._read_files(files, in_memory=in_memory)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py", line 173, in _read_files
pa_table: Table = self._get_table_from_filename(f_dict, in_memory=in_memory)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py", line 308, in _get_table_from_filename
table = ArrowReader.read_table(filename, in_memory=in_memory)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py", line 327, in read_table
return table_cls.from_file(filename)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/table.py", line 458, in from_file
table = _memory_mapped_arrow_table_from_file(filename)
File "/home/lysandre/Workspaces/Python/datasets/src/datasets/table.py", line 45, in _memory_mapped_arrow_table_from_file
pa_table = opened_stream.read_all()
File "pyarrow/ipc.pxi", line 563, in pyarrow.lib.RecordBatchReader.read_all
File "pyarrow/error.pxi", line 114, in pyarrow.lib.check_status
OSError: Header-type of flatbuffer-encoded Message is not RecordBatch.
```
## Environment info
- `datasets` version: 1.12.2.dev0
- Platform: Linux-5.14.7-arch1-1-x86_64-with-glibc2.33
- Python version: 3.9.7
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/2968
|
[
"This is because you have to specify which split corresponds to what file:\r\n```python\r\ndata_files = {\"train\": \"train/split.parquet\", \"validation\": \"validation/split.parquet\"}\r\nbrand_new_dataset_2 = load_dataset(\"ds\", data_files=data_files)\r\n```\r\n\r\nOtherwise it tries to concatenate the two splits, and it fails because they don't have the same features.\r\n\r\nIt works with save_to_disk/load_from_disk because it also stores json files that contain the information about which files goes into which split",
"Wonderful, thanks for the help!",
"I may be mistaken but I think the following doesn't work either:\r\n\r\n```python\r\nfrom datasets import load_dataset\r\n\r\nds = load_dataset(\"lhoestq/custom_squad\")\r\n\r\n\r\ndef identical_answers(e):\r\n e['identical_answers'] = len(set(e['answers']['text'])) == 1\r\n return e\r\n\r\n\r\nds['validation'] = ds['validation'].map(identical_answers)\r\nds['train'].to_parquet(\"./ds/train/split.parquet\")\r\nds['validation'].to_parquet(\"./ds/validation/split.parquet\")\r\n\r\ndata_files = {\"train\": \"train/split.parquet\", \"validation\": \"validation/split.parquet\"}\r\nbrand_new_dataset_2 = load_dataset(\"ds\", data_files=data_files)\r\n```",
"It works on my side as soon as the directories named `ds/train` and `ds/validation` exist (otherwise it returns a FileNotFoundError). What error are you getting ?",
"Also we may introduce a default mapping for the data files:\r\n```python\r\n{\r\n \"train\": [\"*train*\"],\r\n \"test\": [\"*test*\"],\r\n \"validation\": [\"*dev*\", \"valid\"],\r\n}\r\n```\r\nthis way if you name your files according to the splits you won't have to specify the data_files parameter. What do you think ?\r\n\r\nI moved this discussion to #3027 ",
"I'm getting the following error:\r\n\r\n```\r\nDownloading and preparing dataset custom_squad/plain_text to /home/lysandre/.cache/huggingface/datasets/lhoestq___custom_squad)/plain_text/1.0.0/397916d1ae99584877e0fb4f5b8b6f01e66fcbbeff4d178afb30c933a8d0d93a...\r\n100%|██████████| 2/2 [00:00<00:00, 7760.04it/s]\r\n100%|██████████| 2/2 [00:00<00:00, 2020.38it/s]\r\n 0%| | 0/2 [00:00<?, ?it/s]Traceback (most recent call last):\r\n File \"<input>\", line 1, in <module>\r\n File \"/opt/pycharm-professional/plugins/python/helpers/pydev/_pydev_bundle/pydev_umd.py\", line 198, in runfile\r\n pydev_imports.execfile(filename, global_vars, local_vars) # execute the script\r\n File \"/opt/pycharm-professional/plugins/python/helpers/pydev/_pydev_imps/_pydev_execfile.py\", line 18, in execfile\r\n exec(compile(contents+\"\\n\", file, 'exec'), glob, loc)\r\n File \"/home/lysandre/.config/JetBrains/PyCharm2021.2/scratches/datasets/upload_dataset.py\", line 12, in <module>\r\n ds = load_dataset(\"lhoestq/custom_squad\")\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/load.py\", line 1207, in load_dataset\r\n ds = builder_instance.as_dataset(split=split, ignore_verifications=ignore_verifications, in_memory=keep_in_memory)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py\", line 823, in as_dataset\r\n datasets = utils.map_nested(\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/utils/py_utils.py\", line 207, in map_nested\r\n mapped = [\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/utils/py_utils.py\", line 208, in <listcomp>\r\n _single_map_nested((function, obj, types, None, True))\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/utils/py_utils.py\", line 143, in _single_map_nested\r\n return function(data_struct)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py\", line 854, in _build_single_dataset\r\n ds = self._as_dataset(\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/builder.py\", line 924, in _as_dataset\r\n dataset_kwargs = ArrowReader(self._cache_dir, self.info).read(\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py\", line 217, in read\r\n return