html_url stringlengths 48 51 | title stringlengths 5 280 | comments stringlengths 63 51.8k | body stringlengths 0 36.2k ⌀ | comment_length int64 16 1.52k | text stringlengths 159 54.1k | embeddings listlengths 768 768 |
|---|---|---|---|---|---|---|
https://github.com/huggingface/datasets/issues/6906 | irc_disentangle - Issue with splitting data | I still find out that there are some strange bug in v2.15.0 of datasets. it seems like that the *.arrow file cannot be established. it may be an index of the subsets. well I still try to debug it. but, one of the most efficient way may be using the google colab to build this index in the ~/huggingface/datasets, and than download them to replace the local file.....lol......it works! | ### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10 | 69 | irc_disentangle - Issue with splitting data
### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10
I still find out that there are some strange bug in v2.15.0 of datasets. it seems like that the *.arrow file cannot be established. it may be an index of the subsets. well I still try to debug it. but, one of the most efficient way may be using the google colab to build this index in the ~/huggingface/datasets, and than download them to replace the local file.....lol......it works! | [
-0.1607564091682434,
-0.12835319340229034,
-0.03919059410691261,
0.5721808671951294,
-0.06806319952011108,
0.062165386974811554,
0.42442840337753296,
0.2841145098209381,
0.25551173090934753,
0.24333840608596802,
-0.06968239694833755,
0.14003582298755646,
-0.13992339372634888,
0.09073927253... |
https://github.com/huggingface/datasets/issues/6906 | irc_disentangle - Issue with splitting data | Yeah I did try what you suggested and it didn’t work. I was able to get it
on a local from someone who access the dataset in the past. Let me know
when you end up fixing this bug.
On Tue, Jun 11, 2024 at 10:33 PM Vincent Lau ***@***.***>
wrote:
> I still find out that there are some strange bug in v2.15.0 of datasets.
> it seems like that the *.arrow file cannot be established. it may be an
> index of the subsets. well I still try to debug it. but, one of the most
> efficient way may be using the google colab to build this index in the
> ~/huggingface/datasets, and than download them to replace the local
> file.....lol......it works!
>
> —
> Reply to this email directly, view it on GitHub
> <https://github.com/huggingface/datasets/issues/6906#issuecomment-2161988798>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/A3HXU7BCJE2LOCWRVWPMNODZG6XPJAVCNFSM6AAAAABH45CNPWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRRHE4DQNZZHA>
> .
> You are receiving this because you authored the thread.Message ID:
> ***@***.***>
>
| ### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10 | 162 | irc_disentangle - Issue with splitting data
### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10
Yeah I did try what you suggested and it didn’t work. I was able to get it
on a local from someone who access the dataset in the past. Let me know
when you end up fixing this bug.
On Tue, Jun 11, 2024 at 10:33 PM Vincent Lau ***@***.***>
wrote:
> I still find out that there are some strange bug in v2.15.0 of datasets.
> it seems like that the *.arrow file cannot be established. it may be an
> index of the subsets. well I still try to debug it. but, one of the most
> efficient way may be using the google colab to build this index in the
> ~/huggingface/datasets, and than download them to replace the local
> file.....lol......it works!
>
> —
> Reply to this email directly, view it on GitHub
> <https://github.com/huggingface/datasets/issues/6906#issuecomment-2161988798>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/A3HXU7BCJE2LOCWRVWPMNODZG6XPJAVCNFSM6AAAAABH45CNPWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRRHE4DQNZZHA>
> .
> You are receiving this because you authored the thread.Message ID:
> ***@***.***>
>
| [
-0.11062994599342346,
-0.20289093255996704,
-0.020420651882886887,
0.5355083346366882,
-0.01568976789712906,
0.027922704815864563,
0.40039125084877014,
0.27474549412727356,
0.3411843776702881,
0.2529586851596832,
-0.13087789714336395,
0.21824510395526886,
-0.07574411481618881,
0.1518655568... |
https://github.com/huggingface/datasets/issues/6906 | irc_disentangle - Issue with splitting data | Could you please provide more information, as required by the Bug template: https://github.com/huggingface/datasets/issues/new?assignees=&labels=&projects=&template=bug-report.yml
Without all that information, it is very difficult for us to understand the underlying issue and to give a pertinent answer.
What are the versions of the libraries you are using? Datasets, pyarrow, fsspec,...
> Environment info
> Please share your environemnt info with us. You can run the command datasets-cli env and copy-paste its output below.
What is the output you get after executing these code lines?
```python
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
```
| ### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10 | 89 | irc_disentangle - Issue with splitting data
### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10
Could you please provide more information, as required by the Bug template: https://github.com/huggingface/datasets/issues/new?assignees=&labels=&projects=&template=bug-report.yml
Without all that information, it is very difficult for us to understand the underlying issue and to give a pertinent answer.
What are the versions of the libraries you are using? Datasets, pyarrow, fsspec,...
> Environment info
> Please share your environemnt info with us. You can run the command datasets-cli env and copy-paste its output below.
What is the output you get after executing these code lines?
```python
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
```
| [
-0.058951303362846375,
-0.29593199491500854,
-0.0023607350885868073,
0.5987377762794495,
0.011229859665036201,
0.017998501658439636,
0.46196457743644714,
0.32676392793655396,
0.331938773393631,
0.21582576632499695,
-0.04904395341873169,
0.23178730905056,
-0.05281625688076019,
0.14836469292... |
https://github.com/huggingface/datasets/issues/6906 | irc_disentangle - Issue with splitting data | We have made the following fixes:
- [Fix source data URL](https://huggingface.co/datasets/jkkummerfeld/irc_disentangle/discussions/4)
- [Convert dataset to Parquet](https://huggingface.co/datasets/jkkummerfeld/irc_disentangle/discussions/5) | ### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10 | 16 | irc_disentangle - Issue with splitting data
### Describe the bug
I am trying to access your database through python using "datasets.load_dataset("irc_disentangle")" and I am getting this error message:
ValueError: Instruction "train" corresponds to no data!
### Steps to reproduce the bug
import datasets
ds = datasets.load_dataset('irc_disentangle')
ds
### Expected behavior
The data is supposed to load into ds and be accessable as such:
ds['train'][1050], ds['train'][1055]
### Environment info
I tired Python 3.12 and 3.10
We have made the following fixes:
- [Fix source data URL](https://huggingface.co/datasets/jkkummerfeld/irc_disentangle/discussions/4)
- [Convert dataset to Parquet](https://huggingface.co/datasets/jkkummerfeld/irc_disentangle/discussions/5) | [
-0.02859576605260372,
-0.22506079077720642,
-0.026607561856508255,
0.6058424711227417,
0.04811711609363556,
0.004391923546791077,
0.3899601399898529,
0.3061465919017792,
0.19714923202991486,
0.2205083668231964,
-0.02636583335697651,
0.28163039684295654,
-0.07801733911037445,
0.154011383652... |
https://github.com/huggingface/datasets/issues/6903 | Add the option of saving in parquet instead of arrow | No, it does not save the metadata json.
We have to recode all meta json load/save
with another custome functions.
save_to_disk
and load should have option with
“Parquet” instead of “arrow”
since “arrow” is never user for production
(only parquet).
Thanks !
> On May 17, 2024, at 5:38, Frédéric Branchaud-Charron ***@***.***> wrote:
>
>
> I think Dataset.to_parquet is what you're looking for.
>
> Let me know if I'm wrong
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| ### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing ! | 98 | Add the option of saving in parquet instead of arrow
### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing !
No, it does not save the metadata json.
We have to recode all meta json load/save
with another custome functions.
save_to_disk
and load should have option with
“Parquet” instead of “arrow”
since “arrow” is never user for production
(only parquet).
Thanks !
> On May 17, 2024, at 5:38, Frédéric Branchaud-Charron ***@***.***> wrote:
>
>
> I think Dataset.to_parquet is what you're looking for.
>
> Let me know if I'm wrong
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| [
-0.2602289021015167,
0.25236329436302185,
-0.08383029699325562,
0.1211492270231247,
0.11433389037847519,
-0.14599336683750153,
0.055254269391298294,
0.21268214285373688,
-0.23365715146064758,
0.050044991075992584,
0.1188136637210846,
0.8451120853424072,
-0.17672879993915558,
0.211130380630... |
https://github.com/huggingface/datasets/issues/6903 | Add the option of saving in parquet instead of arrow | Ok,
What about loading ?
Should we do in 2 steps ?
> On Jun 14, 2024, at 1:09, Quentin Lhoest ***@***.***> wrote:
>
>
> You can use to_parquet and ds.info.write_to_directory() to save the dataset info
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| ### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing ! | 63 | Add the option of saving in parquet instead of arrow
### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing !
Ok,
What about loading ?
Should we do in 2 steps ?
> On Jun 14, 2024, at 1:09, Quentin Lhoest ***@***.***> wrote:
>
>
> You can use to_parquet and ds.info.write_to_directory() to save the dataset info
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| [
-0.4148215651512146,
0.05486324429512024,
-0.13479964435100555,
0.21776516735553741,
0.14533403515815735,
-0.1493590623140335,
0.15330255031585693,
0.2757137715816498,
0.057868171483278275,
0.31751492619514465,
0.09430436044931412,
0.6173985600471497,
-0.012661146931350231,
0.3191124200820... |
https://github.com/huggingface/datasets/issues/6903 | Add the option of saving in parquet instead of arrow | Isn’t easier to combine both
into load_dataset and save_dataset
with parquet options.
2) another question,
How can we download large dataset into disk directly without loading all in memory (!)
> On Jun 14, 2024, at 19:54, Quentin Lhoest ***@***.***> wrote:
>
>
> Yes, and there is DatasetInfo.from_directory(). to reload the info
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| ### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing ! | 79 | Add the option of saving in parquet instead of arrow
### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing !
Isn’t easier to combine both
into load_dataset and save_dataset
with parquet options.
2) another question,
How can we download large dataset into disk directly without loading all in memory (!)
> On Jun 14, 2024, at 19:54, Quentin Lhoest ***@***.***> wrote:
>
>
> Yes, and there is DatasetInfo.from_directory(). to reload the info
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| [
-0.5679495930671692,
0.04482512176036835,
-0.11140307784080505,
0.22289925813674927,
0.19645869731903076,
-0.03018975630402565,
-0.04859768971800804,
0.34565630555152893,
0.01152343675494194,
0.2370610237121582,
0.0372697189450264,
0.5357140302658081,
-0.0758049339056015,
0.209399476647377... |
https://github.com/huggingface/datasets/issues/6903 | Add the option of saving in parquet instead of arrow | `load_dataset` doesn't load the dataset in memory, it progressively writes to disk in Arrow format and then memory maps the Arrow files. This allows to load datasets bigger than memory and without filling your RAM | ### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing ! | 35 | Add the option of saving in parquet instead of arrow
### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing !
`load_dataset` doesn't load the dataset in memory, it progressively writes to disk in Arrow format and then memory maps the Arrow files. This allows to load datasets bigger than memory and without filling your RAM | [
-0.5085446238517761,
-0.04656665772199631,
-0.13971364498138428,
0.270813524723053,
0.21689365804195404,
-0.0961124524474144,
0.0268910750746727,
0.24537453055381775,
0.043092723935842514,
0.23568403720855713,
0.07143481075763702,
0.494878888130188,
-0.09851767867803574,
0.1642365157604217... |
https://github.com/huggingface/datasets/issues/6903 | Add the option of saving in parquet instead of arrow | Sure.
How memory map is managed ?
Managed by the OS ?
Why the need of save_dataset() ?
> On Jun 15, 2024, at 0:06, Quentin Lhoest ***@***.***> wrote:
>
>
> load_dataset doesn't load the dataset in memory, it progressively writes to disk in Arrow format and then memory maps the Arrow files. This allows to load datasets bigger than memory and without filling your RAM
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| ### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing ! | 93 | Add the option of saving in parquet instead of arrow
### Feature request
In dataset.save_to_disk('/path/to/save/dataset'),
add the option to save in parquet format
dataset.save_to_disk('/path/to/save/dataset', format="parquet"),
because arrow is not used for Production Big data.... (only parquet)
### Motivation
because arrow is not used for Production Big data.... (only parquet)
### Your contribution
I can do the testing !
Sure.
How memory map is managed ?
Managed by the OS ?
Why the need of save_dataset() ?
> On Jun 15, 2024, at 0:06, Quentin Lhoest ***@***.***> wrote:
>
>
> load_dataset doesn't load the dataset in memory, it progressively writes to disk in Arrow format and then memory maps the Arrow files. This allows to load datasets bigger than memory and without filling your RAM
>
> —
> Reply to this email directly, view it on GitHub, or unsubscribe.
> You are receiving this because you authored the thread.
| [
-0.4061277210712433,
-0.07804780453443527,
-0.15096910297870636,
0.3015359044075012,
0.17626085877418518,
-0.01711559295654297,
0.01719146966934204,
0.16975663602352142,
0.1115928441286087,
0.2302323579788208,
0.08540602028369904,
0.5447483658790588,
-0.1307019293308258,
-0.006384987384080... |
https://github.com/huggingface/datasets/issues/6900 | [WebDataset] KeyError with user-defined `Features` when a field is missing in an example | It shouldn't be difficult, I think it's just a matter of adding the missing fields from `self.config.features` in `example` here: before it iterates on image_field_names and audio_field_names. A missing field should have a value set to None
https://github.com/huggingface/datasets/blob/768cb35ede5a6c35fa7545aa3671f3e321c96440/src/datasets/packaged_modules/webdataset/webdataset.py#L113-L116 | reported at https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions/discussions/1
```
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
``` | 38 | [WebDataset] KeyError with user-defined `Features` when a field is missing in an example
reported at https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions/discussions/1
```
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
```
It shouldn't be difficult, I think it's just a matter of adding the missing fields from `self.config.features` in `example` here: before it iterates on image_field_names and audio_field_names. A missing field should have a value set to None
https://github.com/huggingface/datasets/blob/768cb35ede5a6c35fa7545aa3671f3e321c96440/src/datasets/packaged_modules/webdataset/webdataset.py#L113-L116 | [
0.25792214274406433,
-0.18959563970565796,
-0.07479274272918701,
0.0006574913859367371,
-0.0015557631850242615,
0.10873619467020035,
0.34772181510925293,
0.27355822920799255,
0.12412917613983154,
-0.008390381932258606,
0.17592963576316833,
0.19717289507389069,
-0.14997316896915436,
0.39989... |
https://github.com/huggingface/datasets/issues/6900 | [WebDataset] KeyError with user-defined `Features` when a field is missing in an example | @lhoestq So like this then?
```
def _generate_examples(self, tar_paths, tar_iterators):
image_field_names = [
field_name for field_name, feature in self.info.features.items() if isinstance(feature, datasets.Image)
]
audio_field_names = [
field_name for field_name, feature in self.info.features.items() if isinstance(feature, datasets.Audio)
]
all_field_names = list(self.config.features.keys())
for tar_idx, (tar_path, tar_iterator) in enumerate(zip(tar_paths, tar_iterators)):
for example_idx, example in enumerate(self._get_pipeline_from_tar(tar_path, tar_iterator)):
for field_name in all_field_names:
if field_name not in example:
if field_name in self.config.features:
example[field_name] = self.config.features[field_name]
else:
example[field_name] = None
# Process image and audio fields
for field_name in image_field_names + audio_field_names:
if example[field_name] is not None:
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
yield f"{tar_idx}_{example_idx}", example
```
Or should we avoid trying add the missing values and just set them to None?
```
for field_name in all_field_names:
if field_name not in example:
example[field_name] = None
``` | reported at https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions/discussions/1
```
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
``` | 132 | [WebDataset] KeyError with user-defined `Features` when a field is missing in an example
reported at https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions/discussions/1
```
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
```
@lhoestq So like this then?
```
def _generate_examples(self, tar_paths, tar_iterators):
image_field_names = [
field_name for field_name, feature in self.info.features.items() if isinstance(feature, datasets.Image)
]
audio_field_names = [
field_name for field_name, feature in self.info.features.items() if isinstance(feature, datasets.Audio)
]
all_field_names = list(self.config.features.keys())
for tar_idx, (tar_path, tar_iterator) in enumerate(zip(tar_paths, tar_iterators)):
for example_idx, example in enumerate(self._get_pipeline_from_tar(tar_path, tar_iterator)):
for field_name in all_field_names:
if field_name not in example:
if field_name in self.config.features:
example[field_name] = self.config.features[field_name]
else:
example[field_name] = None
# Process image and audio fields
for field_name in image_field_names + audio_field_names:
if example[field_name] is not None:
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
yield f"{tar_idx}_{example_idx}", example
```
Or should we avoid trying add the missing values and just set them to None?
```
for field_name in all_field_names:
if field_name not in example:
example[field_name] = None
``` | [
0.1758723258972168,
-0.21770384907722473,
-0.1031029224395752,
0.07011441141366959,
0.021451488137245178,
0.002216920256614685,
0.3708716332912445,
0.26003575325012207,
0.13557389378547668,
0.040011271834373474,
0.3267216682434082,
0.2618832290172577,
-0.17561253905296326,
0.23694916069507... |
https://github.com/huggingface/datasets/issues/6900 | [WebDataset] KeyError with user-defined `Features` when a field is missing in an example | Yup this is the solution !
```python
for field_name in all_field_names:
if field_name not in example:
example[field_name] = None
``` | reported at https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions/discussions/1
```
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
``` | 20 | [WebDataset] KeyError with user-defined `Features` when a field is missing in an example
reported at https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions/discussions/1
```
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
```
Yup this is the solution !
```python
for field_name in all_field_names:
if field_name not in example:
example[field_name] = None
``` | [
0.3126630485057831,
-0.23326580226421356,
-0.09404422342777252,
-0.03234681487083435,
-0.03845921903848648,
0.09963660687208176,
0.3824291229248047,
0.3597545325756073,
0.18889564275741577,
-0.04843384027481079,
0.41676995158195496,
0.20619578659534454,
-0.0827454999089241,
0.4317778646945... |
https://github.com/huggingface/datasets/issues/6897 | datasets template guide :: issue in documentation YAML | Hello, @bghira.
Thanks for reporting. Please note that the text originating the error is not supposed to be valid YAML: it contains the instructions to generate the actual YAML content, that should replace the instructions comment.
On the other hand, I agree that it is not nice to have that YAML error message at the top of the page:
I am proposing a change to make the YAML error disappear. | ### Describe the bug
There is a YAML error at the top of the page, and I don't think it's supposed to be there
### Steps to reproduce the bug
1. Browse to [this tutorial document](https://github.com/huggingface/datasets/blob/main/templates/README_guide.md)
2. Observe a big red error at the top
3. The rest of the document remains functional
### Expected behavior
I think the YAML block should be displayed or ignored.
### Environment info
N/A | 74 | datasets template guide :: issue in documentation YAML
### Describe the bug
There is a YAML error at the top of the page, and I don't think it's supposed to be there
### Steps to reproduce the bug
1. Browse to [this tutorial document](https://github.com/huggingface/datasets/blob/main/templates/README_guide.md)
2. Observe a big red error at the top
3. The rest of the document remains functional
### Expected behavior
I think the YAML block should be displayed or ignored.
### Environment info
N/A
Hello, @bghira.
Thanks for reporting. Please note that the text originating the error is not supposed to be valid YAML: it contains the instructions to generate the actual YAML content, that should replace the instructions comment.
On the other hand, I agree that it is not nice to have that YAML error message at the top of the page:
I am proposing a change to make the YAML error disappear. | [
0.0628998875617981,
-0.5750184059143066,
0.08095027506351471,
0.06528156995773315,
0.3983103036880493,
0.501067042350769,
0.3459073603153229,
0.14436297118663788,
-0.11757896840572357,
-0.02074657380580902,
0.2931133508682251,
0.1422312706708908,
0.135603666305542,
0.0730132982134819,
0.... |
https://github.com/huggingface/datasets/issues/6897 | datasets template guide :: issue in documentation YAML | thanks albert! i looked at it for a while to figure it out. i think the `raw` view option is the correct way to look at it? | ### Describe the bug
There is a YAML error at the top of the page, and I don't think it's supposed to be there
### Steps to reproduce the bug
1. Browse to [this tutorial document](https://github.com/huggingface/datasets/blob/main/templates/README_guide.md)
2. Observe a big red error at the top
3. The rest of the document remains functional
### Expected behavior
I think the YAML block should be displayed or ignored.
### Environment info
N/A | 27 | datasets template guide :: issue in documentation YAML
### Describe the bug
There is a YAML error at the top of the page, and I don't think it's supposed to be there
### Steps to reproduce the bug
1. Browse to [this tutorial document](https://github.com/huggingface/datasets/blob/main/templates/README_guide.md)
2. Observe a big red error at the top
3. The rest of the document remains functional
### Expected behavior
I think the YAML block should be displayed or ignored.
### Environment info
N/A
thanks albert! i looked at it for a while to figure it out. i think the `raw` view option is the correct way to look at it? | [
0.005896255373954773,
-0.6684544086456299,
0.04609465226531029,
0.2047625333070755,
0.2285977602005005,
0.43979182839393616,
0.33936700224876404,
0.15852387249469757,
-0.06923028081655502,
-0.08908259868621826,
0.28198954463005066,
0.08838176727294922,
0.15670068562030792,
0.14199063181877... |
https://github.com/huggingface/datasets/issues/6891 | Unable to load JSON saved using `to_json` | Hi @DarshanDeshpande,
Please note that the default format of the method `Dataset.to_json` is [JSON-Lines](https://jsonlines.org/): it passes `orient="records", lines=True` to `pandas.DataFrame.to_json`. This format is specially useful for large datasets, since unlike regular JSON files, it does not require loading all the data into memory at once, but can be done iteratively by batches.
In order to read this file using the `json` library, you should parse line by line:
```python
with open("full_dataset.json", "r") as f:
data = [json.loads(line) for line in f]
len(data)
```
Maybe we should explain this better in our docs. | ### Describe the bug
Datasets stored in the JSON format cannot be loaded using `json.load()`
### Steps to reproduce the bug
```
import json
from datasets import load_dataset
dataset = load_dataset("squad")
train_dataset, test_dataset = dataset["train"], dataset["validation"]
test_dataset.to_json("full_dataset.json")
# This works
loaded_test = load_dataset("json", data_files="full_dataset.json")
# This fails
loaded_test = json.load(open("full_dataset.json", "r"))
```
### Expected behavior
The JSON should be correctly formatted when writing so that it can be loaded using `json.load()`.
### Environment info
Colab: https://colab.research.google.com/drive/1st1iStFUVgu9ZPvnzSzL4vDeYWDwYpUm?usp=sharing | 92 | Unable to load JSON saved using `to_json`
### Describe the bug
Datasets stored in the JSON format cannot be loaded using `json.load()`
### Steps to reproduce the bug
```
import json
from datasets import load_dataset
dataset = load_dataset("squad")
train_dataset, test_dataset = dataset["train"], dataset["validation"]
test_dataset.to_json("full_dataset.json")
# This works
loaded_test = load_dataset("json", data_files="full_dataset.json")
# This fails
loaded_test = json.load(open("full_dataset.json", "r"))
```
### Expected behavior
The JSON should be correctly formatted when writing so that it can be loaded using `json.load()`.
### Environment info
Colab: https://colab.research.google.com/drive/1st1iStFUVgu9ZPvnzSzL4vDeYWDwYpUm?usp=sharing
Hi @DarshanDeshpande,
Please note that the default format of the method `Dataset.to_json` is [JSON-Lines](https://jsonlines.org/): it passes `orient="records", lines=True` to `pandas.DataFrame.to_json`. This format is specially useful for large datasets, since unlike regular JSON files, it does not require loading all the data into memory at once, but can be done iteratively by batches.
In order to read this file using the `json` library, you should parse line by line:
```python
with open("full_dataset.json", "r") as f:
data = [json.loads(line) for line in f]
len(data)
```
Maybe we should explain this better in our docs. | [
0.05648532137274742,
0.1627628207206726,
-0.0438183918595314,
0.3540229797363281,
0.39994484186172485,
0.25341832637786865,
0.2940066456794739,
0.5735443234443665,
0.12257605791091919,
-0.001523330807685852,
-0.12310168892145157,
0.6201069355010986,
0.027151502668857574,
0.0723390430212020... |
https://github.com/huggingface/datasets/issues/6887 | FAISS load to None | Hello,
I'm not sure I understand.
The return value of `ds.load_faiss_index` is None as expected.
I see that loading an Index on a dataset that doesn't have an `embedding` column doesn't raise an Issue. Is that the issue?
So `ds` doesn't have an `embedding` column, but we load an index that looks for it. But this will raise an issue only when calling `ds.search`. | ### Describe the bug
I've use FAISS with Datasets and save to FAISS.
Then load to save FAISS then no error, then ds to None
```python
ds.load_faiss_index('embeddings', 'my_index.faiss')
```
### Steps to reproduce the bug
# 1.
```python
ds_with_embeddings = ds.map(lambda example: {'embeddings': model(transforms(example['image']).unsqueeze(0)).squeeze()}, batch_size=64)
ds_with_embeddings.add_faiss_index(column='embeddings')
ds_with_embeddings.save_faiss_index('embeddings', 'index.faiss')
```
# 2.
```python
ds.load_faiss_index('embeddings', 'my_index.faiss')
```
### Expected behavior
Add column in Datasets.
### Environment info
Google Colab, SageMaker Notebook | 64 | FAISS load to None
### Describe the bug
I've use FAISS with Datasets and save to FAISS.
Then load to save FAISS then no error, then ds to None
```python
ds.load_faiss_index('embeddings', 'my_index.faiss')
```
### Steps to reproduce the bug
# 1.
```python
ds_with_embeddings = ds.map(lambda example: {'embeddings': model(transforms(example['image']).unsqueeze(0)).squeeze()}, batch_size=64)
ds_with_embeddings.add_faiss_index(column='embeddings')
ds_with_embeddings.save_faiss_index('embeddings', 'index.faiss')
```
# 2.
```python
ds.load_faiss_index('embeddings', 'my_index.faiss')
```
### Expected behavior
Add column in Datasets.
### Environment info
Google Colab, SageMaker Notebook
Hello,
I'm not sure I understand.
The return value of `ds.load_faiss_index` is None as expected.
I see that loading an Index on a dataset that doesn't have an `embedding` column doesn't raise an Issue. Is that the issue?
So `ds` doesn't have an `embedding` column, but we load an index that looks for it. But this will raise an issue only when calling `ds.search`. | [
0.03885560855269432,
-0.4406752288341522,
0.05914180353283882,
0.245367169380188,
0.20897629857063293,
0.07248102873563766,
0.5071743726730347,
0.1786344349384308,
0.8528186082839966,
0.45393699407577515,
0.018829818814992905,
0.10844770073890686,
0.06420552730560303,
-0.07409307360649109,... |
https://github.com/huggingface/datasets/issues/6882 | Connection Error When Using By-pass Proxies | Changing the supplier of the proxy will solve this problem, or you can visit and follow the instructions in https://hf-mirror.com | ### Describe the bug
I'm currently using Clash for Windows as my proxy tunnel, after exporting HTTP_PROXY and HTTPS_PROXY to the port that clash provides🤔, it runs into a connection error saying "Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (ConnectionError(MaxRetryError("HTTPSConnectionPool(host='raw.githubusercontent.com', port=443): Max retries exceeded with url: /huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7f969d391870>: Failed to establish a new connection: [Errno 111] Connection refused'))")))"
I have already read the documentation provided on the hugginface, but I think I didn't see the detailed instruction on how to set up proxies for this library.
### Steps to reproduce the bug
1. Turn on any proxy software like Clash / ShadosocksR etc.
2. export system varibles to the port provided by your proxy software in wsl (It's ok for other applications to use proxy expect dataset-library)
3. load any dataset from hugginface online
### Expected behavior
---------------------------------------------------------------------------
ConnectionError Traceback (most recent call last)
Cell In[33], [line 3](vscode-notebook-cell:?execution_count=33&line=3)
[1](vscode-notebook-cell:?execution_count=33&line=1) from datasets import load_metric
----> [3](vscode-notebook-cell:?execution_count=33&line=3) metric = load_metric("seqeval")
File ~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:46, in deprecated.<locals>.decorator.<locals>.wrapper(*args, **kwargs)
[44](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:44) warnings.warn(warning_msg, category=FutureWarning, stacklevel=2)
[45](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:45) _emitted_deprecation_warnings.add(func_hash)
---> [46](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:46) return deprecated_function(*args, **kwargs)
File ~/.local/lib/python3.10/site-packages/datasets/load.py:2104, in load_metric(path, config_name, process_id, num_process, cache_dir, experiment_id, keep_in_memory, download_config, download_mode, revision, trust_remote_code, **metric_init_kwargs)
[2101](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2101) warnings.filterwarnings("ignore", message=".*https://huggingface.co/docs/evaluate$", category=FutureWarning)
[2103](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2103) download_mode = DownloadMode(download_mode or DownloadMode.REUSE_DATASET_IF_EXISTS)
-> [2104](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2104) metric_module = metric_module_factory(
[2105](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2105) path,
[2106](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2106) revision=revision,
[2107](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2107) download_config=download_config,
[2108](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2108) download_mode=download_mode,
[2109](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2109) trust_remote_code=trust_remote_code,
[2110](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2110) ).module_path
[2111](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2111) metric_cls = import_main_class(metric_module, dataset=False)
[2112](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2112) metric = metric_cls(
[2113](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2113) config_name=config_name,
[2114](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2114) process_id=process_id,
...
--> [633](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/file_utils.py:633) raise ConnectionError(f"Couldn't reach {url} ({repr(head_error)})")
[634](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/file_utils.py:634) elif response is not None:
[635](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/file_utils.py:635) raise ConnectionError(f"Couldn't reach {url} (error {response.status_code})")
ConnectionError: Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (SSLError(MaxRetryError("HTTPSConnectionPool(host='raw.githubusercontent.com', port=443): Max retries exceeded with url: /huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (Caused by SSLError(SSLEOFError(8, '[SSL: UNEXPECTED_EOF_WHILE_READING] EOF occurred in violation of protocol (_ssl.c:1007)')))")))
### Environment info
- `datasets` version: 2.19.1
- Platform: Linux-5.10.102.1-microsoft-standard-WSL2-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.2.0 | 20 | Connection Error When Using By-pass Proxies
### Describe the bug
I'm currently using Clash for Windows as my proxy tunnel, after exporting HTTP_PROXY and HTTPS_PROXY to the port that clash provides🤔, it runs into a connection error saying "Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (ConnectionError(MaxRetryError("HTTPSConnectionPool(host='raw.githubusercontent.com', port=443): Max retries exceeded with url: /huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7f969d391870>: Failed to establish a new connection: [Errno 111] Connection refused'))")))"
I have already read the documentation provided on the hugginface, but I think I didn't see the detailed instruction on how to set up proxies for this library.
### Steps to reproduce the bug
1. Turn on any proxy software like Clash / ShadosocksR etc.
2. export system varibles to the port provided by your proxy software in wsl (It's ok for other applications to use proxy expect dataset-library)
3. load any dataset from hugginface online
### Expected behavior
---------------------------------------------------------------------------
ConnectionError Traceback (most recent call last)
Cell In[33], [line 3](vscode-notebook-cell:?execution_count=33&line=3)
[1](vscode-notebook-cell:?execution_count=33&line=1) from datasets import load_metric
----> [3](vscode-notebook-cell:?execution_count=33&line=3) metric = load_metric("seqeval")
File ~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:46, in deprecated.<locals>.decorator.<locals>.wrapper(*args, **kwargs)
[44](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:44) warnings.warn(warning_msg, category=FutureWarning, stacklevel=2)
[45](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:45) _emitted_deprecation_warnings.add(func_hash)
---> [46](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/deprecation_utils.py:46) return deprecated_function(*args, **kwargs)
File ~/.local/lib/python3.10/site-packages/datasets/load.py:2104, in load_metric(path, config_name, process_id, num_process, cache_dir, experiment_id, keep_in_memory, download_config, download_mode, revision, trust_remote_code, **metric_init_kwargs)
[2101](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2101) warnings.filterwarnings("ignore", message=".*https://huggingface.co/docs/evaluate$", category=FutureWarning)
[2103](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2103) download_mode = DownloadMode(download_mode or DownloadMode.REUSE_DATASET_IF_EXISTS)
-> [2104](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2104) metric_module = metric_module_factory(
[2105](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2105) path,
[2106](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2106) revision=revision,
[2107](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2107) download_config=download_config,
[2108](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2108) download_mode=download_mode,
[2109](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2109) trust_remote_code=trust_remote_code,
[2110](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2110) ).module_path
[2111](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2111) metric_cls = import_main_class(metric_module, dataset=False)
[2112](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2112) metric = metric_cls(
[2113](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2113) config_name=config_name,
[2114](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/load.py:2114) process_id=process_id,
...
--> [633](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/file_utils.py:633) raise ConnectionError(f"Couldn't reach {url} ({repr(head_error)})")
[634](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/file_utils.py:634) elif response is not None:
[635](https://vscode-remote+wsl-002bubuntu-002d22-002e04.vscode-resource.vscode-cdn.net/home/noodle/Transformers-Tutorials/LayoutLMv3/~/.local/lib/python3.10/site-packages/datasets/utils/file_utils.py:635) raise ConnectionError(f"Couldn't reach {url} (error {response.status_code})")
ConnectionError: Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (SSLError(MaxRetryError("HTTPSConnectionPool(host='raw.githubusercontent.com', port=443): Max retries exceeded with url: /huggingface/datasets/2.19.1/metrics/seqeval/seqeval.py (Caused by SSLError(SSLEOFError(8, '[SSL: UNEXPECTED_EOF_WHILE_READING] EOF occurred in violation of protocol (_ssl.c:1007)')))")))
### Environment info
- `datasets` version: 2.19.1
- Platform: Linux-5.10.102.1-microsoft-standard-WSL2-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.2.0
Changing the supplier of the proxy will solve this problem, or you can visit and follow the instructions in https://hf-mirror.com | [
-0.20472918450832367,
-0.37931641936302185,
0.0601181723177433,
0.2547002136707306,
0.047565702348947525,
-0.19099479913711548,
0.08826708048582077,
-0.08290274441242218,
0.14276599884033203,
0.07145500928163528,
0.022649560123682022,
-0.14429670572280884,
0.2713148891925812,
0.40864679217... |
https://github.com/huggingface/datasets/issues/6880 | Webdataset: KeyError: 'png' on some datasets when streaming | The error is caused by malformed basenames of the files within the TARs:
- `15_Cohen_1-s2.0-S0929664620300449-gr3_lrg-b.png` becomes `15_Cohen_1-s2` as the grouping `__key__`, and `0-S0929664620300449-gr3_lrg-b.png` as the additional key to be added to the example
- whereas the intended behavior was to use `15_Cohen_1-s2.0-S0929664620300449-gr3_lrg-b` as the grouping `__key__`, and `png` as the additional key to be added to the example
To get the expected behavior, the basenames of the files within the TARs should be fixed so that they only contain a single dot, the one separating the file extension. | reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
``` | 88 | Webdataset: KeyError: 'png' on some datasets when streaming
reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
```
The error is caused by malformed basenames of the files within the TARs:
- `15_Cohen_1-s2.0-S0929664620300449-gr3_lrg-b.png` becomes `15_Cohen_1-s2` as the grouping `__key__`, and `0-S0929664620300449-gr3_lrg-b.png` as the additional key to be added to the example
- whereas the intended behavior was to use `15_Cohen_1-s2.0-S0929664620300449-gr3_lrg-b` as the grouping `__key__`, and `png` as the additional key to be added to the example
To get the expected behavior, the basenames of the files within the TARs should be fixed so that they only contain a single dot, the one separating the file extension. | [
-0.311941534280777,
-0.2119862288236618,
-0.1765904724597931,
0.1928684115409851,
0.29888659715652466,
0.041977062821388245,
0.16332295536994934,
0.4688340425491333,
0.017786100506782532,
-0.06651580333709717,
-0.092808797955513,
0.0658211037516594,
-0.20402628183364868,
0.1120481416583061... |
https://github.com/huggingface/datasets/issues/6880 | Webdataset: KeyError: 'png' on some datasets when streaming | I reopen it because I think we should try to give a clearer error message with a specific error code.
For now, it's hard for the user to understand where the error comes from (not everybody knows the subtleties of the webdataset filename structure).
(we can transfer it to https://github.com/huggingface/dataset-viewer if it fits better there) | reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
``` | 55 | Webdataset: KeyError: 'png' on some datasets when streaming
reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
```
I reopen it because I think we should try to give a clearer error message with a specific error code.
For now, it's hard for the user to understand where the error comes from (not everybody knows the subtleties of the webdataset filename structure).
(we can transfer it to https://github.com/huggingface/dataset-viewer if it fits better there) | [
-0.311941534280777,
-0.2119862288236618,
-0.1765904724597931,
0.1928684115409851,
0.29888659715652466,
0.041977062821388245,
0.16332295536994934,
0.4688340425491333,
0.017786100506782532,
-0.06651580333709717,
-0.092808797955513,
0.0658211037516594,
-0.20402628183364868,
0.1120481416583061... |
https://github.com/huggingface/datasets/issues/6880 | Webdataset: KeyError: 'png' on some datasets when streaming | same with .jpg -> https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions
```
Error code: DatasetGenerationError
Exception: DatasetGenerationError
Message: An error occurred while generating the dataset
Traceback: Traceback (most recent call last):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1748, in _prepare_split_single
for key, record in generator:
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 818, in wrapped
for item in generator(*args, **kwargs):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
KeyError: 'jpg'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1316, in compute_config_parquet_and_info_response
parquet_operations, partial = stream_convert_to_parquet(
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 909, in stream_convert_to_parquet
builder._prepare_split(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1627, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1784, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
```
| reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
``` | 135 | Webdataset: KeyError: 'png' on some datasets when streaming
reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
```
same with .jpg -> https://huggingface.co/datasets/ProGamerGov/synthetic-dataset-1m-dalle3-high-quality-captions
```
Error code: DatasetGenerationError
Exception: DatasetGenerationError
Message: An error occurred while generating the dataset
Traceback: Traceback (most recent call last):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1748, in _prepare_split_single
for key, record in generator:
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 818, in wrapped
for item in generator(*args, **kwargs):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 109, in _generate_examples
example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
KeyError: 'jpg'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1316, in compute_config_parquet_and_info_response
parquet_operations, partial = stream_convert_to_parquet(
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 909, in stream_convert_to_parquet
builder._prepare_split(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1627, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1784, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
```
| [
-0.311941534280777,
-0.2119862288236618,
-0.1765904724597931,
0.1928684115409851,
0.29888659715652466,
0.041977062821388245,
0.16332295536994934,
0.4688340425491333,
0.017786100506782532,
-0.06651580333709717,
-0.092808797955513,
0.0658211037516594,
-0.20402628183364868,
0.1120481416583061... |
https://github.com/huggingface/datasets/issues/6880 | Webdataset: KeyError: 'png' on some datasets when streaming | More details in the spec (https://docs.google.com/document/d/18OdLjruFNX74ILmgrdiCI9J1fQZuhzzRBCHV9URWto0/edit#heading=h.hkptaq2kct2s)
> The prefix of a file is all directory components of the file plus the file name component up to the first “.” in the file name.
> The last extension (i.e., the portion after the last “.”) in a file name determines the file type.
> Example:
images17/image194.left.jpg
images17/image194.right.jpg
images17/image194.json
images17/image12.left.jpg
images17/image12.json
images17/image12.right.jpg
images3/image1459.left.jpg
> …
> When reading this with a WebDataset library, you would get the following two dictionaries back in sequence:
{ “__key__”: “images17/image194”, “left.jpg”: b”...”, “right.jpg”: b”...”, “json”: b”...”}
{ “__key__”: “images17/image12”, “left.jpg”: b”...”, “right.jpg”: b”...”, “json”: b”...”}
| reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
``` | 99 | Webdataset: KeyError: 'png' on some datasets when streaming
reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
```
More details in the spec (https://docs.google.com/document/d/18OdLjruFNX74ILmgrdiCI9J1fQZuhzzRBCHV9URWto0/edit#heading=h.hkptaq2kct2s)
> The prefix of a file is all directory components of the file plus the file name component up to the first “.” in the file name.
> The last extension (i.e., the portion after the last “.”) in a file name determines the file type.
> Example:
images17/image194.left.jpg
images17/image194.right.jpg
images17/image194.json
images17/image12.left.jpg
images17/image12.json
images17/image12.right.jpg
images3/image1459.left.jpg
> …
> When reading this with a WebDataset library, you would get the following two dictionaries back in sequence:
{ “__key__”: “images17/image194”, “left.jpg”: b”...”, “right.jpg”: b”...”, “json”: b”...”}
{ “__key__”: “images17/image12”, “left.jpg”: b”...”, “right.jpg”: b”...”, “json”: b”...”}
| [
-0.311941534280777,
-0.2119862288236618,
-0.1765904724597931,
0.1928684115409851,
0.29888659715652466,
0.041977062821388245,
0.16332295536994934,
0.4688340425491333,
0.017786100506782532,
-0.06651580333709717,
-0.092808797955513,
0.0658211037516594,
-0.20402628183364868,
0.1120481416583061... |
https://github.com/huggingface/datasets/issues/6880 | Webdataset: KeyError: 'png' on some datasets when streaming | OK, the issue is different in the latter case: some files are suffixed as `.jpeg`, and others as `.jpg` :)
Is it a limitation of the webdataset format, or of the datasets library @lhoestq? And could we be able to give a clearer error? | reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
``` | 44 | Webdataset: KeyError: 'png' on some datasets when streaming
reported at https://huggingface.co/datasets/tbone5563/tar_images/discussions/1
```python
>>> from datasets import load_dataset
>>> ds = load_dataset("tbone5563/tar_images")
Downloading data: 100%
1.41G/1.41G [00:48<00:00, 17.2MB/s]
Downloading data: 100%
619M/619M [00:11<00:00, 57.4MB/s]
Generating train split:
970/0 [00:02<00:00, 534.94 examples/s]
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1747 _time = time.time()
-> 1748 for key, record in generator:
1749 if max_shard_size is not None and writer._num_bytes > max_shard_size:
7 frames
[/usr/local/lib/python3.10/dist-packages/datasets/packaged_modules/webdataset/webdataset.py](https://localhost:8080/#) in _generate_examples(self, tar_paths, tar_iterators)
108 for field_name in image_field_names + audio_field_names:
--> 109 example[field_name] = {"path": example["__key__"] + "." + field_name, "bytes": example[field_name]}
110 yield f"{tar_idx}_{example_idx}", example
KeyError: 'png'
The above exception was the direct cause of the following exception:
DatasetGenerationError Traceback (most recent call last)
[<ipython-input-2-8e0fbb7badc9>](https://localhost:8080/#) in <cell line: 3>()
1 from datasets import load_dataset
2
----> 3 ds = load_dataset("tbone5563/tar_images")
[/usr/local/lib/python3.10/dist-packages/datasets/load.py](https://localhost:8080/#) in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2607
2608 # Download and prepare data
-> 2609 builder_instance.download_and_prepare(
2610 download_config=download_config,
2611 download_mode=download_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in download_and_prepare(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)
1025 if num_proc is not None:
1026 prepare_split_kwargs["num_proc"] = num_proc
-> 1027 self._download_and_prepare(
1028 dl_manager=dl_manager,
1029 verification_mode=verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs)
1787
1788 def _download_and_prepare(self, dl_manager, verification_mode, **prepare_splits_kwargs):
-> 1789 super()._download_and_prepare(
1790 dl_manager,
1791 verification_mode,
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _download_and_prepare(self, dl_manager, verification_mode, **prepare_split_kwargs)
1120 try:
1121 # Prepare split will record examples associated to the split
-> 1122 self._prepare_split(split_generator, **prepare_split_kwargs)
1123 except OSError as e:
1124 raise OSError(
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split(self, split_generator, check_duplicate_keys, file_format, num_proc, max_shard_size)
1625 job_id = 0
1626 with pbar:
-> 1627 for job_id, done, content in self._prepare_split_single(
1628 gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
1629 ):
[/usr/local/lib/python3.10/dist-packages/datasets/builder.py](https://localhost:8080/#) in _prepare_split_single(self, gen_kwargs, fpath, file_format, max_shard_size, split_info, check_duplicate_keys, job_id)
1782 if isinstance(e, SchemaInferenceError) and e.__context__ is not None:
1783 e = e.__context__
-> 1784 raise DatasetGenerationError("An error occurred while generating the dataset") from e
1785
1786 yield job_id, True, (total_num_examples, total_num_bytes, writer._features, num_shards, shard_lengths)
DatasetGenerationError: An error occurred while generating the dataset
```
OK, the issue is different in the latter case: some files are suffixed as `.jpeg`, and others as `.jpg` :)
Is it a limitation of the webdataset format, or of the datasets library @lhoestq? And could we be able to give a clearer error? | [
-0.311941534280777,
-0.2119862288236618,
-0.1765904724597931,
0.1928684115409851,
0.29888659715652466,
0.041977062821388245,
0.16332295536994934,
0.4688340425491333,
0.017786100506782532,
-0.06651580333709717,
-0.092808797955513,
0.0658211037516594,
-0.20402628183364868,
0.1120481416583061... |
https://github.com/huggingface/datasets/issues/6877 | OSError: [Errno 24] Too many open files | > ulimit -n 8192 can solve this problem
Would there be a systematic way to do this ? The data loading is part of the [MTEB](https://github.com/embeddings-benchmark/mteb) library | ### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
| 27 | OSError: [Errno 24] Too many open files
### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
> ulimit -n 8192 can solve this problem
Would there be a systematic way to do this ? The data loading is part of the [MTEB](https://github.com/embeddings-benchmark/mteb) library | [
-0.4518268406391144,
-0.28309592604637146,
-0.07178039848804474,
0.4186365604400635,
0.15448938310146332,
0.09496546536684036,
0.23318471014499664,
0.40026846528053284,
-0.014156319200992584,
0.32510289549827576,
-0.19934286177158356,
0.011135597713291645,
-0.23308409750461578,
0.115100480... |
https://github.com/huggingface/datasets/issues/6877 | OSError: [Errno 24] Too many open files | > > ulimit -n 8192 can solve this problem
>
> Would there be a systematic way to do this ? The data loading is part of the [MTEB](https://github.com/embeddings-benchmark/mteb) library
I think we could modify the _prepare_split_single function | ### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
| 38 | OSError: [Errno 24] Too many open files
### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
> > ulimit -n 8192 can solve this problem
>
> Would there be a systematic way to do this ? The data loading is part of the [MTEB](https://github.com/embeddings-benchmark/mteb) library
I think we could modify the _prepare_split_single function | [
-0.4518268406391144,
-0.28309592604637146,
-0.07178039848804474,
0.4186365604400635,
0.15448938310146332,
0.09496546536684036,
0.23318471014499664,
0.40026846528053284,
-0.014156319200992584,
0.32510289549827576,
-0.19934286177158356,
0.011135597713291645,
-0.23308409750461578,
0.115100480... |
https://github.com/huggingface/datasets/issues/6877 | OSError: [Errno 24] Too many open files | I fixed it with https://github.com/huggingface/datasets/pull/6893, feel free to re-open if you're still having the issue :) | ### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
| 16 | OSError: [Errno 24] Too many open files
### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
I fixed it with https://github.com/huggingface/datasets/pull/6893, feel free to re-open if you're still having the issue :) | [
-0.4518268406391144,
-0.28309592604637146,
-0.07178039848804474,
0.4186365604400635,
0.15448938310146332,
0.09496546536684036,
0.23318471014499664,
0.40026846528053284,
-0.014156319200992584,
0.32510289549827576,
-0.19934286177158356,
0.011135597713291645,
-0.23308409750461578,
0.115100480... |
https://github.com/huggingface/datasets/issues/6877 | OSError: [Errno 24] Too many open files | > I fixed it with #6893, feel free to re-open if you're still having the issue :)
Thanks a lot! | ### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
| 20 | OSError: [Errno 24] Too many open files
### Describe the bug
I am trying to load the 'default' subset of the following dataset which contains lots of files (828 per split): [https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb](https://huggingface.co/datasets/mteb/biblenlp-corpus-mmteb)
When trying to load it using the `load_dataset` function I get the following error
```python
>>> from datasets import load_dataset
>>> d = load_dataset('mteb/biblenlp-corpus-mmteb')
Downloading readme: 100%|████████████████████████████████████████████████████████████████████████| 201k/201k [00:00<00:00, 1.07MB/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 1069.15it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 436182.33it/s]
Resolving data files: 100%|█████████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 2228.75it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 646478.73it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 831032.24it/s]
Resolving data files: 100%|███████████████████████████████████████████████████████████████████| 828/828 [00:00<00:00, 517645.51it/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:33<00:00, 24.87files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 27.48files/s]
Downloading data: 100%|████████████████████████████████████████████████████████████████████████| 828/828 [00:30<00:00, 26.94files/s]
Generating train split: 1571592 examples [00:03, 461438.97 examples/s]
Generating test split: 11163 examples [00:00, 118190.72 examples/s]
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1995, in _prepare_split_single
for _, table in generator:
File ".env/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 99, in _generate_tables
with open(file, "rb") as f:
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/streaming.py", line 75, in wrapper
return function(*args, download_config=download_config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/utils/file_utils.py", line 1224, in xopen
file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/datasets/filesystems/compression.py", line 81, in _open
return self.file.open()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 135, in open
return self.__enter__()
^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/spec.py", line 1293, in open
f = self._open(
^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 197, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 322, in __init__
self._open()
File ".env/lib/python3.12/site-packages/fsspec/implementations/local.py", line 327, in _open
self.f = open(self.path, mode=self.mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/downloads/3a347186abfc0f9c924dde0221d246db758c7232c0101523f04a87c17d696618'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 981, in incomplete_dir
yield tmp_dir
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File ".env/lib/python3.12/site-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 1007, in download_and_prepare
with incomplete_dir(self._output_dir) as tmp_output_dir:
File "/usr/lib/python3.12/contextlib.py", line 158, in __exit__
self.gen.throw(value)
File ".env/lib/python3.12/site-packages/datasets/builder.py", line 988, in incomplete_dir
shutil.rmtree(tmp_dir)
File "/usr/lib/python3.12/shutil.py", line 785, in rmtree
_rmtree_safe_fd(fd, path, onexc)
File "/usr/lib/python3.12/shutil.py", line 661, in _rmtree_safe_fd
onexc(os.scandir, path, err)
File "/usr/lib/python3.12/shutil.py", line 657, in _rmtree_safe_fd
with os.scandir(topfd) as scandir_it:
^^^^^^^^^^^^^^^^^
OSError: [Errno 24] Too many open files: '.cache/huggingface/datasets/mteb___biblenlp-corpus-mmteb/default/0.0.0/3912ed967b0834547f35b2da9470c4976b357c9a.incomplete'
```
I looked for the maximum number of open files on my machine (Ubuntu 24.04) and it seems to be 1024, but even when I try to load a single split (`load_dataset('mteb/biblenlp-corpus-mmteb', split='train')`) I get the same error
### Steps to reproduce the bug
```python
from datasets import load_dataset
d = load_dataset('mteb/biblenlp-corpus-mmteb')
```
### Expected behavior
Load the dataset without error
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.8.0-31-generic-x86_64-with-glibc2.39
- Python version: 3.12.3
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
> I fixed it with #6893, feel free to re-open if you're still having the issue :)
Thanks a lot! | [
-0.4518268406391144,
-0.28309592604637146,
-0.07178039848804474,
0.4186365604400635,
0.15448938310146332,
0.09496546536684036,
0.23318471014499664,
0.40026846528053284,
-0.014156319200992584,
0.32510289549827576,
-0.19934286177158356,
0.011135597713291645,
-0.23308409750461578,
0.115100480... |
https://github.com/huggingface/datasets/issues/6868 | datasets.BuilderConfig does not work. | I guess the issue is caused by the customization of BuilderConfig that you use from the repo [https://github.com/BeyonderXX/InstructUIE](https://github.com/BeyonderXX/InstructUIE/blob/master/src/uie_dataset.py). You should report to them.