self.read_files(files=files, original_instructions=instructions, in_memory=in_memory)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py\", line 238, in read_files\r\n pa_table = self._read_files(files, in_memory=in_memory)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py\", line 173, in _read_files\r\n pa_table: Table = self._get_table_from_filename(f_dict, in_memory=in_memory)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py\", line 308, in _get_table_from_filename\r\n table = ArrowReader.read_table(filename, in_memory=in_memory)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/arrow_reader.py\", line 327, in read_table\r\n return table_cls.from_file(filename)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/table.py\", line 458, in from_file\r\n table = _memory_mapped_arrow_table_from_file(filename)\r\n File \"/home/lysandre/Workspaces/Python/datasets/src/datasets/table.py\", line 45, in _memory_mapped_arrow_table_from_file\r\n pa_table = opened_stream.read_all()\r\n File \"pyarrow/ipc.pxi\", line 563, in pyarrow.lib.RecordBatchReader.read_all\r\n File \"pyarrow/error.pxi\", line 114, in pyarrow.lib.check_status\r\nOSError: Header-type of flatbuffer-encoded Message is not RecordBatch.\r\n```\r\n\r\nTried on current master, after updating latest dependencies and obtained the same result",
"The proposal in #3027 sounds good to me!",
"I just tried again on colab by installing `datasets` from source with pyarrow 3.0.0 and didn't get any error.\r\n\r\nYou error seems to happen when doing\r\n```python\r\nds = load_dataset(\"lhoestq/custom_squad\")\r\n```\r\n\r\nMore specifically it fails when trying to read the arrow file that just got generated. I haven't issues like this before. Can you make sure you have a recent version of `pyarrow` ? Maybe it was an old version that wrote the arrow file and some header was missing.",
"Thank you for your pointer! This seems to have been linked to Python 3.9.7: it works flawlessly with Python 3.8.6. This can be closed, thanks a lot for your help."
] | null | 2,968
| false
|
Adding vision-and-language datasets (e.g., VQA, VCR) to Datasets
|
**Is your feature request related to a problem? Please describe.**
Would you like to add any vision-and-language datasets (e.g., VQA, VCR) to Huggingface Datasets?
**Describe the solution you'd like**
N/A
**Describe alternatives you've considered**
N/A
**Additional context**
This is Da Yin at UCLA. Recently, we have published an EMNLP 2021 paper about geo-diverse visual commonsense reasoning (https://arxiv.org/abs/2109.06860). We propose a new dataset called GD-VCR, a vision-and-language dataset to evaluate how well V&L models perform on scenarios involving geo-location-specific commonsense. We hope to have our V&L dataset incorporated into Huggingface to further promote our project, but I haven't seen much V&L datasets in the current package. Is it possible to add V&L datasets, and if so, how should we prepare for the loading? Thank you very much!
|
https://github.com/huggingface/datasets/issues/2967
|
[] | null | 2,967
| false
|
Upload greek-legal-code dataset
| null |
https://github.com/huggingface/datasets/pull/2966
|
[
"@albertvillanova @lhoestq thank you very much for reviewing! :hugs: \r\n\r\nI 've pushed some updates/changes as requested."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2966",
"html_url": "https://github.com/huggingface/datasets/pull/2966",
"diff_url": "https://github.com/huggingface/datasets/pull/2966.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2966.patch",
"merged_at": "2021-10-13T13:37:30"
}
| 2,966
| true
|
Invalid download URL of WMT17 `zh-en` data
|
## Describe the bug
Partial data (wmt17 zh-en) cannot be downloaded due to an invalid URL.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset('wmt17','zh-en')
```
## Expected results
ConnectionError: Couldn't reach ftp://cwmt-wmt:cwmt-wmt@datasets.nju.edu.cn/parallel/casia2015.zip
|
https://github.com/huggingface/datasets/issues/2965
|
[
"Fixed in the current release. Close this issue."
] | null | 2,965
| false
|
Error when calculating Matthews Correlation Coefficient loaded with `load_metric`
|
## Describe the bug
After loading the metric named "[Matthews Correlation Coefficient](https://huggingface.co/metrics/matthews_correlation)" from `🤗datasets`, the `.compute` method fails with the following exception `AttributeError: 'float' object has no attribute 'item'` (complete stack trace can be provided if required).
## Steps to reproduce the bug
```python
import torch
predictions = torch.ones((10,))
references = torch.zeros((10,))
from datasets import load_metric
METRIC = load_metric("matthews_correlation")
result = METRIC.compute(predictions=predictions, references=references)
```
## Expected results
We should expect a Python `dict` as it follows:
```
{
"matthews_correlation": float()
}
```
as defined in https://github.com/huggingface/datasets/blob/master/metrics/matthews_correlation/matthews_correlation.py, so the fix will imply removing `.item()`, since the value returned by the `scikit-learn` function is not a `torch.Tensor` but a `float`, which means that the `.item()` will fail.