I see you already opened an issue in their repo:
- https://github.com/BeyonderXX/InstructUIE/issues/40 | ### Describe the bug
I custom a BuilderConfig and GeneratorBasedBuilder.
Here is the code for BuilderConfig
```
class UIEConfig(datasets.BuilderConfig):
def __init__(
self,
*args,
data_dir=None,
instruction_file=None,
instruction_strategy=None,
task_config_dir=None,
num_examples=None,
max_num_instances_per_task=None,
max_num_instances_per_eval_task=None,
over_sampling=None,
**kwargs
):
super().__init__(*args, **kwargs)
self.data_dir = data_dir
self.num_examples = num_examples
self.over_sampling = over_sampling
self.instructions = self._parse_instruction(instruction_file)
self.task_configs = self._parse_task_config(task_config_dir)
self.instruction_strategy = instruction_strategy
self.max_num_instances_per_task = max_num_instances_per_task
self.max_num_instances_per_eval_task = max_num_instances_per_eval_task
```
Besides, here is the code for GeneratorBasedBuilder.
```
class UIEInstructions(datasets.GeneratorBasedBuilder):
VERSION = datasets.Version("2.0.0")
BUILDER_CONFIG_CLASS = UIEConfig
BUILDER_CONFIGS = [
UIEConfig(name="default", description="Default config for NaturalInstructions")
]
DEFAULT_CONFIG_NAME = "default"
```
Here is the load_dataset
```
raw_datasets = load_dataset(
os.path.join(CURRENT_DIR, "uie_dataset.py"),
data_dir=data_args.data_dir,
task_config_dir=data_args.task_config_dir,
instruction_file=data_args.instruction_file,
instruction_strategy=data_args.instruction_strategy,
cache_dir=data_cache_dir, # for debug, change dataset size, otherwise open it
max_num_instances_per_task=data_args.max_num_instances_per_task,
max_num_instances_per_eval_task=data_args.max_num_instances_per_eval_task,
num_examples=data_args.num_examples,
over_sampling=data_args.over_sampling
)
```
Finally, I met the error.
```
BuilderConfig UIEConfig(name='default', version=0.0.0, data_dir=None, data_files=None, description='Default config for NaturalInstructions') doesn't have a 'task_config_dir' key.
```
I debugged the code, but I find the parameters added by me may not work.
### Steps to reproduce the bug
https://github.com/BeyonderXX/InstructUIE/blob/master/src/uie_dataset.py
### Expected behavior
```
BuilderConfig UIEConfig(name='default', version=0.0.0, data_dir=None, data_files=None, description='Default config for NaturalInstructions') doesn't have a 'task_config_dir' key.
```
### Environment info
torch 2.3.0+cu118
transformers 4.40.1
python 3.8 | 35 | datasets.BuilderConfig does not work.
### Describe the bug
I custom a BuilderConfig and GeneratorBasedBuilder.
Here is the code for BuilderConfig
```
class UIEConfig(datasets.BuilderConfig):
def __init__(
self,
*args,
data_dir=None,
instruction_file=None,
instruction_strategy=None,
task_config_dir=None,
num_examples=None,
max_num_instances_per_task=None,
max_num_instances_per_eval_task=None,
over_sampling=None,
**kwargs
):
super().__init__(*args, **kwargs)
self.data_dir = data_dir
self.num_examples = num_examples
self.over_sampling = over_sampling
self.instructions = self._parse_instruction(instruction_file)
self.task_configs = self._parse_task_config(task_config_dir)
self.instruction_strategy = instruction_strategy
self.max_num_instances_per_task = max_num_instances_per_task
self.max_num_instances_per_eval_task = max_num_instances_per_eval_task
```
Besides, here is the code for GeneratorBasedBuilder.
```
class UIEInstructions(datasets.GeneratorBasedBuilder):
VERSION = datasets.Version("2.0.0")
BUILDER_CONFIG_CLASS = UIEConfig
BUILDER_CONFIGS = [
UIEConfig(name="default", description="Default config for NaturalInstructions")
]
DEFAULT_CONFIG_NAME = "default"
```
Here is the load_dataset
```
raw_datasets = load_dataset(
os.path.join(CURRENT_DIR, "uie_dataset.py"),
data_dir=data_args.data_dir,
task_config_dir=data_args.task_config_dir,
instruction_file=data_args.instruction_file,
instruction_strategy=data_args.instruction_strategy,
cache_dir=data_cache_dir, # for debug, change dataset size, otherwise open it
max_num_instances_per_task=data_args.max_num_instances_per_task,
max_num_instances_per_eval_task=data_args.max_num_instances_per_eval_task,
num_examples=data_args.num_examples,
over_sampling=data_args.over_sampling
)
```
Finally, I met the error.
```
BuilderConfig UIEConfig(name='default', version=0.0.0, data_dir=None, data_files=None, description='Default config for NaturalInstructions') doesn't have a 'task_config_dir' key.
```
I debugged the code, but I find the parameters added by me may not work.
### Steps to reproduce the bug
https://github.com/BeyonderXX/InstructUIE/blob/master/src/uie_dataset.py
### Expected behavior
```
BuilderConfig UIEConfig(name='default', version=0.0.0, data_dir=None, data_files=None, description='Default config for NaturalInstructions') doesn't have a 'task_config_dir' key.
```
### Environment info
torch 2.3.0+cu118
transformers 4.40.1
python 3.8
I guess the issue is caused by the customization of BuilderConfig that you use from the repo [https://github.com/BeyonderXX/InstructUIE](https://github.com/BeyonderXX/InstructUIE/blob/master/src/uie_dataset.py). You should report to them.
I see you already opened an issue in their repo:
- https://github.com/BeyonderXX/InstructUIE/issues/40 | [
-0.6874491572380066,
0.22365805506706238,
0.019618958234786987,
0.18071717023849487,
0.13669389486312866,
-0.022435389459133148,
0.18786385655403137,
0.03343265503644943,
-0.19063644111156464,
0.21181687712669373,
0.17358554899692535,
0.3953530192375183,
-0.14363087713718414,
-0.0016871523... |
https://github.com/huggingface/datasets/issues/6867 | Improve performance of JSON loader | Thanks! Feel free to ping me for examples. May not respond immediately because we're all busy but would like to help. | As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson | 21 | Improve performance of JSON loader
As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson
Thanks! Feel free to ping me for examples. May not respond immediately because we're all busy but would like to help. | [
-0.03919713571667671,
0.2434317171573639,
-0.17889869213104248,
0.1696496307849884,
0.09725135564804077,
0.08846461027860641,
0.2441028356552124,
0.4211943745613098,
0.5134557485580444,
-0.07182447612285614,
0.08842527121305466,
0.6556617617607117,
-0.03109866753220558,
0.21843014657497406... |
https://github.com/huggingface/datasets/issues/6867 | Improve performance of JSON loader | Hi @natolambert, could you please give some examples of JSON files to benchmark?
Please note that this JSON file (https://huggingface.co/datasets/allenai/reward-bench-results/blob/main/eval-set-scores/Ray2333/reward-model-Mistral-7B-instruct-Unified-Feedback.json) is not in "records" orient; instead it has the following structure:
```json
{
"chat_template": "tulu",
"id": [30, 34, 35,...],
"model": "Ray2333/reward-model-Mistral-7B-instruct-Unified-Feedback",
"model_type": "Seq. Classifier",
"results": [1, 1, 1, ...],
"scores_chosen": [4.421875, 1.8916015625, 3.8515625,...],
"scores_rejected": [-2.416015625, -1.47265625, -0.9912109375,...],
"subset": ["alpacaeval-easy", "alpacaeval-easy", "alpacaeval-easy",...]
"text_chosen": ["<s>[INST] How do I detail a...",...],
"text_rejected": ["<s>[INST] How do I detail a...",...]
}
```
Note that "records" orient should be a list (not a dict) with each row as one item of the list:
```json
[
{"chat_template": "tulu", "id": 30,... },
{"chat_template": "tulu", "id": 34,... },
...
]
``` | As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson | 112 | Improve performance of JSON loader
As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson
Hi @natolambert, could you please give some examples of JSON files to benchmark?
Please note that this JSON file (https://huggingface.co/datasets/allenai/reward-bench-results/blob/main/eval-set-scores/Ray2333/reward-model-Mistral-7B-instruct-Unified-Feedback.json) is not in "records" orient; instead it has the following structure:
```json
{
"chat_template": "tulu",
"id": [30, 34, 35,...],
"model": "Ray2333/reward-model-Mistral-7B-instruct-Unified-Feedback",
"model_type": "Seq. Classifier",
"results": [1, 1, 1, ...],
"scores_chosen": [4.421875, 1.8916015625, 3.8515625,...],
"scores_rejected": [-2.416015625, -1.47265625, -0.9912109375,...],
"subset": ["alpacaeval-easy", "alpacaeval-easy", "alpacaeval-easy",...]
"text_chosen": ["<s>[INST] How do I detail a...",...],
"text_rejected": ["<s>[INST] How do I detail a...",...]
}
```
Note that "records" orient should be a list (not a dict) with each row as one item of the list:
```json
[
{"chat_template": "tulu", "id": 30,... },
{"chat_template": "tulu", "id": 34,... },
...
]
``` | [
-0.04002209007740021,
0.2650395333766937,
-0.1358243077993393,
0.30585044622421265,
-0.07673455774784088,
0.08662878721952438,
0.10383865237236023,
0.5155295133590698,
0.3769634962081909,
-0.12184146046638489,
0.016211044043302536,
0.693203330039978,
0.02511027827858925,
0.0849677994847297... |
https://github.com/huggingface/datasets/issues/6867 | Improve performance of JSON loader | We use a mix (which is a mess), here's an example with the records orient
https://huggingface.co/datasets/allenai/reward-bench-results/blob/main/best-of-n/alpaca_eval/tulu-13b/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5.json
There are more in that folder, ~40mb maybe? | As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson | 24 | Improve performance of JSON loader
As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson
We use a mix (which is a mess), here's an example with the records orient
https://huggingface.co/datasets/allenai/reward-bench-results/blob/main/best-of-n/alpaca_eval/tulu-13b/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5.json
There are more in that folder, ~40mb maybe? | [
-0.045264944434165955,
0.34829017519950867,
-0.19149582087993622,
0.3720752000808716,
0.08937397599220276,
0.10532016307115555,
0.16974525153636932,
0.5300548076629639,
0.38451218605041504,
-0.018146663904190063,
-0.07998569309711456,
0.566435694694519,
-0.05782806873321533,
0.168301939964... |
https://github.com/huggingface/datasets/issues/6867 | Improve performance of JSON loader | @albertvillanova here's a snippet so you don't need to click
```
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
0
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.076171875
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
1
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.87890625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
2
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.287109375
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
3
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 1.6337890625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
4
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 5.27734375
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
5
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.0625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
6
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 2.29296875
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
7
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 6.77734375
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
8
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.853515625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
9
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 4.86328125
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
10
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 2.890625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
11
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 4.70703125
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
12
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 4.45703125
}
``` | As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson | 207 | Improve performance of JSON loader
As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson
@albertvillanova here's a snippet so you don't need to click
```
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
0
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.076171875
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
1
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.87890625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
2
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.287109375
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
3
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 1.6337890625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
4
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 5.27734375
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
5
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.0625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
6
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 2.29296875
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
7
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 6.77734375
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
8
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 3.853515625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
9
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 4.86328125
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
10
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 2.890625
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
11
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 4.70703125
}
{
"config": "top_p=0.9;temp=1.0",
"dataset_details": "helpful_base",
"id": [
0,
12
],
"model": "allenai/tulu-2-dpo-13b",
"scores": 4.45703125
}
``` | [
-0.2909109890460968,
0.18723493814468384,
-0.21109522879123688,
0.18613563477993011,
0.23768350481987,
0.09412505477666855,
0.2980051040649414,
0.541363537311554,
0.3936551809310913,
0.02537570893764496,
-0.09020436555147171,
0.6553610563278198,
0.1419888138771057,
0.24535055458545685,
-... |
https://github.com/huggingface/datasets/issues/6867 | Improve performance of JSON loader | Thanks again for your feedback, @natolambert.
However, strictly speaking, the last file is not in JSON format but in kind of JSON-Lines like format (although not properly either because there are multiple newline characters within each object). Not even pandas can read that file format.
Anyway, for JSON-Lines, I would expect that `datasets` and `pandas` have the same performance for JSON Lines files, as both use `pyarrow` under the hood...
A proper JSON file in records orient should be a list (a JSON array): the first character should be `[`.
Anyway, I am generating a JSON file from your JSON-Lines file to test performance. | As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson | 104 | Improve performance of JSON loader
As reported by @natolambert, loading regular JSON files with `datasets` shows poor performance.
The cause is that we use the `json` Python standard library instead of other faster libraries. See my old comment: https://github.com/huggingface/datasets/pull/2638#pullrequestreview-706983714
> There are benchmarks that compare different JSON packages, with the Standard Library one among the worst performant:
> - https://github.com/ultrajson/ultrajson#benchmarks
> - https://github.com/ijl/orjson#performance
I remember having a discussion about this and it was decided that it was better not to include an additional dependency on a 3rd-party library.
However:
- We already depend on `pandas` and `pandas` depends on `ujson`: so we have an indirect dependency on `ujson`
- Even if the above were not the case, we always could include `ujson` as an optional extra dependency, and check at runtime if it is installed to decide which library to use, either json or ujson
Thanks again for your feedback, @natolambert.
However, strictly speaking, the last file is not in JSON format but in kind of JSON-Lines like format (although not properly either because there are multiple newline characters within each object). Not even pandas can read that file format.
Anyway, for JSON-Lines, I would expect that `datasets` and `pandas` have the same performance for JSON Lines files, as both use `pyarrow` under the hood...
A proper JSON file in records orient should be a list (a JSON array): the first character should be `[`.
Anyway, I am generating a JSON file from your JSON-Lines file to test performance. | [
-0.29499319195747375,
0.25726717710494995,
-0.18426750600337982,
0.1834651231765747,
0.199994757771492,
0.025203242897987366,
0.3841243386268616,
0.4907195568084717,
0.2896990478038788,
-0.06192803382873535,
-0.14312726259231567,
0.5697386264801025,
0.13296867907047272,
0.06747961044311523... |
https://github.com/huggingface/datasets/issues/6866 | DataFilesNotFoundError for datasets in the open-llm-leaderboard | Hi @jerome-white, thnaks for reporting.
However, I cannot reproduce your issue:
```python
>>> from datasets import get_dataset_config_names
>>> get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5")
['harness_arc_challenge_25',
'harness_gsm8k_5',
'harness_hellaswag_10',
'harness_hendrycksTest_5',
'harness_hendrycksTest_abstract_algebra_5',
'harness_hendrycksTest_anatomy_5',
'harness_hendrycksTest_astronomy_5',
'harness_hendrycksTest_business_ethics_5',
'harness_hendrycksTest_clinical_knowledge_5',
'harness_hendrycksTest_college_biology_5',
'harness_hendrycksTest_college_chemistry_5',
'harness_hendrycksTest_college_computer_science_5',
'harness_hendrycksTest_college_mathematics_5',
'harness_hendrycksTest_college_medicine_5',
'harness_hendrycksTest_college_physics_5',
'harness_hendrycksTest_computer_security_5',
'harness_hendrycksTest_conceptual_physics_5',
'harness_hendrycksTest_econometrics_5',
'harness_hendrycksTest_electrical_engineering_5',
'harness_hendrycksTest_elementary_mathematics_5',
'harness_hendrycksTest_formal_logic_5',
'harness_hendrycksTest_global_facts_5',
'harness_hendrycksTest_high_school_biology_5',
'harness_hendrycksTest_high_school_chemistry_5',
'harness_hendrycksTest_high_school_computer_science_5',
'harness_hendrycksTest_high_school_european_history_5',
'harness_hendrycksTest_high_school_geography_5',
'harness_hendrycksTest_high_school_government_and_politics_5',
'harness_hendrycksTest_high_school_macroeconomics_5',
'harness_hendrycksTest_high_school_mathematics_5',
'harness_hendrycksTest_high_school_microeconomics_5',
'harness_hendrycksTest_high_school_physics_5',
'harness_hendrycksTest_high_school_psychology_5',
'harness_hendrycksTest_high_school_statistics_5',
'harness_hendrycksTest_high_school_us_history_5',
'harness_hendrycksTest_high_school_world_history_5',
'harness_hendrycksTest_human_aging_5',
'harness_hendrycksTest_human_sexuality_5',
'harness_hendrycksTest_international_law_5',
'harness_hendrycksTest_jurisprudence_5',
'harness_hendrycksTest_logical_fallacies_5',
'harness_hendrycksTest_machine_learning_5',
'harness_hendrycksTest_management_5',
'harness_hendrycksTest_marketing_5',
'harness_hendrycksTest_medical_genetics_5',
'harness_hendrycksTest_miscellaneous_5',
'harness_hendrycksTest_moral_disputes_5',
'harness_hendrycksTest_moral_scenarios_5',
'harness_hendrycksTest_nutrition_5',
'harness_hendrycksTest_philosophy_5',
'harness_hendrycksTest_prehistory_5',
'harness_hendrycksTest_professional_accounting_5',
'harness_hendrycksTest_professional_law_5',
'harness_hendrycksTest_professional_medicine_5',
'harness_hendrycksTest_professional_psychology_5',
'harness_hendrycksTest_public_relations_5',
'harness_hendrycksTest_security_studies_5',
'harness_hendrycksTest_sociology_5',
'harness_hendrycksTest_us_foreign_policy_5',
'harness_hendrycksTest_virology_5',
'harness_hendrycksTest_world_religions_5',
'harness_truthfulqa_mc_0',
'harness_winogrande_5',
'results']
```
Maybe it was just a temporary issue... | ### Describe the bug
When trying to get config names or load any dataset within the open-llm-leaderboard ecosystem (`open-llm-leaderboard/details_`) I receive the DataFilesNotFoundError. For the last month or so I've been loading datasets from the leaderboard almost everyday; yesterday was the first time I started seeing this.
### Steps to reproduce the bug
This snippet has three cells:
1. Loads the modules
2. Tries to get config names
3. Tries to load the dataset
I've chosen "davidkim205"'s Rhea-72b-v0.5 model because it is one of the best performers on the leaderboard should likely have no dataset issues:
```python
In [1]: from datasets import load_dataset, get_dataset_config_names
In [2]: get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea
...: -72b-v0.5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[2], line 1
----> 1 get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/inspect.py:347, in get_dataset_config_names(path, revision, download_config, download_mode, dynamic_modules_path, data_files, **download_kwargs)
291 def get_dataset_config_names(
292 path: str,
293 revision: Optional[Union[str, Version]] = None,
(...)
298 **download_kwargs,
299 ):
300 """Get the list of available config names for a particular dataset.
301
302 Args:
(...)
345 ```
346 """
--> 347 dataset_module = dataset_module_factory(
348 path,
349 revision=revision,
350 download_config=download_config,
351 download_mode=download_mode,
352 dynamic_modules_path=dynamic_modules_path,
353 data_files=data_files,
354 **download_kwargs,
355 )
356 builder_cls = get_dataset_builder_class(dataset_module, dataset_name=os.path.basename(path))
357 return list(builder_cls.builder_configs.keys()) or [
358 dataset_module.builder_kwargs.get("config_name", builder_cls.DEFAULT_CONFIG_NAME or "default")
359 ]
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
In [3]: data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-
...: v0.5", "harness_winogrande_5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[3], line 1
----> 1 data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5", "harness_winogrande_5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2587, in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2582 verification_mode = VerificationMode(
2583 (verification_mode or VerificationMode.BASIC_CHECKS) if not save_infos else VerificationMode.ALL_CHECKS
2584 )
2586 # Create a dataset builder
-> 2587 builder_instance = load_dataset_builder(
2588 path=path,
2589 name=name,
2590 data_dir=data_dir,
2591 data_files=data_files,
2592 cache_dir=cache_dir,
2593 features=features,
2594 download_config=download_config,
2595 download_mode=download_mode,
2596 revision=revision,
2597 token=token,
2598 storage_options=storage_options,
2599 trust_remote_code=trust_remote_code,
2600 _require_default_config_name=name is None,
2601 **config_kwargs,
2602 )
2604 # Return iterable dataset in case of streaming
2605 if streaming:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2259, in load_dataset_builder(path, name, data_dir, data_files, cache_dir, features, download_config, download_mode, revision, token, use_auth_token, storage_options, trust_remote_code, _require_default_config_name, **config_kwargs)
2257 download_config = download_config.copy() if download_config else DownloadConfig()
2258 download_config.storage_options.update(storage_options)
-> 2259 dataset_module = dataset_module_factory(
2260 path,
2261 revision=revision,
2262 download_config=download_config,
2263 download_mode=download_mode,
2264 data_dir=data_dir,
2265 data_files=data_files,
2266 cache_dir=cache_dir,
2267 trust_remote_code=trust_remote_code,
2268 _require_default_config_name=_require_default_config_name,
2269 _require_custom_configs=bool(config_kwargs),
2270 )
2271 # Get dataset builder class from the processing script
2272 builder_kwargs = dataset_module.builder_kwargs
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
```
### Expected behavior
No exceptions from `get_dataset_config_names` or `load_dataset`
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.5.0-1018-aws-aarch64-with-glibc2.35
- Python version: 3.11.8
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1 | 91 | DataFilesNotFoundError for datasets in the open-llm-leaderboard
### Describe the bug
When trying to get config names or load any dataset within the open-llm-leaderboard ecosystem (`open-llm-leaderboard/details_`) I receive the DataFilesNotFoundError. For the last month or so I've been loading datasets from the leaderboard almost everyday; yesterday was the first time I started seeing this.
### Steps to reproduce the bug
This snippet has three cells:
1. Loads the modules
2. Tries to get config names
3. Tries to load the dataset
I've chosen "davidkim205"'s Rhea-72b-v0.5 model because it is one of the best performers on the leaderboard should likely have no dataset issues:
```python
In [1]: from datasets import load_dataset, get_dataset_config_names
In [2]: get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea
...: -72b-v0.5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[2], line 1
----> 1 get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/inspect.py:347, in get_dataset_config_names(path, revision, download_config, download_mode, dynamic_modules_path, data_files, **download_kwargs)
291 def get_dataset_config_names(
292 path: str,
293 revision: Optional[Union[str, Version]] = None,
(...)
298 **download_kwargs,
299 ):
300 """Get the list of available config names for a particular dataset.
301
302 Args:
(...)
345 ```
346 """
--> 347 dataset_module = dataset_module_factory(
348 path,
349 revision=revision,
350 download_config=download_config,
351 download_mode=download_mode,
352 dynamic_modules_path=dynamic_modules_path,
353 data_files=data_files,
354 **download_kwargs,
355 )
356 builder_cls = get_dataset_builder_class(dataset_module, dataset_name=os.path.basename(path))
357 return list(builder_cls.builder_configs.keys()) or [
358 dataset_module.builder_kwargs.get("config_name", builder_cls.DEFAULT_CONFIG_NAME or "default")
359 ]
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
In [3]: data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-
...: v0.5", "harness_winogrande_5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[3], line 1
----> 1 data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5", "harness_winogrande_5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2587, in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2582 verification_mode = VerificationMode(
2583 (verification_mode or VerificationMode.BASIC_CHECKS) if not save_infos else VerificationMode.ALL_CHECKS
2584 )
2586 # Create a dataset builder
-> 2587 builder_instance = load_dataset_builder(
2588 path=path,
2589 name=name,
2590 data_dir=data_dir,
2591 data_files=data_files,
2592 cache_dir=cache_dir,
2593 features=features,
2594 download_config=download_config,
2595 download_mode=download_mode,
2596 revision=revision,
2597 token=token,
2598 storage_options=storage_options,
2599 trust_remote_code=trust_remote_code,
2600 _require_default_config_name=name is None,
2601 **config_kwargs,
2602 )
2604 # Return iterable dataset in case of streaming
2605 if streaming:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2259, in load_dataset_builder(path, name, data_dir, data_files, cache_dir, features, download_config, download_mode, revision, token, use_auth_token, storage_options, trust_remote_code, _require_default_config_name, **config_kwargs)
2257 download_config = download_config.copy() if download_config else DownloadConfig()
2258 download_config.storage_options.update(storage_options)
-> 2259 dataset_module = dataset_module_factory(
2260 path,
2261 revision=revision,
2262 download_config=download_config,
2263 download_mode=download_mode,
2264 data_dir=data_dir,
2265 data_files=data_files,
2266 cache_dir=cache_dir,
2267 trust_remote_code=trust_remote_code,
2268 _require_default_config_name=_require_default_config_name,
2269 _require_custom_configs=bool(config_kwargs),
2270 )
2271 # Get dataset builder class from the processing script
2272 builder_kwargs = dataset_module.builder_kwargs
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
```
### Expected behavior
No exceptions from `get_dataset_config_names` or `load_dataset`
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.5.0-1018-aws-aarch64-with-glibc2.35
- Python version: 3.11.8
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
Hi @jerome-white, thnaks for reporting.
However, I cannot reproduce your issue:
```python
>>> from datasets import get_dataset_config_names
>>> get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5")
['harness_arc_challenge_25',
'harness_gsm8k_5',
'harness_hellaswag_10',
'harness_hendrycksTest_5',
'harness_hendrycksTest_abstract_algebra_5',
'harness_hendrycksTest_anatomy_5',
'harness_hendrycksTest_astronomy_5',
'harness_hendrycksTest_business_ethics_5',
'harness_hendrycksTest_clinical_knowledge_5',
'harness_hendrycksTest_college_biology_5',
'harness_hendrycksTest_college_chemistry_5',
'harness_hendrycksTest_college_computer_science_5',
'harness_hendrycksTest_college_mathematics_5',
'harness_hendrycksTest_college_medicine_5',
'harness_hendrycksTest_college_physics_5',
'harness_hendrycksTest_computer_security_5',
'harness_hendrycksTest_conceptual_physics_5',
'harness_hendrycksTest_econometrics_5',
'harness_hendrycksTest_electrical_engineering_5',
'harness_hendrycksTest_elementary_mathematics_5',
'harness_hendrycksTest_formal_logic_5',
'harness_hendrycksTest_global_facts_5',
'harness_hendrycksTest_high_school_biology_5',
'harness_hendrycksTest_high_school_chemistry_5',
'harness_hendrycksTest_high_school_computer_science_5',
'harness_hendrycksTest_high_school_european_history_5',
'harness_hendrycksTest_high_school_geography_5',
'harness_hendrycksTest_high_school_government_and_politics_5',
'harness_hendrycksTest_high_school_macroeconomics_5',
'harness_hendrycksTest_high_school_mathematics_5',
'harness_hendrycksTest_high_school_microeconomics_5',
'harness_hendrycksTest_high_school_physics_5',
'harness_hendrycksTest_high_school_psychology_5',
'harness_hendrycksTest_high_school_statistics_5',
'harness_hendrycksTest_high_school_us_history_5',
'harness_hendrycksTest_high_school_world_history_5',
'harness_hendrycksTest_human_aging_5',
'harness_hendrycksTest_human_sexuality_5',
'harness_hendrycksTest_international_law_5',
'harness_hendrycksTest_jurisprudence_5',
'harness_hendrycksTest_logical_fallacies_5',
'harness_hendrycksTest_machine_learning_5',
'harness_hendrycksTest_management_5',
'harness_hendrycksTest_marketing_5',
'harness_hendrycksTest_medical_genetics_5',
'harness_hendrycksTest_miscellaneous_5',
'harness_hendrycksTest_moral_disputes_5',
'harness_hendrycksTest_moral_scenarios_5',
'harness_hendrycksTest_nutrition_5',
'harness_hendrycksTest_philosophy_5',
'harness_hendrycksTest_prehistory_5',
'harness_hendrycksTest_professional_accounting_5',
'harness_hendrycksTest_professional_law_5',
'harness_hendrycksTest_professional_medicine_5',
'harness_hendrycksTest_professional_psychology_5',
'harness_hendrycksTest_public_relations_5',
'harness_hendrycksTest_security_studies_5',
'harness_hendrycksTest_sociology_5',
'harness_hendrycksTest_us_foreign_policy_5',
'harness_hendrycksTest_virology_5',
'harness_hendrycksTest_world_religions_5',
'harness_truthfulqa_mc_0',
'harness_winogrande_5',
'results']
```
Maybe it was just a temporary issue... | [
-0.49482882022857666,
0.3736054599285126,
0.0811573714017868,
0.6338543891906738,
0.08500688523054123,
-0.10712634027004242,
-0.016161341220140457,
0.34829655289649963,
0.10186594724655151,
0.34748297929763794,
0.05879886448383331,
0.21926254034042358,
-0.1991540938615799,
-0.1078597977757... |
https://github.com/huggingface/datasets/issues/6866 | DataFilesNotFoundError for datasets in the open-llm-leaderboard | > Maybe it was just a temporary issue...
Perhaps. I've changed my workflow to use the hub's `HfFileSystem`, so for now this is no longer a blocker for me. I'll reopen the issue if that changes. | ### Describe the bug
When trying to get config names or load any dataset within the open-llm-leaderboard ecosystem (`open-llm-leaderboard/details_`) I receive the DataFilesNotFoundError. For the last month or so I've been loading datasets from the leaderboard almost everyday; yesterday was the first time I started seeing this.
### Steps to reproduce the bug
This snippet has three cells:
1. Loads the modules
2. Tries to get config names
3. Tries to load the dataset
I've chosen "davidkim205"'s Rhea-72b-v0.5 model because it is one of the best performers on the leaderboard should likely have no dataset issues:
```python
In [1]: from datasets import load_dataset, get_dataset_config_names
In [2]: get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea
...: -72b-v0.5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[2], line 1
----> 1 get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/inspect.py:347, in get_dataset_config_names(path, revision, download_config, download_mode, dynamic_modules_path, data_files, **download_kwargs)
291 def get_dataset_config_names(
292 path: str,
293 revision: Optional[Union[str, Version]] = None,
(...)
298 **download_kwargs,
299 ):
300 """Get the list of available config names for a particular dataset.
301
302 Args:
(...)
345 ```
346 """
--> 347 dataset_module = dataset_module_factory(
348 path,
349 revision=revision,
350 download_config=download_config,
351 download_mode=download_mode,
352 dynamic_modules_path=dynamic_modules_path,
353 data_files=data_files,
354 **download_kwargs,
355 )
356 builder_cls = get_dataset_builder_class(dataset_module, dataset_name=os.path.basename(path))
357 return list(builder_cls.builder_configs.keys()) or [
358 dataset_module.builder_kwargs.get("config_name", builder_cls.DEFAULT_CONFIG_NAME or "default")
359 ]
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
In [3]: data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-
...: v0.5", "harness_winogrande_5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[3], line 1
----> 1 data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5", "harness_winogrande_5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2587, in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2582 verification_mode = VerificationMode(
2583 (verification_mode or VerificationMode.BASIC_CHECKS) if not save_infos else VerificationMode.ALL_CHECKS
2584 )
2586 # Create a dataset builder
-> 2587 builder_instance = load_dataset_builder(
2588 path=path,
2589 name=name,
2590 data_dir=data_dir,
2591 data_files=data_files,
2592 cache_dir=cache_dir,
2593 features=features,
2594 download_config=download_config,
2595 download_mode=download_mode,
2596 revision=revision,
2597 token=token,
2598 storage_options=storage_options,
2599 trust_remote_code=trust_remote_code,
2600 _require_default_config_name=name is None,
2601 **config_kwargs,
2602 )
2604 # Return iterable dataset in case of streaming
2605 if streaming:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2259, in load_dataset_builder(path, name, data_dir, data_files, cache_dir, features, download_config, download_mode, revision, token, use_auth_token, storage_options, trust_remote_code, _require_default_config_name, **config_kwargs)
2257 download_config = download_config.copy() if download_config else DownloadConfig()
2258 download_config.storage_options.update(storage_options)
-> 2259 dataset_module = dataset_module_factory(
2260 path,
2261 revision=revision,
2262 download_config=download_config,
2263 download_mode=download_mode,
2264 data_dir=data_dir,
2265 data_files=data_files,
2266 cache_dir=cache_dir,
2267 trust_remote_code=trust_remote_code,
2268 _require_default_config_name=_require_default_config_name,
2269 _require_custom_configs=bool(config_kwargs),
2270 )
2271 # Get dataset builder class from the processing script
2272 builder_kwargs = dataset_module.builder_kwargs
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
```
### Expected behavior
No exceptions from `get_dataset_config_names` or `load_dataset`
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.5.0-1018-aws-aarch64-with-glibc2.35
- Python version: 3.11.8
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1 | 36 | DataFilesNotFoundError for datasets in the open-llm-leaderboard
### Describe the bug
When trying to get config names or load any dataset within the open-llm-leaderboard ecosystem (`open-llm-leaderboard/details_`) I receive the DataFilesNotFoundError. For the last month or so I've been loading datasets from the leaderboard almost everyday; yesterday was the first time I started seeing this.
### Steps to reproduce the bug
This snippet has three cells:
1. Loads the modules
2. Tries to get config names
3. Tries to load the dataset
I've chosen "davidkim205"'s Rhea-72b-v0.5 model because it is one of the best performers on the leaderboard should likely have no dataset issues:
```python
In [1]: from datasets import load_dataset, get_dataset_config_names
In [2]: get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea
...: -72b-v0.5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[2], line 1
----> 1 get_dataset_config_names("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/inspect.py:347, in get_dataset_config_names(path, revision, download_config, download_mode, dynamic_modules_path, data_files, **download_kwargs)
291 def get_dataset_config_names(
292 path: str,
293 revision: Optional[Union[str, Version]] = None,
(...)
298 **download_kwargs,
299 ):
300 """Get the list of available config names for a particular dataset.
301
302 Args:
(...)
345 ```
346 """
--> 347 dataset_module = dataset_module_factory(
348 path,
349 revision=revision,
350 download_config=download_config,
351 download_mode=download_mode,
352 dynamic_modules_path=dynamic_modules_path,
353 data_files=data_files,
354 **download_kwargs,
355 )
356 builder_cls = get_dataset_builder_class(dataset_module, dataset_name=os.path.basename(path))
357 return list(builder_cls.builder_configs.keys()) or [
358 dataset_module.builder_kwargs.get("config_name", builder_cls.DEFAULT_CONFIG_NAME or "default")
359 ]
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
In [3]: data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-
...: v0.5", "harness_winogrande_5")
---------------------------------------------------------------------------
DataFilesNotFoundError Traceback (most recent call last)
Cell In[3], line 1
----> 1 data = load_dataset("open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5", "harness_winogrande_5")
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2587, in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)
2582 verification_mode = VerificationMode(
2583 (verification_mode or VerificationMode.BASIC_CHECKS) if not save_infos else VerificationMode.ALL_CHECKS
2584 )
2586 # Create a dataset builder
-> 2587 builder_instance = load_dataset_builder(
2588 path=path,
2589 name=name,
2590 data_dir=data_dir,
2591 data_files=data_files,
2592 cache_dir=cache_dir,
2593 features=features,
2594 download_config=download_config,
2595 download_mode=download_mode,
2596 revision=revision,
2597 token=token,
2598 storage_options=storage_options,
2599 trust_remote_code=trust_remote_code,
2600 _require_default_config_name=name is None,
2601 **config_kwargs,
2602 )
2604 # Return iterable dataset in case of streaming
2605 if streaming:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:2259, in load_dataset_builder(path, name, data_dir, data_files, cache_dir, features, download_config, download_mode, revision, token, use_auth_token, storage_options, trust_remote_code, _require_default_config_name, **config_kwargs)
2257 download_config = download_config.copy() if download_config else DownloadConfig()
2258 download_config.storage_options.update(storage_options)
-> 2259 dataset_module = dataset_module_factory(
2260 path,
2261 revision=revision,
2262 download_config=download_config,
2263 download_mode=download_mode,
2264 data_dir=data_dir,
2265 data_files=data_files,
2266 cache_dir=cache_dir,
2267 trust_remote_code=trust_remote_code,
2268 _require_default_config_name=_require_default_config_name,
2269 _require_custom_configs=bool(config_kwargs),
2270 )
2271 # Get dataset builder class from the processing script
2272 builder_kwargs = dataset_module.builder_kwargs
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1821, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, cache_dir, trust_remote_code, _require_default_config_name, _require_custom_configs, **download_kwargs)
1812 return LocalDatasetModuleFactoryWithScript(
1813 combined_path,
1814 download_mode=download_mode,
1815 dynamic_modules_path=dynamic_modules_path,
1816 trust_remote_code=trust_remote_code,
1817 ).get_module()
1818 elif os.path.isdir(path):
1819 return LocalDatasetModuleFactoryWithoutScript(
1820 path, data_dir=data_dir, data_files=data_files, download_mode=download_mode
-> 1821 ).get_module()
1822 # Try remotely
1823 elif is_relative_path(path) and path.count("/") <= 1:
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:1039, in LocalDatasetModuleFactoryWithoutScript.get_module(self)
1033 patterns = get_data_patterns(base_path)
1034 data_files = DataFilesDict.from_patterns(
1035 patterns,
1036 base_path=base_path,
1037 allowed_extensions=ALL_ALLOWED_EXTENSIONS,
1038 )
-> 1039 module_name, default_builder_kwargs = infer_module_for_data_files(
1040 data_files=data_files,
1041 path=self.path,
1042 )
1043 data_files = data_files.filter_extensions(_MODULE_TO_EXTENSIONS[module_name])
1044 # Collect metadata files if the module supports them
File ~/open-llm-bda/venv/lib/python3.11/site-packages/datasets/load.py:597, in infer_module_for_data_files(data_files, path, download_config)
595 raise ValueError(f"Couldn't infer the same data file format for all splits. Got {split_modules}")
596 if not module_name:
--> 597 raise DataFilesNotFoundError("No (supported) data files found" + (f" in {path}" if path else ""))
598 return module_name, default_builder_kwargs
DataFilesNotFoundError: No (supported) data files found in open-llm-leaderboard/details_davidkim205__Rhea-72b-v0.5
```
### Expected behavior
No exceptions from `get_dataset_config_names` or `load_dataset`
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-6.5.0-1018-aws-aarch64-with-glibc2.35
- Python version: 3.11.8
- `huggingface_hub` version: 0.23.0
- PyArrow version: 16.0.0
- Pandas version: 2.2.2
- `fsspec` version: 2024.3.1
> Maybe it was just a temporary issue...
Perhaps. I've changed my workflow to use the hub's `HfFileSystem`, so for now this is no longer a blocker for me. I'll reopen the issue if that changes. | [
-0.49482882022857666,
0.3736054599285126,
0.0811573714017868,
0.6338543891906738,
0.08500688523054123,
-0.10712634027004242,
-0.016161341220140457,
0.34829655289649963,
0.10186594724655151,
0.34748297929763794,
0.05879886448383331,
0.21926254034042358,
-0.1991540938615799,
-0.1078597977757... |
https://github.com/huggingface/datasets/issues/6864 | Dataset 'rewardsignal/reddit_writing_prompts' doesn't exist on the Hub | Hi @vinodrajendran001, thanks for reporting.
Indeed the dataset no longer exists on the Hub. The URL https://huggingface.co/datasets/rewardsignal/reddit_writing_prompts gives 404 Not Found error. | ### Describe the bug
The dataset `rewardsignal/reddit_writing_prompts` is missing in Huggingface Hub.
### Steps to reproduce the bug
```
from datasets import load_dataset
prompt_response_dataset = load_dataset("rewardsignal/reddit_writing_prompts", data_files="prompt_responses_full.csv", split='train[:80%]')
```
### Expected behavior
DatasetNotFoundError: Dataset 'rewardsignal/reddit_writing_prompts' doesn't exist on the Hub or cannot be accessed
### Environment info
Nothing to do with versions | 22 | Dataset 'rewardsignal/reddit_writing_prompts' doesn't exist on the Hub
### Describe the bug
The dataset `rewardsignal/reddit_writing_prompts` is missing in Huggingface Hub.
### Steps to reproduce the bug
```
from datasets import load_dataset
prompt_response_dataset = load_dataset("rewardsignal/reddit_writing_prompts", data_files="prompt_responses_full.csv", split='train[:80%]')
```
### Expected behavior
DatasetNotFoundError: Dataset 'rewardsignal/reddit_writing_prompts' doesn't exist on the Hub or cannot be accessed
### Environment info
Nothing to do with versions
Hi @vinodrajendran001, thanks for reporting.
Indeed the dataset no longer exists on the Hub. The URL https://huggingface.co/datasets/rewardsignal/reddit_writing_prompts gives 404 Not Found error. | [
-0.03555293753743172,
-0.45973971486091614,
-0.0076712388545274734,
0.46709901094436646,
0.15462878346443176,
0.25667402148246765,
0.2523685097694397,
0.18569418787956238,
0.18151366710662842,
0.16432809829711914,
-0.07102049887180328,
0.102827288210392,
-0.05865245312452316,
0.31427040696... |
https://github.com/huggingface/datasets/issues/6858 | Segmentation fault | I downloaded the jsonl file and extract it manually.
The issue seems to be related to pyarrow.json
python3 -q -X faulthandler -c "from datasets import load_dataset; load_dataset('json', data_files='/Users/scampion/Downloads/1998-09.jsonl')"
Generating train split: 0 examples [00:00, ? examples/s]Fatal Python error: Segmentation fault
Thread 0x00007000000c1000 (most recent call first):
<no Python frame>
Thread 0x00007000024df000 (most recent call first):
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 331 in wait
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 629 in wait
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/tqdm/_monitor.py", line 60 in run
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 1045 in _bootstrap_inner
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 1002 in _bootstrap
Thread 0x00007ff845c66640 (most recent call first):
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/packaged_modules/json/json.py", line 122 in _generate_tables
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1995 in _prepare_split_single
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1882 in _prepare_split
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1122 in _download_and_prepare
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1027 in download_and_prepare
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/load.py", line 2609 in load_dataset
File "<string>", line 1 in <module>
Extension modules: numpy.core._multiarray_umath, numpy.core._multiarray_tests, numpy.linalg._umath_linalg, numpy.fft._pocketfft_internal, numpy.random._common, numpy.random.bit_generator, numpy.random._bounded_integers, numpy.random._mt19937, numpy.random.mtrand, numpy.random._philox, numpy.random._pcg64, numpy.random._sfc64, numpy.random._generator, pyarrow.lib, pyarrow._hdfsio, pandas._libs.tslibs.ccalendar, pandas._libs.tslibs.np_datetime, pandas._libs.tslibs.dtypes, pandas._libs.tslibs.base, pandas._libs.tslibs.nattype, pandas._libs.tslibs.timezones, pandas._libs.tslibs.fields, pandas._libs.tslibs.timedeltas, pandas._libs.tslibs.tzconversion, pandas._libs.tslibs.timestamps, pandas._libs.properties, pandas._libs.tslibs.offsets, pandas._libs.tslibs.strptime, pandas._libs.tslibs.parsing, pandas._libs.tslibs.conversion, pandas._libs.tslibs.period, pandas._libs.tslibs.vectorized, pandas._libs.ops_dispatch, pandas._libs.missing, pandas._libs.hashtable, pandas._libs.algos, pandas._libs.interval, pandas._libs.lib, pyarrow._compute, pandas._libs.ops, pandas._libs.hashing, pandas._libs.arrays, pandas._libs.tslib, pandas._libs.sparse, pandas._libs.internals, pandas._libs.indexing, pandas._libs.index, pandas._libs.writers, pandas._libs.join, pandas._libs.window.aggregations, pandas._libs.window.indexers, pandas._libs.reshape, pandas._libs.groupby, pandas._libs.json, pandas._libs.parsers, pandas._libs.testing, charset_normalizer.md, yaml._yaml, pyarrow._parquet, pyarrow._fs, pyarrow._hdfs, pyarrow._gcsfs, pyarrow._s3fs, multidict._multidict, yarl._quoting_c, aiohttp._helpers, aiohttp._http_writer, aiohttp._http_parser, aiohttp._websocket, frozenlist._frozenlist, xxhash._xxhash, pyarrow._json (total: 72)
[1] 56678 segmentation fault python3 -q -X faulthandler -c
/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:254: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d '
(venv_test) | ### Describe the bug
Using various version for datasets, I'm no more longer able to load that dataset without a segmentation fault.
Several others files are also concerned.