## Actual results
```
Traceback (most recent call last):
File "/home/alvaro.bartolome/XXX/xxx/cli.py", line 59, in main
app()
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/typer/main.py", line 214, in __call__
return get_command(self)(*args, **kwargs)
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/click/core.py", line 1137, in __call__
return self.main(*args, **kwargs)
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/click/core.py", line 1062, in main
rv = self.invoke(ctx)
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/click/core.py", line 1668, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/click/core.py", line 1404, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/click/core.py", line 763, in invoke
return __callback(*args, **kwargs)
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/typer/main.py", line 500, in wrapper
return callback(**use_params) # type: ignore
File "/home/alvaro.bartolome/XXX/xxx/cli.py", line 43, in train
metrics = trainer.evaluate()
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/transformers/trainer.py", line 2051, in evaluate
output = eval_loop(
File "/home/alvaro.bartolome/miniconda3/envs/xxx/lib/python3.9/site-packages/transformers/trainer.py", line 2292, in evaluation_loop
metrics = self.compute_metrics(EvalPrediction(predictions=all_preds, label_ids=all_labels))
File "/home/alvaro.bartolome/XXX/xxx/metrics.py", line 20, in compute_metrics
res = METRIC.compute(predictions=predictions, references=eval_preds.label_ids)
File "/home/alvaro.bartolome/miniconda3/envs/lang/lib/python3.9/site-packages/datasets/metric.py", line 402, in compute
output = self._compute(predictions=predictions, references=references, **kwargs)
File "/home/alvaro.bartolome/.cache/huggingface/modules/datasets_modules/metrics/matthews_correlation/0275f1e9a4d318e3ea8cdd87547ee0d58d894966616052e3d18444ac8ddd2357/matthews_correlation.py", line 88, in _compute
"matthews_correlation": matthews_corrcoef(references, predictions, sample_weight=sample_weight).item(),
AttributeError: 'float' object has no attribute 'item'
```
## Environment info
- `datasets` version: 1.12.1
- Platform: Linux-4.15.0-1113-azure-x86_64-with-glibc2.23
- Python version: 3.9.7
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/2964
|
[
"After some more tests I've realized that this \"issue\" is due to the `numpy.float64` to `float` conversion, but when defining a function named `compute_metrics` as it follows:\r\n\r\n```python\r\ndef compute_metrics(eval_preds):\r\n metric = load_metric(\"matthews_correlation\")\r\n logits, labels = eval_preds\r\n predictions = np.argmax(logits, axis=1)\r\n return metric.compute(predictions=predictions, references=labels)\r\n```\r\n\r\nIt fails when the evaluation metrics are computed in the `Trainer` with the same error code `AttributeError: 'float' object has no attribute 'item'` as the output is not a `numpy.float64`... Maybe I'm doing something wrong, not sure!",
"Ok after some more experiments I've realized that it's an issue from my side, at first I thought it was due to `fp16=True` in `TrainingArguments`, but in the end that may not be the issue, so I'll close this for now and check later, since the mistake is on my side :weary: Sorry for the inconvenience!"
] | null | 2,964
| false
|
raise TypeError( TypeError: Provided `function` which is applied to all elements of table returns a variable of type <class 'list'>. Make sure provided `function` returns a variable of type `dict` to update the dataset or `None` if you are only interested in side effects.
|
## Describe the bug
A clear and concise description of what the bug is.
I am trying to use Dataset to load my file in order to use Bert embeddings model baut when I finished loading using dataset and I want to pass to the tokenizer using the function map; I get the following error : raise TypeError(
TypeError: Provided `function` which is applied to all elements of table returns a variable of type <class 'list'>. Make sure provided `function` returns a variable of type `dict` to update the dataset or `None` if you are only interested in side effects.
I was able to load my file using dataset before but since this morning , I keep getting this erreor.
## Steps to reproduce the bug
```python
# Xtrain, ytrain, filename, len_labels = read_file_2(fic)
# Xtrain, lge_size = get_flaubert_layer(Xtrain, path_to_model_lge)
data_preprocessed = make_new_traindata(Xtrain)
my_dict = {"verbatim": data_preprocessed[1], "label": ytrain} # lemme avec conjonction
dataset = Dataset.from_dict(my_dict)
```
## Expected results
A clear and concise description of the expected results.
## Actual results
Specify the actual results or traceback.
## Environment info
<!-- You can run the command `datasets-cli env` and copy-and-paste its output below. -->
- `datasets` version:
- Platform:
- Python version:
- PyArrow version:
|
https://github.com/huggingface/datasets/issues/2963
|
[] | null | 2,963
| false
|
Enable splits during streaming the dataset
|
## Describe the Problem
I'd like to stream only a specific percentage or part of the dataset.
I want to do splitting when I'm streaming dataset as well.
## Solution
Enabling splits when `streaming = True` as well.
`e.g. dataset = load_dataset('dataset', split='train[:100]', streaming = True)`
## Alternatives
Below is the alternative of doing it.
`dataset = load_dataset("dataset", split='train', streaming = True).take(100)`
|
https://github.com/huggingface/datasets/issues/2962
|
[] | null | 2,962
| false
|
Fix CI doc build
|
Pin `fsspec`.
Before the issue: 'fsspec-2021.8.1', 's3fs-2021.8.1'
Generating the issue: 'fsspec-2021.9.0', 's3fs-0.5.1'
|
https://github.com/huggingface/datasets/pull/2961
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2961",
"html_url": "https://github.com/huggingface/datasets/pull/2961",
"diff_url": "https://github.com/huggingface/datasets/pull/2961.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2961.patch",
"merged_at": "2021-09-24T13:18:07"
}
| 2,961
| true
|
Support pandas 1.3 new `read_csv` parameters
|
Support two new arguments introduced in pandas v1.3.0:
- `encoding_errors`
- `on_bad_lines`
`read_csv` reference: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_csv.html
|
https://github.com/huggingface/datasets/pull/2960
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2960",
"html_url": "https://github.com/huggingface/datasets/pull/2960",
"diff_url": "https://github.com/huggingface/datasets/pull/2960.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2960.patch",
"merged_at": "2021-09-24T11:22:30"
}
| 2,960
| true
|
Added computer vision tasks
|
Added various image processing/computer vision tasks.
|
https://github.com/huggingface/datasets/pull/2959
|
[
"Looks great, thanks ! If the 3d ones are really rare we can remove them for now.\r\n\r\nAnd I can see that `object-detection` and `semantic-segmentation` are both task categories (top-level) and task ids (bottom-level). Maybe there's a way to group them and have less granularity for the task categories. For example `speech-processing` is a high level task category. What do you think ?\r\n\r\nWe can still update the list of tasks later if needed when we have more vision datasets\r\n",
"@lhoestq @osanseviero I used the categories (there were main ones and subcategories) in the paperswithcode, I got rid of some of them that could be too granular. I can put it there if you'd like (I'll wait for your reply before committing it again)",
"We can ignore the ones that are too granular IMO. What we did for audio tasks is to have them all under \"audio-processing\". Maybe we can do the same here for now until we have more comprehensive tasks/applications ?",
"Following the discussion in (private) https://github.com/huggingface/moon-landing/issues/2020, what do you think of aligning the top level tasks list with the model tasks taxonomy ?\r\n\r\n* Image Classification\r\n* Object Detection\r\n* Image Segmentation\r\n* Text-to-Image\r\n* Image-to-Text\r\n",
"I moved it to [a branch](https://github.com/huggingface/datasets/pull/3800) for ease."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2959",
"html_url": "https://github.com/huggingface/datasets/pull/2959",
"diff_url": "https://github.com/huggingface/datasets/pull/2959.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2959.patch",
"merged_at": null
}
| 2,959
| true
|
Add security policy to the project
|
Add security policy to the project, as recommended by GitHub: https://docs.github.com/en/code-security/getting-started/adding-a-security-policy-to-your-repository
Close #2953.
|
https://github.com/huggingface/datasets/pull/2958
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2958",
"html_url": "https://github.com/huggingface/datasets/pull/2958",
"diff_url": "https://github.com/huggingface/datasets/pull/2958.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2958.patch",
"merged_at": "2021-10-21T15:16:43"
}
| 2,958
| true
|
MultiWOZ Dataset NonMatchingChecksumError
|
## Describe the bug
The checksums for the downloaded MultiWOZ dataset and source MultiWOZ dataset aren't matching.
## Steps to reproduce the bug
Both of the below dataset versions yield the checksum error:
```python
from datasets import load_dataset
dataset = load_dataset('multi_woz_v22', 'v2.2')
dataset = load_dataset('multi_woz_v22', 'v2.2_active_only')
```
## Expected results
For the above calls to `load_dataset` to work.
## Actual results
NonMatchingChecksumError. Traceback:
> Traceback (most recent call last):
File "/Users/brady/anaconda3/envs/elysium/lib/python3.8/site-packages/IPython/core/interactiveshell.py", line 3441, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File "<ipython-input-15-4e91280e112e>", line 1, in <module>
dataset = load_dataset('multi_woz_v22', 'v2.2')
File "/Users/brady/anaconda3/envs/elysium/lib/python3.8/site-packages/datasets/load.py", line 847, in load_dataset
builder_instance.download_and_prepare(
File "/Users/brady/anaconda3/envs/elysium/lib/python3.8/site-packages/datasets/builder.py", line 615, in download_and_prepare
self._download_and_prepare(
File "/Users/brady/anaconda3/envs/elysium/lib/python3.8/site-packages/datasets/builder.py", line 675, in _download_and_prepare
verify_checksums(
File "/Users/brady/anaconda3/envs/elysium/lib/python3.8/site-packages/datasets/utils/info_utils.py", line 40, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json']
## Environment info
- `datasets` version: 1.11.0
- Platform: macOS-10.15.7-x86_64-i386-64bit
- Python version: 3.8.10
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/2957
|
[
"Hi Brady! I met the similar issue, it stuck in the downloading stage instead of download anything, maybe it is broken. After I change the downloading from URLs to one url of the [Multiwoz project](https://github.com/budzianowski/multiwoz/archive/44f0f8479f11721831c5591b839ad78827da197b.zip) and use dirs to get separate files, the problems gone."
] | null | 2,957
| false
|
Cache problem in the `load_dataset` method for local compressed file(s)
|
## Describe the bug
Cache problem in the `load_dataset` method: when modifying a compressed file in a local folder `load_dataset` doesn't detect the change and load the previous version.