### Steps to reproduce the bug
# Create a new venv
python3 -m venv venv_test
source venv_test/bin/activate
# Install the latest version
pip install datasets
# Load that dataset
python3 -q -X faulthandler -c "from datasets import load_dataset; load_dataset('EuropeanParliament/Eurovoc', '1998-09')"
### Expected behavior
Data must be loaded
### Environment info
datasets==2.19.0
Python 3.11.7
Darwin 22.5.0 Darwin Kernel Version 22.5.0: Mon Apr 24 20:51:50 PDT 2023; root:xnu-8796.121.2~5/RELEASE_X86_64 x86_64 | 242 | Segmentation fault
### Describe the bug
Using various version for datasets, I'm no more longer able to load that dataset without a segmentation fault.
Several others files are also concerned.
### Steps to reproduce the bug
# Create a new venv
python3 -m venv venv_test
source venv_test/bin/activate
# Install the latest version
pip install datasets
# Load that dataset
python3 -q -X faulthandler -c "from datasets import load_dataset; load_dataset('EuropeanParliament/Eurovoc', '1998-09')"
### Expected behavior
Data must be loaded
### Environment info
datasets==2.19.0
Python 3.11.7
Darwin 22.5.0 Darwin Kernel Version 22.5.0: Mon Apr 24 20:51:50 PDT 2023; root:xnu-8796.121.2~5/RELEASE_X86_64 x86_64
I downloaded the jsonl file and extract it manually.
The issue seems to be related to pyarrow.json
python3 -q -X faulthandler -c "from datasets import load_dataset; load_dataset('json', data_files='/Users/scampion/Downloads/1998-09.jsonl')"
Generating train split: 0 examples [00:00, ? examples/s]Fatal Python error: Segmentation fault
Thread 0x00007000000c1000 (most recent call first):
<no Python frame>
Thread 0x00007000024df000 (most recent call first):
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 331 in wait
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 629 in wait
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/tqdm/_monitor.py", line 60 in run
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 1045 in _bootstrap_inner
File "/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/threading.py", line 1002 in _bootstrap
Thread 0x00007ff845c66640 (most recent call first):
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/packaged_modules/json/json.py", line 122 in _generate_tables
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1995 in _prepare_split_single
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1882 in _prepare_split
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1122 in _download_and_prepare
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/builder.py", line 1027 in download_and_prepare
File "/Users/scampion/src/test/venv_test/lib/python3.11/site-packages/datasets/load.py", line 2609 in load_dataset
File "<string>", line 1 in <module>
Extension modules: numpy.core._multiarray_umath, numpy.core._multiarray_tests, numpy.linalg._umath_linalg, numpy.fft._pocketfft_internal, numpy.random._common, numpy.random.bit_generator, numpy.random._bounded_integers, numpy.random._mt19937, numpy.random.mtrand, numpy.random._philox, numpy.random._pcg64, numpy.random._sfc64, numpy.random._generator, pyarrow.lib, pyarrow._hdfsio, pandas._libs.tslibs.ccalendar, pandas._libs.tslibs.np_datetime, pandas._libs.tslibs.dtypes, pandas._libs.tslibs.base, pandas._libs.tslibs.nattype, pandas._libs.tslibs.timezones, pandas._libs.tslibs.fields, pandas._libs.tslibs.timedeltas, pandas._libs.tslibs.tzconversion, pandas._libs.tslibs.timestamps, pandas._libs.properties, pandas._libs.tslibs.offsets, pandas._libs.tslibs.strptime, pandas._libs.tslibs.parsing, pandas._libs.tslibs.conversion, pandas._libs.tslibs.period, pandas._libs.tslibs.vectorized, pandas._libs.ops_dispatch, pandas._libs.missing, pandas._libs.hashtable, pandas._libs.algos, pandas._libs.interval, pandas._libs.lib, pyarrow._compute, pandas._libs.ops, pandas._libs.hashing, pandas._libs.arrays, pandas._libs.tslib, pandas._libs.sparse, pandas._libs.internals, pandas._libs.indexing, pandas._libs.index, pandas._libs.writers, pandas._libs.join, pandas._libs.window.aggregations, pandas._libs.window.indexers, pandas._libs.reshape, pandas._libs.groupby, pandas._libs.json, pandas._libs.parsers, pandas._libs.testing, charset_normalizer.md, yaml._yaml, pyarrow._parquet, pyarrow._fs, pyarrow._hdfs, pyarrow._gcsfs, pyarrow._s3fs, multidict._multidict, yarl._quoting_c, aiohttp._helpers, aiohttp._http_writer, aiohttp._http_parser, aiohttp._websocket, frozenlist._frozenlist, xxhash._xxhash, pyarrow._json (total: 72)
[1] 56678 segmentation fault python3 -q -X faulthandler -c
/usr/local/Cellar/python@3.11/3.11.7/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:254: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d '
(venv_test) | [
-0.273857444524765,
0.12453977763652802,
0.029024098068475723,
0.24738529324531555,
0.1829131543636322,
0.0015955716371536255,
0.41411513090133667,
0.4311520457267761,
-0.1906261295080185,
-0.05419503152370453,
0.1505376249551773,
0.5418916940689087,
-0.0870991051197052,
-0.333734720945358... |
https://github.com/huggingface/datasets/issues/6856 | CI fails on Windows for test_delete_from_hub and test_xgetsize_private due to new-line character | After investigation, I have found that when a local file is uploaded to the Hub, the new line character is no longer transformed to "\n": on Windows machine now it is kept as "\r\n".
Any idea why this changed?
CC: @lhoestq | CI fails on Windows for test_delete_from_hub after the merge of:
- #6820
This is weird because the CI was green in the PR branch before merging to main.
```
FAILED tests/test_hub.py::test_delete_from_hub - AssertionError: assert [CommitOperat...\r\n---\r\n')] == [CommitOperat...in/*\n---\n')]
At index 1 diff: CommitOperationAdd(path_in_repo='README.md', path_or_fileobj=b'---\r\nconfigs:\r\n- config_name: cats\r\n data_files:\r\n - split: train\r\n path: cats/train/*\r\n---\r\n') != CommitOperationAdd(path_in_repo='README.md', path_or_fileobj=b'---\nconfigs:\n- config_name: cats\n data_files:\n - split: train\n path: cats/train/*\n---\n')
Full diff:
[
CommitOperationDelete(
path_in_repo='dogs/train/0000.csv',
is_folder=False,
),
CommitOperationAdd(
path_in_repo='README.md',
- path_or_fileobj=b'---\nconfigs:\n- config_name: cats\n data_files:\n '
? --------
+ path_or_fileobj=b'---\r\nconfigs:\r\n- config_name: cats\r\n data_f'
? ++ ++ ++
- b' - split: train\n path: cats/train/*\n---\n',
? ^^^^^^ -
+ b'iles:\r\n - split: train\r\n path: cats/train/*\r'
? ++++++++++ ++ ^
+ b'\n---\r\n',
),
]
``` | 41 | CI fails on Windows for test_delete_from_hub and test_xgetsize_private due to new-line character
CI fails on Windows for test_delete_from_hub after the merge of:
- #6820
This is weird because the CI was green in the PR branch before merging to main.
```
FAILED tests/test_hub.py::test_delete_from_hub - AssertionError: assert [CommitOperat...\r\n---\r\n')] == [CommitOperat...in/*\n---\n')]
At index 1 diff: CommitOperationAdd(path_in_repo='README.md', path_or_fileobj=b'---\r\nconfigs:\r\n- config_name: cats\r\n data_files:\r\n - split: train\r\n path: cats/train/*\r\n---\r\n') != CommitOperationAdd(path_in_repo='README.md', path_or_fileobj=b'---\nconfigs:\n- config_name: cats\n data_files:\n - split: train\n path: cats/train/*\n---\n')
Full diff:
[
CommitOperationDelete(
path_in_repo='dogs/train/0000.csv',
is_folder=False,
),
CommitOperationAdd(
path_in_repo='README.md',
- path_or_fileobj=b'---\nconfigs:\n- config_name: cats\n data_files:\n '
? --------
+ path_or_fileobj=b'---\r\nconfigs:\r\n- config_name: cats\r\n data_f'
? ++ ++ ++
- b' - split: train\n path: cats/train/*\n---\n',
? ^^^^^^ -
+ b'iles:\r\n - split: train\r\n path: cats/train/*\r'
? ++++++++++ ++ ^
+ b'\n---\r\n',
),
]
```
After investigation, I have found that when a local file is uploaded to the Hub, the new line character is no longer transformed to "\n": on Windows machine now it is kept as "\r\n".
Any idea why this changed?
CC: @lhoestq | [
-0.020743783563375473,
0.15854910016059875,
-0.02737545594573021,
-0.05879633128643036,
0.2177542895078659,
-0.24898961186408997,
0.4662143588066101,
0.06262174248695374,
0.19253072142601013,
0.33144664764404297,
-0.030337302014231682,
-0.010529213584959507,
0.10585103183984756,
0.24378193... |
https://github.com/huggingface/datasets/issues/6850 | Problem loading voxpopuli dataset | @Namangarg110 @mohsen-goodarzi The bug appears because the number of urls is less than 16 and the algorithm is meant to work on the previously created mode for a single url as stated on line 314: https://github.com/huggingface/datasets/blob/1bf8a46cc7b096d5c547ea3794f6a4b6c31ea762/src/datasets/download/download_manager.py#L314
In addition, previously `map_nested` function was supported without batching and it is meant to be the default performance.
One of the shortest walk-arounds would be changing the part of the manager with the current setting:
```
if len(url_or_urls) >= 16:
download_func = partial(self._download_batched, download_config=download_config)
else:
download_func = partial(self._download_single, download_config=download_config)
start_time = datetime.now()
with stack_multiprocessing_download_progress_bars():
downloaded_path_or_paths = map_nested(
download_func,
url_or_urls,
map_tuple=True,
num_proc=download_config.num_proc,
desc="Downloading data files",
batched=True if len(url_or_urls) >= 16 else False,
batch_size=-1,
)
```
I would suggest to consider other datasets for similar issues and make a pull-request. | ### Describe the bug
```
Exception has occurred: FileNotFoundError
Couldn't find file at https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/{'en': 'data/en/asr_train.tsv'}
```
Error in logic for link url creation. The link should be https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/en/asr_train.tsv
Basically there should be links directly under ```metadata["train"]```, not under ```metadata["train"][self.config.languages[0]]```
same for audio urls
### Steps to reproduce the bug
```
from datasets import load_dataset
dataset = load_dataset("facebook/voxpopuli","en")
```
### Expected behavior
Dataset should be loaded successfully.
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-5.15.0-1041-aws-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.12.2 | 124 | Problem loading voxpopuli dataset
### Describe the bug
```
Exception has occurred: FileNotFoundError
Couldn't find file at https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/{'en': 'data/en/asr_train.tsv'}
```
Error in logic for link url creation. The link should be https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/en/asr_train.tsv
Basically there should be links directly under ```metadata["train"]```, not under ```metadata["train"][self.config.languages[0]]```
same for audio urls
### Steps to reproduce the bug
```
from datasets import load_dataset
dataset = load_dataset("facebook/voxpopuli","en")
```
### Expected behavior
Dataset should be loaded successfully.
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-5.15.0-1041-aws-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.12.2
@Namangarg110 @mohsen-goodarzi The bug appears because the number of urls is less than 16 and the algorithm is meant to work on the previously created mode for a single url as stated on line 314: https://github.com/huggingface/datasets/blob/1bf8a46cc7b096d5c547ea3794f6a4b6c31ea762/src/datasets/download/download_manager.py#L314
In addition, previously `map_nested` function was supported without batching and it is meant to be the default performance.
One of the shortest walk-arounds would be changing the part of the manager with the current setting:
```
if len(url_or_urls) >= 16:
download_func = partial(self._download_batched, download_config=download_config)
else:
download_func = partial(self._download_single, download_config=download_config)
start_time = datetime.now()
with stack_multiprocessing_download_progress_bars():
downloaded_path_or_paths = map_nested(
download_func,
url_or_urls,
map_tuple=True,
num_proc=download_config.num_proc,
desc="Downloading data files",
batched=True if len(url_or_urls) >= 16 else False,
batch_size=-1,
)
```
I would suggest to consider other datasets for similar issues and make a pull-request. | [
-0.10406342148780823,
-0.19442743062973022,
-0.0057610925287008286,
0.28873103857040405,
0.33215194940567017,
-0.16237778961658478,
0.1272556334733963,
0.14151842892169952,
0.27067849040031433,
0.2080133557319641,
-0.25937405228614807,
-0.14443744719028473,
-0.1255161613225937,
0.110425673... |
https://github.com/huggingface/datasets/issues/6850 | Problem loading voxpopuli dataset | Thanks for reporting @Namangarg110 and thanks for the investigation @MilanaShhanukova.
Apparently, there is an issue with the download functionality.
I am proposing a fix. | ### Describe the bug
```
Exception has occurred: FileNotFoundError
Couldn't find file at https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/{'en': 'data/en/asr_train.tsv'}
```
Error in logic for link url creation. The link should be https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/en/asr_train.tsv
Basically there should be links directly under ```metadata["train"]```, not under ```metadata["train"][self.config.languages[0]]```
same for audio urls
### Steps to reproduce the bug
```
from datasets import load_dataset
dataset = load_dataset("facebook/voxpopuli","en")
```
### Expected behavior
Dataset should be loaded successfully.
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-5.15.0-1041-aws-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.12.2 | 24 | Problem loading voxpopuli dataset
### Describe the bug
```
Exception has occurred: FileNotFoundError
Couldn't find file at https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/{'en': 'data/en/asr_train.tsv'}
```
Error in logic for link url creation. The link should be https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/en/asr_train.tsv
Basically there should be links directly under ```metadata["train"]```, not under ```metadata["train"][self.config.languages[0]]```
same for audio urls
### Steps to reproduce the bug
```
from datasets import load_dataset
dataset = load_dataset("facebook/voxpopuli","en")
```
### Expected behavior
Dataset should be loaded successfully.
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-5.15.0-1041-aws-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.12.2
Thanks for reporting @Namangarg110 and thanks for the investigation @MilanaShhanukova.
Apparently, there is an issue with the download functionality.
I am proposing a fix. | [
0.016976218670606613,
-0.12152555584907532,
-0.025779325515031815,
0.3417309522628784,
0.3796597123146057,
-0.11499855667352676,
0.010786697268486023,
0.10583870112895966,
0.31906557083129883,
0.09532859176397324,
-0.19825315475463867,
-0.16082963347434998,
0.05572669580578804,
0.137243136... |
https://github.com/huggingface/datasets/issues/6847 | [Streaming] Only load requested splits without resolving files for the other splits | I'm having a similar issue when using splices:
<img width="947" alt="image" src="https://github.com/huggingface/datasets/assets/28941213/2153faac-e1fe-4b6d-a79b-30b2699407e8">
<img width="823" alt="image" src="https://github.com/huggingface/datasets/assets/28941213/80919eca-eb6c-407d-8070-52642fdcee54">
<img width="914" alt="image" src="https://github.com/huggingface/datasets/assets/28941213/5219c201-e22e-4536-acc3-a922677785ff">
It seems to be downloading, loading, and generating splits using the entire dataset. | e.g. [thangvip](https://huggingface.co/thangvip)/[cosmopedia_vi_math](https://huggingface.co/datasets/thangvip/cosmopedia_vi_math) has 300 splits and it takes a very long time to load only one split.
This is due to `load_dataset()` resolving the files of all the splits even if only one is needed.
In `dataset-viewer` the splits are loaded in different jobs so it results in 300 jobs that resolve 300 splits -> 90k calls to `/paths-info` | 33 | [Streaming] Only load requested splits without resolving files for the other splits
e.g. [thangvip](https://huggingface.co/thangvip)/[cosmopedia_vi_math](https://huggingface.co/datasets/thangvip/cosmopedia_vi_math) has 300 splits and it takes a very long time to load only one split.
This is due to `load_dataset()` resolving the files of all the splits even if only one is needed.
In `dataset-viewer` the splits are loaded in different jobs so it results in 300 jobs that resolve 300 splits -> 90k calls to `/paths-info`
I'm having a similar issue when using splices:
<img width="947" alt="image" src="https://github.com/huggingface/datasets/assets/28941213/2153faac-e1fe-4b6d-a79b-30b2699407e8">
<img width="823" alt="image" src="https://github.com/huggingface/datasets/assets/28941213/80919eca-eb6c-407d-8070-52642fdcee54">
<img width="914" alt="image" src="https://github.com/huggingface/datasets/assets/28941213/5219c201-e22e-4536-acc3-a922677785ff">
It seems to be downloading, loading, and generating splits using the entire dataset. | [
-0.5164943933486938,
-0.2733079195022583,
-0.03313096985220909,
0.3817753791809082,
0.12230648845434189,
-0.1887190043926239,
0.17538385093212128,
0.24712230265140533,
0.11154185235500336,
0.04875592142343521,
-0.08848661184310913,
0.14890806376934052,
0.13150809705257416,
0.31572917103767... |
https://github.com/huggingface/datasets/issues/6846 | Unimaginable super slow iteration | In every iteration you load the full "random_input" column in memory, only then to access it's i-th element.
You can try using this instead
a,b=dataset[i]['random_input'],dataset[i]['random_output'] | ### Describe the bug
Assuming there is a dataset with 52000 sentences, each with a length of 500, it takes 20 seconds to extract a sentence from the dataset……？Is there something wrong with my iteration?
### Steps to reproduce the bug
```python
import datasets
import time
import random
num_rows = 52000
num_cols = 500
random_input = [[random.randint(1, 100) for _ in range(num_cols)] for _ in range(num_rows)]
random_output = [[random.randint(1, 100) for _ in range(num_cols)] for _ in range(num_rows)]
s=time.time()
d={'random_input':random_input,'random_output':random_output}
dataset=datasets.Dataset.from_dict(d)
print('from dict',time.time()-s)
print(dataset)
for i in range(len(dataset)):
aa=time.time()
a,b=dataset['random_input'][i],dataset['random_output'][i]
print(time.time()-aa)
```
corresponding output
```bash
from dict 9.215498685836792
Dataset({
features: ['random_input', 'random_output'],
num_rows: 52000
})
19.129778146743774
19.329464197158813
19.27668261528015
19.28557538986206
19.247620582580566
19.624247074127197
19.28673791885376
19.301053047180176
19.290496110916138
19.291821718215942
19.357765197753906
```
### Expected behavior
Under normal circumstances, iteration should be very rapid as it does not involve the main tasks other than getting items
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-3.10.0-1160.71.1.el7.x86_64-x86_64-with-glibc2.17
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.0
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0 | 25 | Unimaginable super slow iteration
### Describe the bug
Assuming there is a dataset with 52000 sentences, each with a length of 500, it takes 20 seconds to extract a sentence from the dataset……？Is there something wrong with my iteration?
### Steps to reproduce the bug
```python
import datasets
import time
import random
num_rows = 52000
num_cols = 500
random_input = [[random.randint(1, 100) for _ in range(num_cols)] for _ in range(num_rows)]
random_output = [[random.randint(1, 100) for _ in range(num_cols)] for _ in range(num_rows)]
s=time.time()
d={'random_input':random_input,'random_output':random_output}
dataset=datasets.Dataset.from_dict(d)
print('from dict',time.time()-s)
print(dataset)
for i in range(len(dataset)):
aa=time.time()
a,b=dataset['random_input'][i],dataset['random_output'][i]
print(time.time()-aa)
```
corresponding output
```bash
from dict 9.215498685836792
Dataset({
features: ['random_input', 'random_output'],
num_rows: 52000
})
19.129778146743774
19.329464197158813
19.27668261528015
19.28557538986206
19.247620582580566
19.624247074127197
19.28673791885376
19.301053047180176
19.290496110916138
19.291821718215942
19.357765197753906
```
### Expected behavior
Under normal circumstances, iteration should be very rapid as it does not involve the main tasks other than getting items
### Environment info
- `datasets` version: 2.19.0
- Platform: Linux-3.10.0-1160.71.1.el7.x86_64-x86_64-with-glibc2.17
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.0
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
In every iteration you load the full "random_input" column in memory, only then to access it's i-th element.
You can try using this instead
a,b=dataset[i]['random_input'],dataset[i]['random_output'] | [
-0.11157979816198349,
-0.3910645842552185,
-0.09469738602638245,
0.18703684210777283,
0.22112862765789032,
0.007039368152618408,
0.23812644183635712,
-0.043871838599443436,
-0.09718349575996399,
0.21295076608657837,
0.21351777017116547,
0.5390519499778748,
0.001966496929526329,
-0.02626883... |
https://github.com/huggingface/datasets/issues/6845 | load_dataset doesn't support list column | I encountered this same issue when loading a customized dataset for ORPO training, in which there were three columns and two of them were lists.
I debugged and found that it might be caused by the type-infer mechanism and because in some chunks one of the columns is always an empty list ([]), it was regarded as ```list<item: null>```, however in some other chunk it was ```list<item: string>```. This triggered a TypeError running the function ```table_cast()```.
I temporarily fixed this by re-dumping the file into a regular JSON format instead of lines of JSON dict. I didn't dig deeper for the lack of knowledge and programming ability but I do hope some developer of this repo will find and fix it. | ### Describe the bug
dataset = load_dataset("Doraemon-AI/text-to-neo4j-cypher-chinese")
got exception:
Generating train split: 1834 examples [00:00, 5227.98 examples/s]
Traceback (most recent call last):
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 2011, in _prepare_split_single
writer.write_table(table)
File "/usr/local/lib/python3.11/dist-packages/datasets/arrow_writer.py", line 585, in write_table
pa_table = table_cast(pa_table, self._schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2295, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2254, in cast_table_to_schema
arrays = [cast_array_to_feature(table[name], feature) for name, feature in features.items()]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2254, in <listcomp>
arrays = [cast_array_to_feature(table[name], feature) for name, feature in features.items()]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 1802, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 1802, in <listcomp>
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2018, in cast_array_to_feature
casted_array_values = _c(array.values, feature[0])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 1804, in wrapper
return func(array, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2115, in cast_array_to_feature
raise TypeError(f"Couldn't cast array of type\n{array.type}\nto\n{feature}")
TypeError: Couldn't cast array of type
struct<m.name: string, x.name: string, p.name: string, n.name: string, h.name: string, name: string, c: int64, collect(r.name): list<item: string>, q.name: string, rel.name: string, count(p): int64, 1: int64, p.location: string, max(n.name): null, mn.name: string, p.time: int64, min(q.name): string>
to
{'q.name': Value(dtype='string', id=None), 'mn.name': Value(dtype='string', id=None), 'x.name': Value(dtype='string', id=None), 'p.name': Value(dtype='string', id=None), 'n.name': Value(dtype='string', id=None), 'name': Value(dtype='string', id=None), 'm.name': Value(dtype='string', id=None), 'h.name': Value(dtype='string', id=None), 'count(p)': Value(dtype='int64', id=None), 'rel.name': Value(dtype='string', id=None), 'c': Value(dtype='int64', id=None), 'collect(r.name)': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None), '1': Value(dtype='int64', id=None), 'p.location': Value(dtype='string', id=None), 'substring(h.name,0,5)': Value(dtype='string', id=None), 'p.time': Value(dtype='int64', id=None)}
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/ubuntu/llm/train-2.py", line 150, in <module>
dataset = load_dataset("Doraemon-AI/text-to-neo4j-cypher-chinese")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
### Steps to reproduce the bug
dataset = load_dataset("Doraemon-AI/text-to-neo4j-cypher-chinese")
### Expected behavior
no exception
### Environment info
python 3.11
datasets 2.19.0 | 121 | load_dataset doesn't support list column
### Describe the bug
dataset = load_dataset("Doraemon-AI/text-to-neo4j-cypher-chinese")
got exception:
Generating train split: 1834 examples [00:00, 5227.98 examples/s]
Traceback (most recent call last):
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 2011, in _prepare_split_single
writer.write_table(table)
File "/usr/local/lib/python3.11/dist-packages/datasets/arrow_writer.py", line 585, in write_table
pa_table = table_cast(pa_table, self._schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2295, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2254, in cast_table_to_schema
arrays = [cast_array_to_feature(table[name], feature) for name, feature in features.items()]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2254, in <listcomp>
arrays = [cast_array_to_feature(table[name], feature) for name, feature in features.items()]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 1802, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 1802, in <listcomp>
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2018, in cast_array_to_feature
casted_array_values = _c(array.values, feature[0])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 1804, in wrapper
return func(array, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/table.py", line 2115, in cast_array_to_feature
raise TypeError(f"Couldn't cast array of type\n{array.type}\nto\n{feature}")
TypeError: Couldn't cast array of type
struct<m.name: string, x.name: string, p.name: string, n.name: string, h.name: string, name: string, c: int64, collect(r.name): list<item: string>, q.name: string, rel.name: string, count(p): int64, 1: int64, p.location: string, max(n.name): null, mn.name: string, p.time: int64, min(q.name): string>
to
{'q.name': Value(dtype='string', id=None), 'mn.name': Value(dtype='string', id=None), 'x.name': Value(dtype='string', id=None), 'p.name': Value(dtype='string', id=None), 'n.name': Value(dtype='string', id=None), 'name': Value(dtype='string', id=None), 'm.name': Value(dtype='string', id=None), 'h.name': Value(dtype='string', id=None), 'count(p)': Value(dtype='int64', id=None), 'rel.name': Value(dtype='string', id=None), 'c': Value(dtype='int64', id=None), 'collect(r.name)': Sequence(feature=Value(dtype='string', id=None), length=-1, id=None), '1': Value(dtype='int64', id=None), 'p.location': Value(dtype='string', id=None), 'substring(h.name,0,5)': Value(dtype='string', id=None), 'p.time': Value(dtype='int64', id=None)}
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/ubuntu/llm/train-2.py", line 150, in <module>
dataset = load_dataset("Doraemon-AI/text-to-neo4j-cypher-chinese")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/datasets/load.py", line 2609, in load_dataset
builder_instance.download_and_prepare(
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File "/usr/local/lib/python3.11/dist-packages/datasets/builder.py", line 2038, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset
### Steps to reproduce the bug
dataset = load_dataset("Doraemon-AI/text-to-neo4j-cypher-chinese")
### Expected behavior
no exception
### Environment info
python 3.11
datasets 2.19.0
I encountered this same issue when loading a customized dataset for ORPO training, in which there were three columns and two of them were lists.