## Steps to reproduce the bug
To test it directly, I have prepared a [Google Colaboratory notebook](https://colab.research.google.com/drive/11Em_Amoc-aPGhSBIkSHU2AvEh24nVayy?usp=sharing) that shows this behavior.
For this example, I have created a toy dataset at: https://huggingface.co/datasets/SaulLu/toy_struc_dataset
This dataset is composed of two versions:
- v1 on commit `a6beb46` which has a single example `{'id': 1, 'value': {'tag': 'a', 'value': 1}}` in file `train.jsonl.gz`
- v2 on commit `e7935f4` (`main` head) which has a single example `{'attr': 1, 'id': 1, 'value': 'a'}` in file `train.jsonl.gz`
With a terminal, we can start to get the v1 version of the dataset
```bash
git lfs install
git clone https://huggingface.co/datasets/SaulLu/toy_struc_dataset
cd toy_struc_dataset
git checkout a6beb46
```
Then we can load it with python and look at the content:
```python
from datasets import load_dataset
path = "/content/toy_struc_dataset"
dataset = load_dataset(path, data_files={"train": "*.jsonl.gz"})
print(dataset["train"][0])
```
Output
```
{'id': 1, 'value': {'tag': 'a', 'value': 1}} # This is the example in v1
```
With a terminal, we can now start to get the v1 version of the dataset
```bash
git checkout main
```
Then we can load it with python and look at the content:
```python
from datasets import load_dataset
path = "/content/toy_struc_dataset"
dataset = load_dataset(path, data_files={"train": "*.jsonl.gz"})
print(dataset["train"][0])
```
Output
```
{'id': 1, 'value': {'tag': 'a', 'value': 1}} # This is the example in v1 (not v2)
```
## Expected results
The last output should have been
```
{"id":1, "value": "a", "attr": 1} # This is the example in v2
```
## Ideas
As discussed offline with Quentin, if the cache hash was ever sensitive to changes in a compressed file we would probably not have the problem anymore.
This situation leads me to suggest 2 other features:
- to also have an `load_from_cache_file` argument in the "load_dataset" method
- to reorganize the cache so that we can delete the caches related to a dataset (cf issue #ToBeFilledSoon)
And thanks again for this great library :hugs:
## Environment info
- `datasets` version: 1.12.1
- Platform: Linux-5.4.104+-x86_64-with-Ubuntu-18.04-bionic
- Python version: 3.7.12
- PyArrow version: 3.0.0
|
https://github.com/huggingface/datasets/issues/2956
|
[] | null | 2,956
| false
|
Update legacy Python image for CI tests in Linux
|
Instead of legacy, use next-generation convenience images, built from the ground up with CI, efficiency, and determinism in mind. Here are some of the highlights:
- Faster spin-up time - In Docker terminology, these next-gen images will generally have fewer and smaller layers. Using these new images will lead to faster image downloads when a build starts, and a higher likelihood that the image is already cached on the host.
- Improved reliability and stability - The existing legacy convenience images are rebuilt practically every day with potential changes from upstream that we cannot always test fast enough. This leads to frequent breaking changes, which is not the best environment for stable, deterministic builds. Next-gen images will only be rebuilt for security and critical-bugs, leading to more stable and deterministic images.
More info: https://circleci.com/docs/2.0/circleci-images
|
https://github.com/huggingface/datasets/pull/2955
|
[
"There is an exception when running `pip install .[tests]`:\r\n```\r\nProcessing /home/circleci/datasets\r\nCollecting numpy>=1.17 (from datasets==1.12.2.dev0)\r\n Downloading https://files.pythonhosted.org/packages/45/b2/6c7545bb7a38754d63048c7696804a0d947328125d81bf12beaa692c3ae3/numpy-1.19.5-cp36-cp36m-manylinux1_x86_64.whl (13.4MB)\r\n 100% |████████████████████████████████| 13.4MB 3.9MB/s eta 0:00:011\r\n\r\n...\r\n\r\nCollecting faiss-cpu (from datasets==1.12.2.dev0)\r\n Downloading https://files.pythonhosted.org/packages/87/91/bf8ea0d42733cbb04f98d3bf27808e4919ceb5ec71102e21119398a97237/faiss-cpu-1.7.1.post2.tar.gz (41kB)\r\n 100% |████████████████████████████████| 51kB 30.9MB/s ta 0:00:01\r\n Complete output from command python setup.py egg_info:\r\n Traceback (most recent call last):\r\n File \"/home/circleci/.pyenv/versions/3.6.14/lib/python3.6/site-packages/setuptools/sandbox.py\", line 154, in save_modules\r\n yield saved\r\n File \"/home/circleci/.pyenv/versions/3.6.14/lib/python3.6/site-packages/setuptools/sandbox.py\", line 195, in setup_context\r\n yield\r\n File \"/home/circleci/.pyenv/versions/3.6.14/lib/python3.6/site-packages/setuptools/sandbox.py\", line 250, in run_setup\r\n _execfile(setup_script, ns)\r\n File \"/home/circleci/.pyenv/versions/3.6.14/lib/python3.6/site-packages/setuptools/sandbox.py\", line 45, in _execfile\r\n exec(code, globals, locals)\r\n File \"/tmp/easy_install-1pop4blm/numpy-1.21.2/setup.py\", line 34, in <module>\r\n method can be invoked.\r\n RuntimeError: Python version >= 3.7 required.\r\n```\r\n\r\nApparently, `numpy-1.21.2` tries to be installed in the temporary directory `/tmp/easy_install-1pop4blm` instead of the downloaded `numpy-1.19.5` (requirement of `datasets`).\r\n\r\nThis is caused because `pip` downloads the `.tar.gz` (instead of the `.whl`) and tries to build it in a tmp dir."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2955",
"html_url": "https://github.com/huggingface/datasets/pull/2955",
"diff_url": "https://github.com/huggingface/datasets/pull/2955.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2955.patch",
"merged_at": "2021-09-24T10:36:05"
}
| 2,955
| true
|
Run tests in parallel
|
Run CI tests in parallel to speed up the test suite.
Speed up results:
- Linux: from `7m 30s` to `5m 32s`
- Windows: from `13m 52s` to `11m 10s`
|
https://github.com/huggingface/datasets/pull/2954
|
[
"There is a speed up in Windows machines:\r\n- From `13m 52s` to `11m 10s`\r\n\r\nIn Linux machines, some workers crash with error message:\r\n```\r\nOSError: [Errno 12] Cannot allocate memory\r\n```",
"There is also a speed up in Linux machines:\r\n- From `7m 30s` to `5m 32s`"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2954",
"html_url": "https://github.com/huggingface/datasets/pull/2954",
"diff_url": "https://github.com/huggingface/datasets/pull/2954.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2954.patch",
"merged_at": "2021-09-28T06:55:51"
}
| 2,954
| true
|
Trying to get in touch regarding a security issue
|
Hey there!
I'd like to report a security issue but cannot find contact instructions on your repository.
If not a hassle, might you kindly add a `SECURITY.md` file with an email, or another contact method? GitHub [recommends](https://docs.github.com/en/code-security/getting-started/adding-a-security-policy-to-your-repository) this best practice to ensure security issues are responsibly disclosed, and it would serve as a simple instruction for security researchers in the future.
Thank you for your consideration, and I look forward to hearing from you!
(cc @huntr-helper)
|
https://github.com/huggingface/datasets/issues/2953
|
[
"Hi @JamieSlome,\r\n\r\nThanks for reaching out. Yes, you are right: I'm opening a PR to add the `SECURITY.md` file and a contact method.\r\n\r\nIn the meantime, please feel free to report the security issue to: feedback@huggingface.co"
] | null | 2,953
| false
|
Fix missing conda deps
|
`aiohttp` was added as a dependency in #2662 but was missing for the conda build, which causes the 1.12.0 and 1.12.1 to fail.
Fix #2932.
|
https://github.com/huggingface/datasets/pull/2952
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2952",
"html_url": "https://github.com/huggingface/datasets/pull/2952",
"diff_url": "https://github.com/huggingface/datasets/pull/2952.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2952.patch",
"merged_at": "2021-09-21T15:30:44"
}
| 2,952
| true
|
Dummy labels no longer on by default in `to_tf_dataset`
|
After more experimentation, I think I have a way to do things that doesn't depend on adding `dummy_labels` - they were quite a hacky solution anyway!
|
https://github.com/huggingface/datasets/pull/2951
|
[
"@lhoestq Let me make sure we never need it, and if not then I'll remove it entirely in a follow-up PR.",
"Thanks ;) it will be less confusing and easier to maintain to not keep unused hacky features"
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2951",
"html_url": "https://github.com/huggingface/datasets/pull/2951",
"diff_url": "https://github.com/huggingface/datasets/pull/2951.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2951.patch",
"merged_at": "2021-09-21T10:14:32"
}
| 2,951
| true
|
Fix fn kwargs in filter
|
#2836 broke the `fn_kwargs` parameter of `filter`, as mentioned in https://github.com/huggingface/datasets/issues/2927
I fixed that and added a test to make sure it doesn't happen again (for either map or filter)
Fix #2927
|
https://github.com/huggingface/datasets/pull/2950
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2950",
"html_url": "https://github.com/huggingface/datasets/pull/2950",
"diff_url": "https://github.com/huggingface/datasets/pull/2950.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2950.patch",
"merged_at": "2021-09-20T15:28:01"
}
| 2,950
| true
|
Introduce web and wiki config in triviaqa dataset
|
The TriviaQA paper suggests that the two subsets (Wikipedia and Web)
should be treated differently. There are also different leaderboards
for the two sets on CodaLab. For that reason, introduce additional
builder configs in the trivia_qa dataset.
|
https://github.com/huggingface/datasets/pull/2949
|
[
"I just made the dummy data smaller :)\r\nOnce github refreshes the change I think we can merge !",
"Thank you so much for reviewing and accepting my pull request!! :)\r\n\r\nI created these rather large dummy data sets to cover all different cases for the row structure. E.g. in the web configuration, it's possible that a row has evidence from wikipedia (\"EntityPages\") and the web (\"SearchResults\"). But it also might happen that either EntityPages or SearchResults is empty. Probably, I will add this thought to the dataset description in the future.",
"Ok I see ! Yes feel free to mention it in the dataset card, this can be useful.\r\n\r\nFor the dummy data though we can keep the small ones, as the tests are mainly about testing the parsing from the dataset script rather than the actual content of the dataset."
] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2949",
"html_url": "https://github.com/huggingface/datasets/pull/2949",
"diff_url": "https://github.com/huggingface/datasets/pull/2949.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2949.patch",
"merged_at": "2021-10-01T15:39:29"
}
| 2,949
| true
|
Fix minor URL format in scitldr dataset
|
While investigating issue #2918, I found this minor format issues in the URLs (if runned in a Windows machine).
|
https://github.com/huggingface/datasets/pull/2948
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2948",
"html_url": "https://github.com/huggingface/datasets/pull/2948",
"diff_url": "https://github.com/huggingface/datasets/pull/2948.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2948.patch",
"merged_at": "2021-09-20T13:18:28"
}
| 2,948
| true
|
Don't use old, incompatible cache for the new `filter`
|
#2836 changed `Dataset.filter` and the resulting data that are stored in the cache are different and incompatible with the ones of the previous `filter` implementation.
However the caching mechanism wasn't able to differentiate between the old and the new implementation of filter (only the method name was taken into account).
This is an issue because anyone that update `datasets` and re-runs some code that uses `filter` would see an error, because the cache would try to load an incompatible `filter` result.
To fix this I added the notion of versioning for dataset transform in the caching mechanism, and bumped the version of the `filter` implementation to 2.0.0
This way the new `filter` outputs are now considered different from the old ones from the caching point of view.
This should fix #2943
cc @anton-l
|
https://github.com/huggingface/datasets/pull/2947
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2947",
"html_url": "https://github.com/huggingface/datasets/pull/2947",
"diff_url": "https://github.com/huggingface/datasets/pull/2947.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2947.patch",
"merged_at": "2021-09-20T13:43:01"
}
| 2,947
| true
|
Update meteor score from nltk update
|
It looks like there were issues in NLTK on the way the METEOR score was computed.
A fix was added in NLTK at https://github.com/nltk/nltk/pull/2763, and therefore the scoring function no longer returns the same values.
I updated the score of the example in the docs
|
https://github.com/huggingface/datasets/pull/2946
|
[] |
{
"url": "https://api.github.com/repos/huggingface/datasets/pulls/2946",
"html_url": "https://github.com/huggingface/datasets/pull/2946",
"diff_url": "https://github.com/huggingface/datasets/pull/2946.diff",
"patch_url": "https://github.com/huggingface/datasets/pull/2946.patch",
"merged_at": "2021-09-20T09:35:59"
}
| 2,946
| true
|
Protect master branch
|
After accidental merge commit (91c55355b634d0dc73350a7ddee1a6776dbbdd69) into `datasets` master branch, all commits present in the feature branch were permanently added to `datasets` master branch history, as e.g.:
- 00cc036fea7c7745cfe722360036ed306796a3f2
- 13ae8c98602bbad8197de3b9b425f4c78f582af1
- ...
I propose to protect our master branch, so that we avoid we can accidentally make this kind of mistakes in the future:
- [x] For Pull Requests using GitHub, allow only squash merging, so that only a single commit per Pull Request is merged into the master branch
- Currently, simple merge commits are already disabled
- I propose to disable rebase merging as well
- ~~Protect the master branch from direct pushes (to avoid accidentally pushing of merge commits)~~
- ~~This protection would reject direct pushes to master branch~~
- ~~If so, for each release (when we need to commit directly to the master branch), we should previously disable the protection and re-enable it again after the release~~
- [x] Protect the master branch only from direct pushing of **merge commits**
- GitHub offers the possibility to protect the master branch only from merge commits (which are the ones that introduce all the commits from the feature branch into the master branch).
- No need to disable/re-enable this protection on each release
This purpose of this Issue is to open a discussion about this problem and to agree in a solution.
|
https://github.com/huggingface/datasets/issues/2945
|
[
"Cool, I think we can do both :)",
"@lhoestq now the 2 are implemented.\r\n\r\nPlease note that for the the second protection, finally I have chosen to protect the master branch only from **merge commits** (see update comment above), so no need to disable/re-enable the protection on each release (direct commits, different from merge commits, can be pushed to the remote master branch; and eventually reverted without messing up the repo history)."
] | null | 2,945
| false
|
Add `remove_columns` to `IterableDataset `
|
**Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is.
```python
from datasets import load_dataset
dataset = load_dataset("c4", 'realnewslike', streaming =True, split='train')
dataset = dataset.remove_columns('url')
```
```
AttributeError: 'IterableDataset' object has no attribute 'remove_columns'
```
**Describe the solution you'd like**
It would be nice to have `.remove_columns()` to match the `Datasets` api.