I debugged and found that it might be caused by the type-infer mechanism and because in some chunks one of the columns is always an empty list ([]), it was regarded as ```list<item: null>```, however in some other chunk it was ```list<item: string>```. This triggered a TypeError running the function ```table_cast()```.
I temporarily fixed this by re-dumping the file into a regular JSON format instead of lines of JSON dict. I didn't dig deeper for the lack of knowledge and programming ability but I do hope some developer of this repo will find and fix it. | [
-0.4311951696872711,
0.1355215311050415,
-0.07370404899120331,
0.33902639150619507,
0.5347718596458435,
0.25930511951446533,
0.6299633979797363,
0.47520458698272705,
0.4229787290096283,
0.07134582847356796,
-0.0033362400718033314,
0.6009827256202698,
-0.17080233991146088,
0.306876003742218... |
https://github.com/huggingface/datasets/issues/6843 | IterableDataset raises exception instead of retrying | Thanks, @mariosasko! Related question (although I guess this is a feature request): could we have some kind of exponential back-off for these retries? Here's my reasoning:
- If a one-time accidental error happens, you should retry immediately and will succeed immediately.
- If the Hub has a small outage on the order of minutes, you don't want to retry on the order of hours.
- If the Hub has a prologned outage of several hours, we don't want to keep retrying on the order of minutes.
There actually already exists an implementation for (clipped) exponential backoff in the HuggingFace suite ([here](https://github.com/huggingface/huggingface_hub/blob/61b156a4f2e5fe1a492ed8712b26803e2122bde0/src/huggingface_hub/utils/_http.py#L306)), but I don't think it is used here.
The requirements are basically that you have an initial minimum waiting time and a maximum waiting time, and with each retry, the waiting time is doubled. We don't want to overload your servers with needless retries, especially when they're down :sweat_smile: | ### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0 | 150 | IterableDataset raises exception instead of retrying
### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0
Thanks, @mariosasko! Related question (although I guess this is a feature request): could we have some kind of exponential back-off for these retries? Here's my reasoning:
- If a one-time accidental error happens, you should retry immediately and will succeed immediately.
- If the Hub has a small outage on the order of minutes, you don't want to retry on the order of hours.
- If the Hub has a prologned outage of several hours, we don't want to keep retrying on the order of minutes.
There actually already exists an implementation for (clipped) exponential backoff in the HuggingFace suite ([here](https://github.com/huggingface/huggingface_hub/blob/61b156a4f2e5fe1a492ed8712b26803e2122bde0/src/huggingface_hub/utils/_http.py#L306)), but I don't think it is used here.
The requirements are basically that you have an initial minimum waiting time and a maximum waiting time, and with each retry, the waiting time is doubled. We don't want to overload your servers with needless retries, especially when they're down :sweat_smile: | [
-0.34127748012542725,
-0.04368980973958969,
-0.03248382732272148,
0.030754581093788147,
0.1366526484489441,
-0.10219722241163254,
0.1777607500553131,
-0.1343664675951004,
-0.15274980664253235,
-0.11436350643634796,
-0.037531763315200806,
-0.08118641376495361,
-0.1617305725812912,
0.1365094... |
https://github.com/huggingface/datasets/issues/6843 | IterableDataset raises exception instead of retrying | Oh, I've just remembered that we added retries to the `HfFileSystem` in `huggingface_hub` 0.21.0 (see [this](https://github.com/huggingface/huggingface_hub/blob/61b156a4f2e5fe1a492ed8712b26803e2122bde0/src/huggingface_hub/hf_file_system.py#L703)), so I'll close the linked PR as we don't want to retry the retries :).
I agree with the exponential backoff suggestion, so I'll open another PR. | ### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0 | 43 | IterableDataset raises exception instead of retrying
### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0
Oh, I've just remembered that we added retries to the `HfFileSystem` in `huggingface_hub` 0.21.0 (see [this](https://github.com/huggingface/huggingface_hub/blob/61b156a4f2e5fe1a492ed8712b26803e2122bde0/src/huggingface_hub/hf_file_system.py#L703)), so I'll close the linked PR as we don't want to retry the retries :).
I agree with the exponential backoff suggestion, so I'll open another PR. | [
-0.34127748012542725,
-0.04368980973958969,
-0.03248382732272148,
0.030754581093788147,
0.1366526484489441,
-0.10219722241163254,
0.1777607500553131,
-0.1343664675951004,
-0.15274980664253235,
-0.11436350643634796,
-0.037531763315200806,
-0.08118641376495361,
-0.1617305725812912,
0.1365094... |
https://github.com/huggingface/datasets/issues/6843 | IterableDataset raises exception instead of retrying | @mariosasko The call you linked indeed points to the implementation I linked in my previous comment, yes, but it has no configurability. Arguably, you want to have this hidden backoff under the hood that catches small network disturbances on the time scale of seconds -- perhaps even with hardcoded limits as is the case currently -- but you also still want to have a separate backoff on top of that with the configurability as suggested by @lhoestq in [the comment I linked](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229).
My particular use-case is that I'm streaming a dataset while training on a university cluster with a very long scheduling queue. This means that when the backoff runs out of retries (which happens in under 30 seconds with the call you linked), I lose my spot on the cluster and have to queue for a whole day or more. Ideally, I should be able to specify that I want to retry for 2 to 3 hours but with more and more time between requests, so that I can smooth over hours-long outages without a setback of days. | ### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0 | 179 | IterableDataset raises exception instead of retrying
### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0
@mariosasko The call you linked indeed points to the implementation I linked in my previous comment, yes, but it has no configurability. Arguably, you want to have this hidden backoff under the hood that catches small network disturbances on the time scale of seconds -- perhaps even with hardcoded limits as is the case currently -- but you also still want to have a separate backoff on top of that with the configurability as suggested by @lhoestq in [the comment I linked](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229).
My particular use-case is that I'm streaming a dataset while training on a university cluster with a very long scheduling queue. This means that when the backoff runs out of retries (which happens in under 30 seconds with the call you linked), I lose my spot on the cluster and have to queue for a whole day or more. Ideally, I should be able to specify that I want to retry for 2 to 3 hours but with more and more time between requests, so that I can smooth over hours-long outages without a setback of days. | [
-0.34127748012542725,
-0.04368980973958969,
-0.03248382732272148,
0.030754581093788147,
0.1366526484489441,
-0.10219722241163254,
0.1777607500553131,
-0.1343664675951004,
-0.15274980664253235,
-0.11436350643634796,
-0.037531763315200806,
-0.08118641376495361,
-0.1617305725812912,
0.1365094... |
https://github.com/huggingface/datasets/issues/6843 | IterableDataset raises exception instead of retrying | I also have my runs crash a surprising amount due to the dataloader crashing because of the hub, some way to address this would be nice. | ### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0 | 26 | IterableDataset raises exception instead of retrying
### Describe the bug
In light of the recent server outages, I decided to look into whether I could somehow wrap my IterableDataset streams to retry rather than error out immediately. To my surprise, `datasets` [already supports retries](https://github.com/huggingface/datasets/issues/6172#issuecomment-1794876229). Since a commit by @lhoestq [last week](https://github.com/huggingface/datasets/commit/a188022dc43a76a119d90c03832d51d6e4a94d91), that code lives here:
https://github.com/huggingface/datasets/blob/fe2bea6a4b09b180bd23b88fe96dfd1a11191a4f/src/datasets/utils/file_utils.py#L1097C1-L1111C19
If GitHub code snippets still aren't working, here's a copy:
```python
def read_with_retries(*args, **kwargs):
disconnect_err = None
for retry in range(1, max_retries + 1):
try:
out = read(*args, **kwargs)
break
except (ClientError, TimeoutError) as err:
disconnect_err = err
logger.warning(
f"Got disconnected from remote data host. Retrying in {config.STREAMING_READ_RETRY_INTERVAL}sec [{retry}/{max_retries}]"
)
time.sleep(config.STREAMING_READ_RETRY_INTERVAL)
else:
raise ConnectionError("Server Disconnected") from disconnect_err
return out
```
With the latest outage, the end of my stack trace looked like this:
```
...
File "/miniconda3/envs/draft/lib/python3.11/site-packages/datasets/download/streaming_download_manager.py", line 342, in read_with_retries
out = read(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 505, in read
buf = self._fp.read(io.DEFAULT_BUFFER_SIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/gzip.py", line 88, in read
return self.file.read(size)
^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/spec.py", line 1856, in read
out = self.cache._fetch(self.loc, self.loc + length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/fsspec/caching.py", line 189, in _fetch
self.cache = self.fetcher(start, end) # new block replaces old
^^^^^^^^^^^^^^^^^^^^^^^^
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/hf_file_system.py", line 626, in _fetch_range
hf_raise_for_status(r)
File "/miniconda3/envs/draft/lib/python3.11/site-packages/huggingface_hub/utils/_errors.py", line 333, in hf_raise_for_status
raise HfHubHTTPError(str(e), response=response) from e
huggingface_hub.utils._errors.HfHubHTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/datasets/allenai/c4/resolve/1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00346-of-01024.json.gz
```
Indeed, the code for retries only catches `ClientError`s and `TimeoutError`s, and all other exceptions, *including HuggingFace's own custom HTTP error class*, **are not caught. Nothing is retried,** and instead the exception is propagated upwards immediately.
### Steps to reproduce the bug
Not sure how you reproduce this. Maybe unplug your Ethernet cable while streaming a dataset; the issue is pretty clear from the stack trace.
### Expected behavior
All HTTP errors while iterating a streamable dataset should cause retries.
### Environment info
Output from `datasets-cli env`:
- `datasets` version: 2.18.0
- Platform: Linux-4.18.0-513.24.1.el8_9.x86_64-x86_64-with-glibc2.28
- Python version: 3.11.7
- `huggingface_hub` version: 0.20.3
- PyArrow version: 15.0.0
- Pandas version: 2.2.0
- `fsspec` version: 2023.10.0
I also have my runs crash a surprising amount due to the dataloader crashing because of the hub, some way to address this would be nice. | [
-0.34127748012542725,
-0.04368980973958969,
-0.03248382732272148,
0.030754581093788147,
0.1366526484489441,
-0.10219722241163254,
0.1777607500553131,
-0.1343664675951004,
-0.15274980664253235,
-0.11436350643634796,
-0.037531763315200806,
-0.08118641376495361,
-0.1617305725812912,
0.1365094... |
https://github.com/huggingface/datasets/issues/6841 | Unable to load wiki_auto_asset_turk from GEM | Hi! I've opened a [PR](https://huggingface.co/datasets/GEM/wiki_auto_asset_turk/discussions/5) with a fix. While waiting for it to be merged, you can load the dataset from the PR branch with `datasets.load_dataset("GEM/wiki_auto_asset_turk", revision="refs/pr/5")` | ### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0 | 27 | Unable to load wiki_auto_asset_turk from GEM
### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0
Hi! I've opened a [PR](https://huggingface.co/datasets/GEM/wiki_auto_asset_turk/discussions/5) with a fix. While waiting for it to be merged, you can load the dataset from the PR branch with `datasets.load_dataset("GEM/wiki_auto_asset_turk", revision="refs/pr/5")` | [
-0.1644713282585144,
0.025188952684402466,
0.050996411591768265,
0.6145248413085938,
0.43248093128204346,
0.11137401312589645,
0.35678306221961975,
0.29374057054519653,
0.1904025375843048,
0.0366792157292366,
0.23593732714653015,
0.47361600399017334,
-0.23568369448184967,
-0.10596699267625... |
https://github.com/huggingface/datasets/issues/6841 | Unable to load wiki_auto_asset_turk from GEM | Thanks Mario. Still getting the same issue though with the suggested fix
#cat gem_sari.py
import datasets
print (datasets.__version__)
dataset =datasets.load_dataset("GEM/wiki_auto_asset_turk", revision="refs/pr/5")
End up with
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType | ### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0 | 109 | Unable to load wiki_auto_asset_turk from GEM
### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0
Thanks Mario. Still getting the same issue though with the suggested fix
#cat gem_sari.py
import datasets
print (datasets.__version__)
dataset =datasets.load_dataset("GEM/wiki_auto_asset_turk", revision="refs/pr/5")
End up with
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType | [
-0.1644713282585144,
0.025188952684402466,
0.050996411591768265,
0.6145248413085938,
0.43248093128204346,
0.11137401312589645,
0.35678306221961975,
0.29374057054519653,
0.1904025375843048,
0.0366792157292366,
0.23593732714653015,
0.47361600399017334,
-0.23568369448184967,
-0.10596699267625... |
https://github.com/huggingface/datasets/issues/6841 | Unable to load wiki_auto_asset_turk from GEM | Tried that a couple of time. It does download the data fresh but end up with same error. Is there a way to see if its using the right version ? | ### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0 | 31 | Unable to load wiki_auto_asset_turk from GEM
### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0
Tried that a couple of time. It does download the data fresh but end up with same error. Is there a way to see if its using the right version ? | [
-0.1644713282585144,
0.025188952684402466,
0.050996411591768265,
0.6145248413085938,
0.43248093128204346,
0.11137401312589645,
0.35678306221961975,
0.29374057054519653,
0.1904025375843048,
0.0366792157292366,
0.23593732714653015,
0.47361600399017334,
-0.23568369448184967,
-0.10596699267625... |
https://github.com/huggingface/datasets/issues/6841 | Unable to load wiki_auto_asset_turk from GEM | the datasets version is 2.18.
I wanted to see if the command datasets.load_dataset("GEM/wiki_auto_asset_turk", revision="refs/pr/5") is using the right revision (refs/pr/5).
| ### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0 | 20 | Unable to load wiki_auto_asset_turk from GEM
### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0
the datasets version is 2.18.
I wanted to see if the command datasets.load_dataset("GEM/wiki_auto_asset_turk", revision="refs/pr/5") is using the right revision (refs/pr/5).
| [
-0.1644713282585144,
0.025188952684402466,
0.050996411591768265,
0.6145248413085938,
0.43248093128204346,
0.11137401312589645,
0.35678306221961975,
0.29374057054519653,
0.1904025375843048,
0.0366792157292366,
0.23593732714653015,
0.47361600399017334,
-0.23568369448184967,
-0.10596699267625... |
https://github.com/huggingface/datasets/issues/6841 | Unable to load wiki_auto_asset_turk from GEM | The issue is fixed once the fixing PR has been merged and the dataset has been converted to Parquet.
If the problem persists on your side, you should update your `datasets` library:
```shell
pip install -U datasets
```
And if you have already the latest version of `datasets`, then you need to delete the old version of this dataset in your cache:
```shell
rm -fr ~/.cache/huggingface/datasets/GEM___wiki_auto_asset_turk
rm -fr ~/.cache/huggingface/modules/datasets_modules/datasets/GEM--wiki_auto_asset_turk
``` | ### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0 | 70 | Unable to load wiki_auto_asset_turk from GEM
### Describe the bug
I am unable to load the wiki_auto_asset_turk dataset. I get a fatal error while trying to access wiki_auto_asset_turk and load it with datasets.load_dataset. The error (TypeError: expected str, bytes or os.PathLike object, not NoneType) is from filenames_for_dataset_split in a os.path.join call
>>import datasets
>>print (datasets.__version__)
>>dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
System output:
Generating train split: 100%|█| 483801/483801 [00:03<00:00, 127164.26 examples/s
Generating validation split: 100%|█| 20000/20000 [00:00<00:00, 116052.94 example
Generating test_asset split: 100%|██| 359/359 [00:00<00:00, 76155.93 examples/s]
Generating test_turk split: 100%|███| 359/359 [00:00<00:00, 87691.76 examples/s]
Traceback (most recent call last):
File "/Users/abhinav.sethy/Code/openai_evals/evals/evals/grammarly_tasks/gem_sari.py", line 3, in <module>
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/load.py", line 2582, in load_dataset
builder_instance.download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1005, in download_and_prepare
self._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1767, in _download_and_prepare
super()._download_and_prepare(
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1100, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/builder.py", line 1565, in _prepare_split
split_info = self.info.splits[split_generator.name]
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/splits.py", line 532, in __getitem__
instructions = make_file_instructions(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/arrow_reader.py", line 121, in make_file_instructions
info.name: filenames_for_dataset_split(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/datasets/naming.py", line 72, in filenames_for_dataset_split
prefix = os.path.join(path, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen posixpath>", line 76, in join
TypeError: expected str, bytes or os.PathLike object, not NoneType
### Steps to reproduce the bug
import datasets
print (datasets.__version__)
dataset = datasets.load_dataset("GEM/wiki_auto_asset_turk")
### Expected behavior
Should be able to load the dataset without any issues
### Environment info
datasets version 2.18.0 (was able to reproduce bug with older versions 2.16 and 2.14 also)
Python 3.12.0
The issue is fixed once the fixing PR has been merged and the dataset has been converted to Parquet.
If the problem persists on your side, you should update your `datasets` library:
```shell
pip install -U datasets
```
And if you have already the latest version of `datasets`, then you need to delete the old version of this dataset in your cache:
```shell
rm -fr ~/.cache/huggingface/datasets/GEM___wiki_auto_asset_turk
rm -fr ~/.cache/huggingface/modules/datasets_modules/datasets/GEM--wiki_auto_asset_turk
``` | [
-0.1644713282585144,
0.025188952684402466,
0.050996411591768265,
0.6145248413085938,
0.43248093128204346,
0.11137401312589645,
0.35678306221961975,
0.29374057054519653,
0.1904025375843048,
0.0366792157292366,
0.23593732714653015,
0.47361600399017334,
-0.23568369448184967,
-0.10596699267625... |
https://github.com/huggingface/datasets/issues/6837 | Cannot use cached dataset without Internet connection (or when servers are down) | There are 2 workarounds, tho:
1. Download datasets from web and just load them locally
2. Use metadata directly (temporal solution, since metadata can change)
```
import datasets
from datasets.data_files import DataFilesDict, DataFilesList
data_files_list = DataFilesList(
[
"hf://datasets/allenai/c4@1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00000-of-01024.json.gz"
],
[("allenai/c4", "1588ec454efa1a09f29cd18ddd04fe05fc8653a2")],
)
data_files = DataFilesDict({"train": data_files_list})
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files=data_files,
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
Second solution also shows where to find the bug. I suggest that the hashing functions should always use only original parameter `data_files`, and not the one they get after connecting to the server and creating `DataFilesDict` | ### Describe the bug
I want to be able to use cached dataset from HuggingFace even when I have no Internet connection (or when HuggingFace servers are down, or my company has network issues).
The problem why I can't use it:
`data_files` argument from `datasets.load_dataset()` function get it updates from the server before calculating hash for caching. As a result, when I run the same code with and without Internet I get different dataset configuration directory name.
### Steps to reproduce the bug
```
import datasets
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files={"train": "en/c4-train.00000-of-01024.json.gz"},
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
1. Run this code with the Internet.
2. Run the same code without the Internet.
### Expected behavior
When running without the Internet connection, the loader should be able to get dataset from cache
### Environment info
- `datasets` version: 2.19.0
- Platform: Windows-10-10.0.19044-SP0
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2 | 93 | Cannot use cached dataset without Internet connection (or when servers are down)
### Describe the bug
I want to be able to use cached dataset from HuggingFace even when I have no Internet connection (or when HuggingFace servers are down, or my company has network issues).
The problem why I can't use it:
`data_files` argument from `datasets.load_dataset()` function get it updates from the server before calculating hash for caching. As a result, when I run the same code with and without Internet I get different dataset configuration directory name.
### Steps to reproduce the bug
```
import datasets
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files={"train": "en/c4-train.00000-of-01024.json.gz"},
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
1. Run this code with the Internet.
2. Run the same code without the Internet.
### Expected behavior
When running without the Internet connection, the loader should be able to get dataset from cache
### Environment info
- `datasets` version: 2.19.0
- Platform: Windows-10-10.0.19044-SP0
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2
There are 2 workarounds, tho:
1. Download datasets from web and just load them locally
2. Use metadata directly (temporal solution, since metadata can change)
```
import datasets
from datasets.data_files import DataFilesDict, DataFilesList
data_files_list = DataFilesList(
[
"hf://datasets/allenai/c4@1588ec454efa1a09f29cd18ddd04fe05fc8653a2/en/c4-train.00000-of-01024.json.gz"
],
[("allenai/c4", "1588ec454efa1a09f29cd18ddd04fe05fc8653a2")],
)
data_files = DataFilesDict({"train": data_files_list})
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files=data_files,
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
Second solution also shows where to find the bug. I suggest that the hashing functions should always use only original parameter `data_files`, and not the one they get after connecting to the server and creating `DataFilesDict` | [
-0.17088213562965393,
0.09447182714939117,
0.08116744458675385,
0.41006356477737427,
0.3210458755493164,
0.06577769666910172,
0.37767836451530457,
0.0363660492002964,
0.2188761830329895,
0.20855626463890076,
0.09173806756734848,
0.13869549334049225,
-0.03013565018773079,
-0.098603710532188... |
https://github.com/huggingface/datasets/issues/6837 | Cannot use cached dataset without Internet connection (or when servers are down) | Hi! You need to set the `HF_DATASETS_OFFLINE` env variable to `1` to load cached datasets offline, as explained in the docs [here](https://huggingface.co/docs/datasets/v2.19.0/en/loading#offline). | ### Describe the bug
I want to be able to use cached dataset from HuggingFace even when I have no Internet connection (or when HuggingFace servers are down, or my company has network issues).
The problem why I can't use it:
`data_files` argument from `datasets.load_dataset()` function get it updates from the server before calculating hash for caching. As a result, when I run the same code with and without Internet I get different dataset configuration directory name.
### Steps to reproduce the bug
```
import datasets
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files={"train": "en/c4-train.00000-of-01024.json.gz"},
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
1. Run this code with the Internet.
2. Run the same code without the Internet.
### Expected behavior
When running without the Internet connection, the loader should be able to get dataset from cache
### Environment info
- `datasets` version: 2.19.0
- Platform: Windows-10-10.0.19044-SP0
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2 | 22 | Cannot use cached dataset without Internet connection (or when servers are down)
### Describe the bug
I want to be able to use cached dataset from HuggingFace even when I have no Internet connection (or when HuggingFace servers are down, or my company has network issues).