**Describe alternatives you've considered**
This can be done with a single call to `.map()`,
I can try to help add this. 🤗
|
https://github.com/huggingface/datasets/issues/2944
|
[
"Hi ! Good idea :)\r\nIf you are interested in contributing, feel free to give it a try and open a Pull Request. Also let me know if I can help you with this or if you have questions"
] | null | 2,944
| false
|
Backwards compatibility broken for cached datasets that use `.filter()`
|
## Describe the bug
After upgrading to datasets `1.12.0`, some cached `.filter()` steps from `1.11.0` started failing with
`ValueError: Keys mismatch: between {'indices': Value(dtype='uint64', id=None)} and {'file': Value(dtype='string', id=None), 'text': Value(dtype='string', id=None), 'speaker_id': Value(dtype='int64', id=None), 'chapter_id': Value(dtype='int64', id=None), 'id': Value(dtype='string', id=None)}`
Related feature: https://github.com/huggingface/datasets/pull/2836
:question: This is probably a `wontfix` bug, since it can be solved by simply cleaning the related cache dirs, but the workaround could be useful for someone googling the error :)
## Workaround
Remove the cache for the given dataset, e.g. `rm -rf ~/.cache/huggingface/datasets/librispeech_asr`.
## Steps to reproduce the bug
1. Delete `~/.cache/huggingface/datasets/librispeech_asr` if it exists.
2. `pip install datasets==1.11.0` and run the following snippet:
```python
from datasets import load_dataset
ids = ["1272-141231-0000"]
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
ds = ds.filter(lambda x: x["id"] in ids)
```
3. `pip install datasets==1.12.1` and re-run the code again
## Expected results
Same result as with the previous `datasets` version.
## Actual results
```bash
Reusing dataset librispeech_asr (./.cache/huggingface/datasets/librispeech_asr/clean/2.1.0/468ec03677f46a8714ac6b5b64dba02d246a228d92cbbad7f3dc190fa039eab1)
Loading cached processed dataset at ./.cache/huggingface/datasets/librispeech_asr/clean/2.1.0/468ec03677f46a8714ac6b5b64dba02d246a228d92cbbad7f3dc190fa039eab1/cache-cd1c29844fdbc87a.arrow
Traceback (most recent call last):
File "./repos/transformers/src/transformers/models/wav2vec2/try_dataset.py", line 5, in <module>
ds = ds.filter(lambda x: x["id"] in ids)
File "./envs/transformers/lib/python3.8/site-packages/datasets/arrow_dataset.py", line 185, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
File "./envs/transformers/lib/python3.8/site-packages/datasets/fingerprint.py", line 398, in wrapper
out = func(self, *args, **kwargs)
File "./envs/transformers/lib/python3.8/site-packages/datasets/arrow_dataset.py", line 2169, in filter
indices = self.map(
File "./envs/transformers/lib/python3.8/site-packages/datasets/arrow_dataset.py", line 1686, in map
return self._map_single(
File "./envs/transformers/lib/python3.8/site-packages/datasets/arrow_dataset.py", line 185, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
File "./envs/transformers/lib/python3.8/site-packages/datasets/fingerprint.py", line 398, in wrapper
out = func(self, *args, **kwargs)
File "./envs/transformers/lib/python3.8/site-packages/datasets/arrow_dataset.py", line 1896, in _map_single
return Dataset.from_file(cache_file_name, info=info, split=self.split)
File "./envs/transformers/lib/python3.8/site-packages/datasets/arrow_dataset.py", line 343, in from_file
return cls(
File "./envs/transformers/lib/python3.8/site-packages/datasets/arrow_dataset.py", line 282, in __init__
self.info.features = self.info.features.reorder_fields_as(inferred_features)
File "./envs/transformers/lib/python3.8/site-packages/datasets/features.py", line 1151, in reorder_fields_as
return Features(recursive_reorder(self, other))
File "./envs/transformers/lib/python3.8/site-packages/datasets/features.py", line 1140, in recursive_reorder
raise ValueError(f"Keys mismatch: between {source} and {target}" + stack_position)
ValueError: Keys mismatch: between {'indices': Value(dtype='uint64', id=None)} and {'file': Value(dtype='string', id=None), 'text': Value(dtype='string', id=None), 'speaker_id': Value(dtype='int64', id=None), 'chapter_id': Value(dtype='int64', id=None), 'id': Value(dtype='string', id=None)}
Process finished with exit code 1
```
## Environment info
- `datasets` version: 1.12.1
- Platform: Linux-5.11.0-34-generic-x86_64-with-glibc2.17
- Python version: 3.8.10
- PyArrow version: 5.0.0
|
https://github.com/huggingface/datasets/issues/2943
|
[
"Hi ! I guess the caching mechanism should have considered the new `filter` to be different from the old one, and don't use cached results from the old `filter`.\r\nTo avoid other users from having this issue we could make the caching differentiate the two, what do you think ?",
"If it's easy enough to implement, then yes please 😄 But this issue can be low-priority, since I've only encountered it in a couple of `transformers` CI tests.",
"Well it can cause issue with anyone that updates `datasets` and re-run some code that uses filter, so I'm creating a PR",
"I just merged a fix, let me know if you're still having this kind of issues :)\r\n\r\nWe'll do a release soon to make this fix available",
"Definitely works on several manual cases with our dummy datasets, thank you @lhoestq !",
"Fixed by #2947."
] | null | 2,943
| false
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.