The problem why I can't use it:
`data_files` argument from `datasets.load_dataset()` function get it updates from the server before calculating hash for caching. As a result, when I run the same code with and without Internet I get different dataset configuration directory name.
### Steps to reproduce the bug
```
import datasets
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files={"train": "en/c4-train.00000-of-01024.json.gz"},
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
1. Run this code with the Internet.
2. Run the same code without the Internet.
### Expected behavior
When running without the Internet connection, the loader should be able to get dataset from cache
### Environment info
- `datasets` version: 2.19.0
- Platform: Windows-10-10.0.19044-SP0
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2
Hi! You need to set the `HF_DATASETS_OFFLINE` env variable to `1` to load cached datasets offline, as explained in the docs [here](https://huggingface.co/docs/datasets/v2.19.0/en/loading#offline). | [
-0.16282786428928375,
-0.0011418312788009644,
0.0749388188123703,
0.3900284171104431,
0.30762583017349243,
0.08525867015123367,
0.37762191891670227,
-0.0045507438480854034,
0.27046650648117065,
0.19738924503326416,
0.08295249938964844,
0.04613589495420456,
-0.0115044629201293,
-0.049671046... |
https://github.com/huggingface/datasets/issues/6837 | Cannot use cached dataset without Internet connection (or when servers are down) | Just tested. It doesn't work, because of the exact problem I described above: hash of dataset config is different.
The only error difference is the reason why it cannot connect to HuggingFace (now it's 'offline mode is enabled')
| ### Describe the bug
I want to be able to use cached dataset from HuggingFace even when I have no Internet connection (or when HuggingFace servers are down, or my company has network issues).
The problem why I can't use it:
`data_files` argument from `datasets.load_dataset()` function get it updates from the server before calculating hash for caching. As a result, when I run the same code with and without Internet I get different dataset configuration directory name.
### Steps to reproduce the bug
```
import datasets
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files={"train": "en/c4-train.00000-of-01024.json.gz"},
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
1. Run this code with the Internet.
2. Run the same code without the Internet.
### Expected behavior
When running without the Internet connection, the loader should be able to get dataset from cache
### Environment info
- `datasets` version: 2.19.0
- Platform: Windows-10-10.0.19044-SP0
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2 | 39 | Cannot use cached dataset without Internet connection (or when servers are down)
### Describe the bug
I want to be able to use cached dataset from HuggingFace even when I have no Internet connection (or when HuggingFace servers are down, or my company has network issues).
The problem why I can't use it:
`data_files` argument from `datasets.load_dataset()` function get it updates from the server before calculating hash for caching. As a result, when I run the same code with and without Internet I get different dataset configuration directory name.
### Steps to reproduce the bug
```
import datasets
c4_dataset = datasets.load_dataset(
path="allenai/c4",
data_files={"train": "en/c4-train.00000-of-01024.json.gz"},
split="train",
cache_dir="/datesets/cache",
download_mode="reuse_cache_if_exists",
token=False,
)
```
1. Run this code with the Internet.
2. Run the same code without the Internet.
### Expected behavior
When running without the Internet connection, the loader should be able to get dataset from cache
### Environment info
- `datasets` version: 2.19.0
- Platform: Windows-10-10.0.19044-SP0
- Python version: 3.10.13
- `huggingface_hub` version: 0.22.2
- PyArrow version: 16.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2
Just tested. It doesn't work, because of the exact problem I described above: hash of dataset config is different.
The only error difference is the reason why it cannot connect to HuggingFace (now it's 'offline mode is enabled')
| [
-0.1979820877313614,
0.004415258765220642,
0.04450315609574318,
0.4112086296081543,
0.3337099552154541,
0.06295847147703171,
0.3772262632846832,
-0.014828983694314957,
0.238040030002594,
0.1506214141845703,
0.08725883811712265,
0.08241887390613556,
-0.020424801856279373,
-0.023453861474990... |
https://github.com/huggingface/datasets/issues/6833 | Super slow iteration with trivial custom transform | Similar issue in text process
```python
tokenizer=AutoTokenizer.from_pretrained(model_dir[args.model])
train_dataset=datasets.load_from_disk(dataset_dir[args.dataset],keep_in_memory=True)['train']
train_dataset=train_dataset.map(partial(dname2func[args.dataset],tokenizer=tokenizer),batched=True,num_proc =50,remove_columns=train_dataset.features.keys(),desc='tokenize',keep_in_memory=True)
```
After this train_dataset will be like
```python
Dataset({
features: ['input_ids', 'labels'],
num_rows: 51760
})
```
In which input_ids and labels are both List[int]
However, per iter on dataset cost 7.412479639053345s ……？
```python
for j in tqdm(range(len(train_dataset)),desc='first stage'):
input_id,label=train_dataset['input_ids'][j],train_dataset['labels'][j]
``` | ### Describe the bug
Dataset is 10X slower when applying trivial transforms:
```
import time
import numpy as np
from datasets import Dataset, Features, Array2D
a = np.zeros((800, 800))
a = np.stack([a] * 1000)
features = Features({"a": Array2D(shape=(800, 800), dtype="uint8")})
ds1 = Dataset.from_dict({"a": a}, features=features).with_format('numpy')
def transform(batch):
return batch
ds2 = ds1.with_transform(transform)
%time sum(1 for _ in ds1)
%time sum(1 for _ in ds2)
```
```
CPU times: user 472 ms, sys: 319 ms, total: 791 ms
Wall time: 794 ms
CPU times: user 9.32 s, sys: 443 ms, total: 9.76 s
Wall time: 9.78 s
```
In my real code I'm using set_transform to apply some post-processing on-the-fly for the 2d array, but it significantly slows down the dataset even if the transform itself is trivial.
Related issue: https://github.com/huggingface/datasets/issues/5841
### Steps to reproduce the bug
Use code in the description to reproduce.
### Expected behavior
Trivial custom transform in the example should not slowdown the dataset iteration.
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.15.0-79-generic-x86_64-with-glibc2.35
- Python version: 3.11.4
- `huggingface_hub` version: 0.20.2
- PyArrow version: 15.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2 | 50 | Super slow iteration with trivial custom transform
### Describe the bug
Dataset is 10X slower when applying trivial transforms:
```
import time
import numpy as np
from datasets import Dataset, Features, Array2D
a = np.zeros((800, 800))
a = np.stack([a] * 1000)
features = Features({"a": Array2D(shape=(800, 800), dtype="uint8")})
ds1 = Dataset.from_dict({"a": a}, features=features).with_format('numpy')
def transform(batch):
return batch
ds2 = ds1.with_transform(transform)
%time sum(1 for _ in ds1)
%time sum(1 for _ in ds2)
```
```
CPU times: user 472 ms, sys: 319 ms, total: 791 ms
Wall time: 794 ms
CPU times: user 9.32 s, sys: 443 ms, total: 9.76 s
Wall time: 9.78 s
```
In my real code I'm using set_transform to apply some post-processing on-the-fly for the 2d array, but it significantly slows down the dataset even if the transform itself is trivial.
Related issue: https://github.com/huggingface/datasets/issues/5841
### Steps to reproduce the bug
Use code in the description to reproduce.
### Expected behavior
Trivial custom transform in the example should not slowdown the dataset iteration.
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.15.0-79-generic-x86_64-with-glibc2.35
- Python version: 3.11.4
- `huggingface_hub` version: 0.20.2
- PyArrow version: 15.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2
Similar issue in text process
```python
tokenizer=AutoTokenizer.from_pretrained(model_dir[args.model])
train_dataset=datasets.load_from_disk(dataset_dir[args.dataset],keep_in_memory=True)['train']
train_dataset=train_dataset.map(partial(dname2func[args.dataset],tokenizer=tokenizer),batched=True,num_proc =50,remove_columns=train_dataset.features.keys(),desc='tokenize',keep_in_memory=True)
```
After this train_dataset will be like
```python
Dataset({
features: ['input_ids', 'labels'],
num_rows: 51760
})
```
In which input_ids and labels are both List[int]
However, per iter on dataset cost 7.412479639053345s ……？
```python
for j in tqdm(range(len(train_dataset)),desc='first stage'):
input_id,label=train_dataset['input_ids'][j],train_dataset['labels'][j]
``` | [
-0.3822912871837616,
-0.16365310549736023,
-0.00981970690190792,
-0.05723545700311661,
0.33164602518081665,
0.010098747909069061,
0.5508282780647278,
0.15641410648822784,
0.10445114970207214,
0.037332676351070404,
-0.09431147575378418,
0.6537461280822754,
-0.2805955410003662,
0.12174079567... |
https://github.com/huggingface/datasets/issues/6833 | Super slow iteration with trivial custom transform | The transform currently replaces the numpy formatting.
So you're back to copying data to long python lists which is super slow.
It would be cool for the transform to not remove the formatting in this case, but this requires a few changes in the lib | ### Describe the bug
Dataset is 10X slower when applying trivial transforms:
```
import time
import numpy as np
from datasets import Dataset, Features, Array2D
a = np.zeros((800, 800))
a = np.stack([a] * 1000)
features = Features({"a": Array2D(shape=(800, 800), dtype="uint8")})
ds1 = Dataset.from_dict({"a": a}, features=features).with_format('numpy')
def transform(batch):
return batch
ds2 = ds1.with_transform(transform)
%time sum(1 for _ in ds1)
%time sum(1 for _ in ds2)
```
```
CPU times: user 472 ms, sys: 319 ms, total: 791 ms
Wall time: 794 ms
CPU times: user 9.32 s, sys: 443 ms, total: 9.76 s
Wall time: 9.78 s
```
In my real code I'm using set_transform to apply some post-processing on-the-fly for the 2d array, but it significantly slows down the dataset even if the transform itself is trivial.
Related issue: https://github.com/huggingface/datasets/issues/5841
### Steps to reproduce the bug
Use code in the description to reproduce.
### Expected behavior
Trivial custom transform in the example should not slowdown the dataset iteration.
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.15.0-79-generic-x86_64-with-glibc2.35
- Python version: 3.11.4
- `huggingface_hub` version: 0.20.2
- PyArrow version: 15.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2 | 45 | Super slow iteration with trivial custom transform
### Describe the bug
Dataset is 10X slower when applying trivial transforms:
```
import time
import numpy as np
from datasets import Dataset, Features, Array2D
a = np.zeros((800, 800))
a = np.stack([a] * 1000)
features = Features({"a": Array2D(shape=(800, 800), dtype="uint8")})
ds1 = Dataset.from_dict({"a": a}, features=features).with_format('numpy')
def transform(batch):
return batch
ds2 = ds1.with_transform(transform)
%time sum(1 for _ in ds1)
%time sum(1 for _ in ds2)
```
```
CPU times: user 472 ms, sys: 319 ms, total: 791 ms
Wall time: 794 ms
CPU times: user 9.32 s, sys: 443 ms, total: 9.76 s
Wall time: 9.78 s
```
In my real code I'm using set_transform to apply some post-processing on-the-fly for the 2d array, but it significantly slows down the dataset even if the transform itself is trivial.
Related issue: https://github.com/huggingface/datasets/issues/5841
### Steps to reproduce the bug
Use code in the description to reproduce.
### Expected behavior
Trivial custom transform in the example should not slowdown the dataset iteration.
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.15.0-79-generic-x86_64-with-glibc2.35
- Python version: 3.11.4
- `huggingface_hub` version: 0.20.2
- PyArrow version: 15.0.0
- Pandas version: 1.5.3
- `fsspec` version: 2023.12.2
The transform currently replaces the numpy formatting.
So you're back to copying data to long python lists which is super slow.
It would be cool for the transform to not remove the formatting in this case, but this requires a few changes in the lib | [
-0.3285689949989319,
-0.135402649641037,
-0.03535167872905731,
-0.082546666264534,
0.3204980492591858,
0.033024854958057404,
0.46783801913261414,
0.22104917466640472,
0.15049003064632416,
-0.0012368634343147278,
-0.11811057478189468,
0.5683466196060181,
-0.2687753140926361,
0.1582103669643... |
https://github.com/huggingface/datasets/issues/6824 | Winogrande does not seem to be compatible with datasets version of 1.18.0 | Hi ! Do you mean 2.18 ? Can you try to update `fsspec` and `huggingface_hub` ?
```
pip install -U fsspec huggingface_hub
``` | ### Describe the bug
I get the following error when simply running `load_dataset('winogrande','winogrande_xl')`.
I do not have such an issue in the 1.17.0 version.
```Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/local/lib/python3.10/dist-packages/datasets/load.py", line 2556, in load_dataset
builder_instance = load_dataset_builder(
File "/usr/local/lib/python3.10/dist-packages/datasets/load.py", line 2265, in load_dataset_builder
builder_instance: DatasetBuilder = builder_cls(
File "/usr/local/lib/python3.10/dist-packages/datasets/builder.py", line 371, in __init__
self.config, self.config_id = self._create_builder_config(
File "/usr/local/lib/python3.10/dist-packages/datasets/builder.py", line 620, in _create_builder_config
builder_config._resolve_data_files(
File "/usr/local/lib/python3.10/dist-packages/datasets/builder.py", line 211, in _resolve_data_files
self.data_files = self.data_files.resolve(base_path, download_config)
File "/usr/local/lib/python3.10/dist-packages/datasets/data_files.py", line 799, in resolve
out[key] = data_files_patterns_list.resolve(base_path, download_config)
File "/usr/local/lib/python3.10/dist-packages/datasets/data_files.py", line 752, in resolve
resolve_pattern(
File "/usr/local/lib/python3.10/dist-packages/datasets/data_files.py", line 393, in resolve_pattern
raise FileNotFoundError(error_msg)
FileNotFoundError: Unable to find 'hf://datasets/winogrande@ebf71e3c7b5880d019ecf6099c0b09311b1084f5/winogrande_xl/train/0000.parquet' with any supported extension ['.csv', '.tsv', '.json', '.jsonl', '.parquet', '.geoparquet', '.gpq', '.arrow', '.txt', '.tar', '.blp', '.bmp', '.dib', '.bufr', '.cur', '.pcx', '.dcx', '.dds', '.ps', '.eps', '.fit', '.fits', '.fli', '.flc', '.ftc', '.ftu', '.gbr', '.gif', '.grib', '.h5', '.hdf', '.png', '.apng', '.jp2', '.j2k', '.jpc', '.jpf', '.jpx', '.j2c', '.icns', '.ico', '.im', '.iim', '.tif', '.tiff', '.jfif', '.jpe', '.jpg', '.jpeg', '.mpg', '.mpeg', '.msp', '.pcd', '.pxr', '.pbm', '.pgm', '.ppm', '.pnm', '.psd', '.bw', '.rgb', '.rgba', '.sgi', '.ras', '.tga', '.icb', '.vda', '.vst', '.webp', '.wmf', '.emf', '.xbm', '.xpm', '.BLP', '.BMP', '.DIB', '.BUFR', '.CUR', '.PCX', '.DCX', '.DDS', '.PS', '.EPS', '.FIT', '.FITS', '.FLI', '.FLC', '.FTC', '.FTU', '.GBR', '.GIF', '.GRIB', '.H5', '.HDF', '.PNG', '.APNG', '.JP2', '.J2K', '.JPC', '.JPF', '.JPX', '.J2C', '.ICNS', '.ICO', '.IM', '.IIM', '.TIF', '.TIFF', '.JFIF', '.JPE', '.JPG', '.JPEG', '.MPG', '.MPEG', '.MSP', '.PCD', '.PXR', '.PBM', '.PGM', '.PPM', '.PNM', '.PSD', '.BW', '.RGB', '.RGBA', '.SGI', '.RAS', '.TGA', '.ICB', '.VDA', '.VST', '.WEBP', '.WMF', '.EMF', '.XBM', '.XPM', '.aiff', '.au', '.avr', '.caf', '.flac', '.htk', '.svx', '.mat4', '.mat5', '.mpc2k', '.ogg', '.paf', '.pvf', '.raw', '.rf64', '.sd2', '.sds', '.ircam', '.voc', '.w64', '.wav', '.nist', '.wavex', '.wve', '.xi', '.mp3', '.opus', '.AIFF', '.AU', '.AVR', '.CAF', '.FLAC', '.HTK', '.SVX', '.MAT4', '.MAT5', '.MPC2K', '.OGG', '.PAF', '.PVF', '.RAW', '.RF64', '.SD2', '.SDS', '.IRCAM', '.VOC', '.W64', '.WAV', '.NIST', '.WAVEX', '.WVE', '.XI', '.MP3', '.OPUS', '.zip']```
### Steps to reproduce the bug
from datasets import load_dataset
datasets = load_dataset('winogrande','winogrande_xl')
### Expected behavior
```Downloading data: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.06M/2.06M [00:00<00:00, 5.16MB/s]
Downloading data: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 118k/118k [00:00<00:00, 360kB/s]
Downloading data: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 85.9k/85.9k [00:00<00:00, 242kB/s]
Generating train split: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████| 40398/40398 [00:00<00:00, 845491.12 examples/s]
Generating test split: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████| 1767/1767 [00:00<00:00, 362501.11 examples/s]
Generating validation split: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 1267/1267 [00:00<00:00, 318768.11 examples/s]```
### Environment info
datasets version: 1.18.0
| 23 | Winogrande does not seem to be compatible with datasets version of 1.18.0
### Describe the bug
I get the following error when simply running `load_dataset('winogrande','winogrande_xl')`.
I do not have such an issue in the 1.17.0 version.
```Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/local/lib/python3.10/dist-packages/datasets/load.py", line 2556, in load_dataset
builder_instance = load_dataset_builder(
File "/usr/local/lib/python3.10/dist-packages/datasets/load.py", line 2265, in load_dataset_builder
builder_instance: DatasetBuilder = builder_cls(
File "/usr/local/lib/python3.10/dist-packages/datasets/builder.py", line 371, in __init__
self.config, self.config_id = self._create_builder_config(
File "/usr/local/lib/python3.10/dist-packages/datasets/builder.py", line 620, in _create_builder_config
builder_config._resolve_data_files(
File "/usr/local/lib/python3.10/dist-packages/datasets/builder.py", line 211, in _resolve_data_files
self.data_files = self.data_files.resolve(base_path, download_config)
File "/usr/local/lib/python3.10/dist-packages/datasets/data_files.py", line 799, in resolve
out[key] = data_files_patterns_list.resolve(base_path, download_config)
File "/usr/local/lib/python3.10/dist-packages/datasets/data_files.py", line 752, in resolve
resolve_pattern(
File "/usr/local/lib/python3.10/dist-packages/datasets/data_files.py", line 393, in resolve_pattern
raise FileNotFoundError(error_msg)
FileNotFoundError: Unable to find 'hf://datasets/winogrande@ebf71e3c7b5880d019ecf6099c0b09311b1084f5/winogrande_xl/train/0000.parquet' with any supported extension ['.csv', '.tsv', '.json', '.jsonl', '.parquet', '.geoparquet', '.gpq', '.arrow', '.txt', '.tar', '.blp', '.bmp', '.dib', '.bufr', '.cur', '.pcx', '.dcx', '.dds', '.ps', '.eps', '.fit', '.fits', '.fli', '.flc', '.ftc', '.ftu', '.gbr', '.gif', '.grib', '.h5', '.hdf', '.png', '.apng', '.jp2', '.j2k', '.jpc', '.jpf', '.jpx', '.j2c', '.icns', '.ico', '.im', '.iim', '.tif', '.tiff', '.jfif', '.jpe', '.jpg', '.jpeg', '.mpg', '.mpeg', '.msp', '.pcd', '.pxr', '.pbm', '.pgm', '.ppm', '.pnm', '.psd', '.bw', '.rgb', '.rgba', '.sgi', '.ras', '.tga', '.icb', '.vda', '.vst', '.webp', '.wmf', '.emf', '.xbm', '.xpm', '.BLP', '.BMP', '.DIB', '.BUFR', '.CUR', '.PCX', '.DCX', '.DDS', '.PS', '.EPS', '.FIT', '.FITS', '.FLI', '.FLC', '.FTC', '.FTU', '.GBR', '.GIF', '.GRIB', '.H5', '.HDF', '.PNG', '.APNG', '.JP2', '.J2K', '.JPC', '.JPF', '.JPX', '.J2C', '.ICNS', '.ICO', '.IM', '.IIM', '.TIF', '.TIFF', '.JFIF', '.JPE', '.JPG', '.JPEG', '.MPG', '.MPEG', '.MSP', '.PCD', '.PXR', '.PBM', '.PGM', '.PPM', '.PNM', '.PSD', '.BW', '.RGB', '.RGBA', '.SGI', '.RAS', '.TGA', '.ICB', '.VDA', '.VST', '.WEBP', '.WMF', '.EMF', '.XBM', '.XPM', '.aiff', '.au', '.avr', '.caf', '.flac', '.htk', '.svx', '.mat4', '.mat5', '.mpc2k', '.ogg', '.paf', '.pvf', '.raw', '.rf64', '.sd2', '.sds', '.ircam', '.voc', '.w64', '.wav', '.nist', '.wavex', '.wve', '.xi', '.mp3', '.opus', '.AIFF', '.AU', '.AVR', '.CAF', '.FLAC', '.HTK', '.SVX', '.MAT4', '.MAT5', '.MPC2K', '.OGG', '.PAF', '.PVF', '.RAW', '.RF64', '.SD2', '.SDS', '.IRCAM', '.VOC', '.W64', '.WAV', '.NIST', '.WAVEX', '.WVE', '.XI', '.MP3', '.OPUS', '.zip']```
### Steps to reproduce the bug
from datasets import load_dataset
datasets = load_dataset('winogrande','winogrande_xl')
### Expected behavior
```Downloading data: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.06M/2.06M [00:00<00:00, 5.16MB/s]
Downloading data: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 118k/118k [00:00<00:00, 360kB/s]
Downloading data: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 85.9k/85.9k [00:00<00:00, 242kB/s]
Generating train split: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████| 40398/40398 [00:00<00:00, 845491.12 examples/s]
Generating test split: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████| 1767/1767 [00:00<00:00, 362501.11 examples/s]
Generating validation split: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 1267/1267 [00:00<00:00, 318768.11 examples/s]```
### Environment info
datasets version: 1.18.0
Hi ! Do you mean 2.18 ? Can you try to update `fsspec` and `huggingface_hub` ?
```
pip install -U fsspec huggingface_hub
``` | [
-0.7672982811927795,
0.12222111225128174,
0.05378527566790581,
0.08904284238815308,
0.15772297978401184,
0.13006216287612915,
0.45174601674079895,
0.33899053931236267,
-0.04396592080593109,
0.0476512685418129,
0.15696606040000916,
0.5043919682502747,
-0.08829959481954575,
0.148674249649047... |
https://github.com/huggingface/datasets/issues/6814 | `map` with `num_proc` > 1 leads to OOM | Hi ! You can try to reduce `writer_batch_size`. It corresponds to the number of samples that stay in RAM before being flushed to disk | ### Describe the bug
When running `map` on parquet dataset loaded from local machine, the RAM usage increases linearly eventually leading to OOM. I was wondering if I should I save the `cache_file` after every n steps in order to prevent this?
### Steps to reproduce the bug
```
ds = load_dataset("parquet", data_files=dataset_path, split="train")
ds = ds.shard(num_shards=4, index=0)
ds = ds.cast_column("audio", datasets.features.Audio(sampling_rate=16_000))
ds = ds.map(prepare_dataset,
num_proc=32,
writer_batch_size=1000,
keep_in_memory=False,
desc="preprocess dataset")
```
```
def prepare_dataset(batch):
# load audio
sample = batch["audio"]
inputs = feature_extractor(sample["array"], sampling_rate=16000)
batch["input_values"] = inputs.input_values[0]
batch["input_length"] = len(sample["array"].squeeze())
return batch
```
### Expected behavior
It shouldn't run into OOM problem.
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-91-generic-x86_64-with-glibc2.17
- Python version: 3.8.19
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.0.3
- `fsspec` version: 2024.2.0 | 24 | `map` with `num_proc` > 1 leads to OOM
### Describe the bug
When running `map` on parquet dataset loaded from local machine, the RAM usage increases linearly eventually leading to OOM. I was wondering if I should I save the `cache_file` after every n steps in order to prevent this?
### Steps to reproduce the bug
```
ds = load_dataset("parquet", data_files=dataset_path, split="train")
ds = ds.shard(num_shards=4, index=0)
ds = ds.cast_column("audio", datasets.features.Audio(sampling_rate=16_000))
ds = ds.map(prepare_dataset,
num_proc=32,
writer_batch_size=1000,
keep_in_memory=False,
desc="preprocess dataset")
```
```
def prepare_dataset(batch):
# load audio
sample = batch["audio"]
inputs = feature_extractor(sample["array"], sampling_rate=16000)
batch["input_values"] = inputs.input_values[0]
batch["input_length"] = len(sample["array"].squeeze())
return batch
```
### Expected behavior
It shouldn't run into OOM problem.
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-91-generic-x86_64-with-glibc2.17
- Python version: 3.8.19
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.0.3
- `fsspec` version: 2024.2.0
Hi ! You can try to reduce `writer_batch_size`. It corresponds to the number of samples that stay in RAM before being flushed to disk | [
-0.4401771128177643,
-0.40566104650497437,
0.08709584176540375,
0.2516528069972992,
0.11869972199201584,
-0.09441425651311874,
0.20353905856609344,
0.11727042496204376,
0.19048455357551575,
0.1920597404241562,
0.36395880579948425,
0.49853959679603577,
-0.38618892431259155,
0.05605891346931... |
https://github.com/huggingface/datasets/issues/6810 | Allow deleting a subset/config from a no-script dataset | Thanks for your comment, @mariosasko. Or maybe both (in Python and as CLI command)? The Python command would be just the reverse of `push_to_hub`...
I am working on a draft implementation, so we can discuss about the API and UX. | As proposed by @BramVanroy, it would be neat to have this functionality through the API. | 40 | Allow deleting a subset/config from a no-script dataset
As proposed by @BramVanroy, it would be neat to have this functionality through the API.
Thanks for your comment, @mariosasko. Or maybe both (in Python and as CLI command)? The Python command would be just the reverse of `push_to_hub`...
I am working on a draft implementation, so we can discuss about the API and UX. | [
-0.22531366348266602,
-0.2988707423210144,
-0.15333543717861176,
-0.26405125856399536,
-0.31296950578689575,
-0.010717570781707764,
0.3022577166557312,
0.32436999678611755,
0.22865477204322815,
0.3413943648338318,
-0.06061657518148422,
0.28991472721099854,
-0.39521804451942444,
0.604092836... |
https://github.com/huggingface/datasets/issues/6805 | Batched mapping of existing string column casts boolean to string | This seems to be hardcoded behavior in table.py `array_cast`.
```python
if (
not allow_number_to_str
and pa.types.is_string(pa_type)
and (pa.types.is_floating(array.type) or pa.types.is_integer(array.type))
):
raise TypeError(
f"Couldn't cast array of type {array.type} to {pa_type} since allow_number_to_str is set to {allow_number_to_str}"
)
if pa.types.is_null(pa_type) and not pa.types.is_null(array.type):
raise TypeError(f"Couldn't cast array of type {array.type} to {pa_type}")
return array.cast(pa_type)
```
where floats and integers are not cast to string but booleans are.
Maybe this should be extended to booleans? | ### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2 | 74 | Batched mapping of existing string column casts boolean to string
### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2
This seems to be hardcoded behavior in table.py `array_cast`.
```python
if (
not allow_number_to_str
and pa.types.is_string(pa_type)
and (pa.types.is_floating(array.type) or pa.types.is_integer(array.type))
):
raise TypeError(
f"Couldn't cast array of type {array.type} to {pa_type} since allow_number_to_str is set to {allow_number_to_str}"
)
if pa.types.is_null(pa_type) and not pa.types.is_null(array.type):
raise TypeError(f"Couldn't cast array of type {array.type} to {pa_type}")
return array.cast(pa_type)
```
where floats and integers are not cast to string but booleans are.
Maybe this should be extended to booleans? | [
-0.05853521451354027,
-0.1928844004869461,
0.049546901136636734,
-0.00908283144235611,
0.1846773475408554,
0.09128903597593307,
0.5908911824226379,
0.332765132188797,
0.3084041476249695,
0.05234541743993759,
-0.040871575474739075,
0.6944421529769897,
0.05864907428622246,
0.0265546571463346... |
https://github.com/huggingface/datasets/issues/6805 | Batched mapping of existing string column casts boolean to string | I'll gladly create a PR but not sure what the behavior should be.
Should a value returned from map be cast to the current feature?
At the moment this seems very inconsistent since `datetime `is also cast (this would only fix `boolean`) but nested structures are not.
```python
dset = Dataset.from_dict({"a": ["Hello world!"]})
dset = dset.map(lambda x: {"a": date(2021, 1, 1)})
# dset[0]["a"] == '2021-01-01'
```
```python
dset = Dataset.from_dict({"a": ["Hello world!"]})
dset = dset.map(lambda x: {"a": [True]})
# dset[0]["a"] == [True]
```
Is there are reason to cast the value if the user doesn't specify it explicitly?
Seems tricky that some things are cast and some are not. | ### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2 | 109 | Batched mapping of existing string column casts boolean to string
### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2
I'll gladly create a PR but not sure what the behavior should be.
Should a value returned from map be cast to the current feature?
At the moment this seems very inconsistent since `datetime `is also cast (this would only fix `boolean`) but nested structures are not.
```python
dset = Dataset.from_dict({"a": ["Hello world!"]})
dset = dset.map(lambda x: {"a": date(2021, 1, 1)})
# dset[0]["a"] == '2021-01-01'
```
```python
dset = Dataset.from_dict({"a": ["Hello world!"]})
dset = dset.map(lambda x: {"a": [True]})
# dset[0]["a"] == [True]
```
Is there are reason to cast the value if the user doesn't specify it explicitly?
Seems tricky that some things are cast and some are not. | [
-0.0327555388212204,
-0.20743200182914734,
0.06965263187885284,
0.04518108069896698,
0.0458211787045002,
0.06987122446298599,
0.5871598124504089,
0.28447648882865906,
0.357866108417511,
-0.008011311292648315,
0.07180257141590118,
0.5964505076408386,
-0.0023967577144503593,
0.05721408873796... |
https://github.com/huggingface/datasets/issues/6805 | Batched mapping of existing string column casts boolean to string | Indeed, it also makes sense to raise a `TypeError` for temporal and decimal types.
> Is there are reason to cast the value if the user doesn't specify it explicitly?
This is how PyArrow's built-in `cast` behaves - it allows casting from primitive types to strings. Hence, we need `allow_number_to_str` to disallow such casts (e.g., in the [scenario](https://github.com/huggingface/datasets/blob/a3bc89d8bfd47c2a175c3ce16d92b7307cdeafd6/src/datasets/arrow_writer.py#L208) when we are "trying a type" to preserve the original type if there is a column in the output dataset with the same name as in the input one).
PS: In the PR, we can introduce `allow_numeric_to_str` (for floats, integers, decimals, booleans) and `allow_temporal_to_str` (for dates, timestamps, ...) and deprecate `allow_number_to_str` to make it clear what each parameter does. | ### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2 | 117 | Batched mapping of existing string column casts boolean to string
### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2
Indeed, it also makes sense to raise a `TypeError` for temporal and decimal types.
> Is there are reason to cast the value if the user doesn't specify it explicitly?
This is how PyArrow's built-in `cast` behaves - it allows casting from primitive types to strings. Hence, we need `allow_number_to_str` to disallow such casts (e.g., in the [scenario](https://github.com/huggingface/datasets/blob/a3bc89d8bfd47c2a175c3ce16d92b7307cdeafd6/src/datasets/arrow_writer.py#L208) when we are "trying a type" to preserve the original type if there is a column in the output dataset with the same name as in the input one).
PS: In the PR, we can introduce `allow_numeric_to_str` (for floats, integers, decimals, booleans) and `allow_temporal_to_str` (for dates, timestamps, ...) and deprecate `allow_number_to_str` to make it clear what each parameter does. | [
-0.14343000948429108,
-0.04710235446691513,
0.09535335004329681,
0.007400814443826675,
0.309475839138031,
0.06637474149465561,
0.5846834778785706,
0.3394041359424591,
0.0745493620634079,
0.14348801970481873,
-0.07709448039531708,
0.6000705361366272,
-0.05066084861755371,
-0.044704377651214... |
https://github.com/huggingface/datasets/issues/6805 | Batched mapping of existing string column casts boolean to string | Would just `allow_primitive_to_str` work?
This should include all `numeric`, `boolean `and `temporal`formats.
Note that at least in the [ C++ implementation](https://arrow.apache.org/docs/cpp/api/utilities.html#_CPPv410is_numericRK8DataType) `numeric `seems to exclude `boolean`.
[](https://arrow.apache.org/docs/cpp/api/utilities.html#_CPPv410is_numericRK8DataType) | ### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2 | 27 | Batched mapping of existing string column casts boolean to string
### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2
Would just `allow_primitive_to_str` work?
This should include all `numeric`, `boolean `and `temporal`formats.
Note that at least in the [ C++ implementation](https://arrow.apache.org/docs/cpp/api/utilities.html#_CPPv410is_numericRK8DataType) `numeric `seems to exclude `boolean`.
[](https://arrow.apache.org/docs/cpp/api/utilities.html#_CPPv410is_numericRK8DataType) | [
-0.1160435676574707,
-0.13926109671592712,
0.1097707599401474,
-0.10254994034767151,
0.20062807202339172,
0.07319114357233047,
0.5531147122383118,
0.3770013451576233,
0.24099013209342957,
0.041817888617515564,
0.03618139028549194,
0.6365835666656494,
0.009841300547122955,
0.059536390006542... |
https://github.com/huggingface/datasets/issues/6805 | Batched mapping of existing string column casts boolean to string | Indeed, `allow_primitive_to_str` sounds better.
PS: PyArrow's `pa.types.is_primitive` returns `False` for decimal types, but I think is okay for us to treat decimals as primitive types (or we can have `allow_decimal_to_str` to be fully consistent with PyArrow) | ### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2 | 36 | Batched mapping of existing string column casts boolean to string
### Describe the bug
Let the dataset contain a column named 'a', which is of the string type.
If 'a' is converted to a boolean using batched mapping, the mapper automatically casts the boolean to a string (e.g., True -> 'true').
It only happens when the original column and the mapped column name are identical.
Thank you!
### Steps to reproduce the bug
```python
from datasets import Dataset
dset = Dataset.from_dict({'a': ['11', '22']})
dset = dset.map(lambda x: {'a': [True for _ in x['a']]}, batched=True)
print(dset['a'])
```
```
> ['true', 'true']
```
### Expected behavior
[True, True]
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-5.4.0-148-generic-x86_64-with-glibc2.31
- Python version: 3.10.13
- `huggingface_hub` version: 0.21.4
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2023.12.2
Indeed, `allow_primitive_to_str` sounds better.
PS: PyArrow's `pa.types.is_primitive` returns `False` for decimal types, but I think is okay for us to treat decimals as primitive types (or we can have `allow_decimal_to_str` to be fully consistent with PyArrow) | [
-0.1349276453256607,
-0.06464986503124237,
0.08175201714038849,
-0.045226141810417175,
0.22103580832481384,
0.06167697161436081,
0.5447356700897217,
0.37978339195251465,
0.13651803135871887,
0.09913846850395203,
-0.034818291664123535,
0.6482187509536743,
0.05565078184008598,
0.007406229153... |
https://github.com/huggingface/datasets/issues/6801 | got fileNotFound | Hi! I'll open a PR on the Hub to fix this, but please use the Hub's [Community tab](https://huggingface.co/datasets/nyanko7/danbooru2023/discussions) to report such issues in the future. | ### Describe the bug
When I use load_dataset to load the nyanko7/danbooru2023 data set, the cache is read in the form of a symlink. There may be a problem with the arrow_dataset initialization process and I get FileNotFoundError: [Errno 2] No such file or directory: '2945000.jpg'
### Steps to reproduce the bug
#code show as below
from datasets import load_dataset
data = load_dataset("nyanko7/danbooru2023",cache_dir=<symlink>)
data["train"][0]
### Expected behavior
I should get this result:
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=365x256 at 0x7FB730CB4070>, 'label': 0}
### Environment info
datasets==2.12.0
python==3.10.14
| 25 | got fileNotFound
### Describe the bug
When I use load_dataset to load the nyanko7/danbooru2023 data set, the cache is read in the form of a symlink. There may be a problem with the arrow_dataset initialization process and I get FileNotFoundError: [Errno 2] No such file or directory: '2945000.jpg'
### Steps to reproduce the bug
#code show as below
from datasets import load_dataset
data = load_dataset("nyanko7/danbooru2023",cache_dir=<symlink>)
data["train"][0]
### Expected behavior
I should get this result:
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=365x256 at 0x7FB730CB4070>, 'label': 0}
### Environment info
datasets==2.12.0
python==3.10.14
Hi! I'll open a PR on the Hub to fix this, but please use the Hub's [Community tab](https://huggingface.co/datasets/nyanko7/danbooru2023/discussions) to report such issues in the future. | [
-0.1854178011417389,
-0.30319511890411377,
0.07290211319923401,
0.46887606382369995,
0.08416895568370819,
-0.0029570162296295166,
0.5426899194717407,
0.2829395532608032,
0.4018489420413971,
0.0466868132352829,
0.11722849309444427,
0.09341808408498764,
-0.19445611536502838,
-0.1978476941585... |
https://github.com/huggingface/datasets/issues/6800 | High overhead when loading lots of subsets from the same dataset | Hi !
It's possible to multiple files at once:
```python
data_files = "data/*.jsonl"
# Or pass a list of files
langs = ['ka-ml', 'br-sr', 'ka-pt', 'id-ko', ..., 'fi-ze_zh', 'he-kk', 'ka-tr']
data_files = [f"data/{lang}.jsonl" for lang in langs]
ds = load_dataset("loicmagne/open-subtitles-250-bitext-mining", data_files=data_files, split="train")
```
Also maybe you can add a subset called "all" for people that want to load all the data without having to list all the languages ?
```yaml
- config_name: all
data_files: data/*.jsonl
```
| ### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
| 76 | High overhead when loading lots of subsets from the same dataset
### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
Hi !
It's possible to multiple files at once:
```python
data_files = "data/*.jsonl"
# Or pass a list of files
langs = ['ka-ml', 'br-sr', 'ka-pt', 'id-ko', ..., 'fi-ze_zh', 'he-kk', 'ka-tr']
data_files = [f"data/{lang}.jsonl" for lang in langs]
ds = load_dataset("loicmagne/open-subtitles-250-bitext-mining", data_files=data_files, split="train")
```
Also maybe you can add a subset called "all" for people that want to load all the data without having to list all the languages ?
```yaml
- config_name: all
data_files: data/*.jsonl
```
| [
-0.3364805281162262,
-0.5688940286636353,
-0.09787166118621826,
0.5463959574699402,
-0.1324806660413742,
-0.007739365100860596,
0.07414510101079941,
0.28510570526123047,
0.3549295663833618,
0.07160203158855438,
-0.4010574221611023,
0.19697675108909607,
0.0950620174407959,
0.097148947417736... |
https://github.com/huggingface/datasets/issues/6800 | High overhead when loading lots of subsets from the same dataset | Thanks for your reply, it is indeed much faster, however the result is a dataset where all the subsets are "merged" together, the language pair is lost:
```
DatasetDict({
train: Dataset({
features: ['sentence1', 'sentence2'],
num_rows: 247809
})
})
```
I guess I could add a 'lang' feature for each row in the dataset, is there a better way to do it ? | ### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
| 62 | High overhead when loading lots of subsets from the same dataset
### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
Thanks for your reply, it is indeed much faster, however the result is a dataset where all the subsets are "merged" together, the language pair is lost:
```
DatasetDict({
train: Dataset({
features: ['sentence1', 'sentence2'],
num_rows: 247809
})
})
```
I guess I could add a 'lang' feature for each row in the dataset, is there a better way to do it ? | [
-0.3364805281162262,
-0.5688940286636353,
-0.09787166118621826,
0.5463959574699402,
-0.1324806660413742,
-0.007739365100860596,
0.07414510101079941,
0.28510570526123047,
0.3549295663833618,
0.07160203158855438,
-0.4010574221611023,
0.19697675108909607,
0.0950620174407959,
0.097148947417736... |
https://github.com/huggingface/datasets/issues/6800 | High overhead when loading lots of subsets from the same dataset | Hi @lhoestq over at https://github.com/embeddings-benchmark/mteb/issues/530 we have started examining these issues and would love to make a PR for datasets if we believe there is a way to improve the speed. As I assume you have a better overview than me @lhoestq, would you be interested in a PR, and might you have an idea about where we would start working on it?
We see a speed comparison of
1. 15 minutes (for ~20% of the languages) when loaded using a for loop
2. 17 minutes using the your suggestion
3. ~30 seconds when using @loicmagne "merged" method.
Worth mentioning is that solution 2 looses the language information. | ### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
| 108 | High overhead when loading lots of subsets from the same dataset
### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
Hi @lhoestq over at https://github.com/embeddings-benchmark/mteb/issues/530 we have started examining these issues and would love to make a PR for datasets if we believe there is a way to improve the speed. As I assume you have a better overview than me @lhoestq, would you be interested in a PR, and might you have an idea about where we would start working on it?
We see a speed comparison of
1. 15 minutes (for ~20% of the languages) when loaded using a for loop
2. 17 minutes using the your suggestion
3. ~30 seconds when using @loicmagne "merged" method.
Worth mentioning is that solution 2 looses the language information. | [
-0.3364805281162262,
-0.5688940286636353,
-0.09787166118621826,
0.5463959574699402,
-0.1324806660413742,
-0.007739365100860596,
0.07414510101079941,
0.28510570526123047,
0.3549295663833618,
0.07160203158855438,
-0.4010574221611023,
0.19697675108909607,
0.0950620174407959,
0.097148947417736... |
https://github.com/huggingface/datasets/issues/6800 | High overhead when loading lots of subsets from the same dataset | Can you retry using `datasets` 2.19 ? We improved a lot the speed of downloading datasets with tons of small files.
```
pip install -U datasets
```
Now this takes 17sec on my side instead of the 17min minutes @loicmagne mentioned :)
```python
>>> %time ds = load_dataset("loicmagne/open-subtitles-250-bitext-mining", data_files="data/*.jsonl")
Downloading readme: 100%|█████████████████████████████████| 13.7k/13.7k [00:00<00:00, 5.47MB/s]
Resolving data files: 100%|█████████████████████████████████| 250/250 [00:00<00:00, 612.51it/s]
Downloading data: 100%|██████████████████████████████████| 250/250 [00:12<00:00, 19.68files/s]
Generating train split: 247809 examples [00:00, 1057071.08 examples/s]
CPU times: user 4.95 s, sys: 3.1 s, total: 8.05 s
Wall time: 17.4 s
``` | ### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
| 92 | High overhead when loading lots of subsets from the same dataset
### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
Can you retry using `datasets` 2.19 ? We improved a lot the speed of downloading datasets with tons of small files.
```
pip install -U datasets
```
Now this takes 17sec on my side instead of the 17min minutes @loicmagne mentioned :)
```python
>>> %time ds = load_dataset("loicmagne/open-subtitles-250-bitext-mining", data_files="data/*.jsonl")
Downloading readme: 100%|█████████████████████████████████| 13.7k/13.7k [00:00<00:00, 5.47MB/s]
Resolving data files: 100%|█████████████████████████████████| 250/250 [00:00<00:00, 612.51it/s]
Downloading data: 100%|██████████████████████████████████| 250/250 [00:12<00:00, 19.68files/s]
Generating train split: 247809 examples [00:00, 1057071.08 examples/s]
CPU times: user 4.95 s, sys: 3.1 s, total: 8.05 s
Wall time: 17.4 s
``` | [
-0.3364805281162262,
-0.5688940286636353,
-0.09787166118621826,
0.5463959574699402,
-0.1324806660413742,
-0.007739365100860596,
0.07414510101079941,
0.28510570526123047,
0.3549295663833618,
0.07160203158855438,
-0.4010574221611023,
0.19697675108909607,
0.0950620174407959,
0.097148947417736... |
https://github.com/huggingface/datasets/issues/6800 | High overhead when loading lots of subsets from the same dataset | > Can you retry using `datasets` 2.19 ? We improved a lot the speed of downloading datasets with tons of small files.
>
> ```
> pip install -U datasets
> ```
>
> Now this takes 17sec on my side instead of the 17min minutes @loicmagne mentioned :)
>
> ```python
> >>> %time ds = load_dataset("loicmagne/open-subtitles-250-bitext-mining", data_files="data/*.jsonl")
> Downloading readme: 100%|█████████████████████████████████| 13.7k/13.7k [00:00<00:00, 5.47MB/s]
> Resolving data files: 100%|█████████████████████████████████| 250/250 [00:00<00:00, 612.51it/s]
> Downloading data: 100%|██████████████████████████████████| 250/250 [00:12<00:00, 19.68files/s]
> Generating train split: 247809 examples [00:00, 1057071.08 examples/s]
> CPU times: user 4.95 s, sys: 3.1 s, total: 8.05 s
> Wall time: 17.4 s
> ```
I was actually just noticing that, I bumped from 2.18 to 2.19 and got a massive speedup, amazing!
About the fact that subset names are lost when loading all files at once, currently my solution is to add a 'lang' feature to each rows, convert to polars and use:
```python
ds_split = ds.to_polars().group_by('lang')
```
It's fast so I think it's an acceptable solution, but is there a better way to do it ? | ### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
| 182 | High overhead when loading lots of subsets from the same dataset
### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
> Can you retry using `datasets` 2.19 ? We improved a lot the speed of downloading datasets with tons of small files.
>
> ```
> pip install -U datasets
> ```
>
> Now this takes 17sec on my side instead of the 17min minutes @loicmagne mentioned :)
>
> ```python
> >>> %time ds = load_dataset("loicmagne/open-subtitles-250-bitext-mining", data_files="data/*.jsonl")
> Downloading readme: 100%|█████████████████████████████████| 13.7k/13.7k [00:00<00:00, 5.47MB/s]
> Resolving data files: 100%|█████████████████████████████████| 250/250 [00:00<00:00, 612.51it/s]
> Downloading data: 100%|██████████████████████████████████| 250/250 [00:12<00:00, 19.68files/s]
> Generating train split: 247809 examples [00:00, 1057071.08 examples/s]
> CPU times: user 4.95 s, sys: 3.1 s, total: 8.05 s
> Wall time: 17.4 s
> ```
I was actually just noticing that, I bumped from 2.18 to 2.19 and got a massive speedup, amazing!
About the fact that subset names are lost when loading all files at once, currently my solution is to add a 'lang' feature to each rows, convert to polars and use:
```python
ds_split = ds.to_polars().group_by('lang')
```
It's fast so I think it's an acceptable solution, but is there a better way to do it ? | [
-0.3364805281162262,
-0.5688940286636353,
-0.09787166118621826,
0.5463959574699402,
-0.1324806660413742,
-0.007739365100860596,
0.07414510101079941,
0.28510570526123047,
0.3549295663833618,
0.07160203158855438,
-0.4010574221611023,
0.19697675108909607,
0.0950620174407959,
0.097148947417736... |
https://github.com/huggingface/datasets/issues/6800 | High overhead when loading lots of subsets from the same dataset | It's the fastest way I think :)
Alternatively you can download the dataset repository locally using [huggingface_hub](https://huggingface.co/docs/huggingface_hub/guides/download) (either via CLI or in python) and load the subsets one by one locally using a for loop as you were doing before (just pass the directory path to load_dataset instead of the dataset_id). | ### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
| 51 | High overhead when loading lots of subsets from the same dataset
### Describe the bug
I have a multilingual dataset that contains a lot of subsets. Each subset corresponds to a pair of languages, you can see here an example with 250 subsets: [https://hf.co/datasets/loicmagne/open-subtitles-250-bitext-mining](). As part of the MTEB benchmark, we may need to load all the subsets of the dataset. The dataset is relatively small and contains only ~45MB of data, but when I try to load every subset, it takes 15 minutes from the HF hub and 13 minutes from the cache
This issue https://github.com/huggingface/datasets/issues/5499 also referenced this overhead, but I'm wondering if there is anything I can do to speedup loading different subsets of the same dataset, both when loading from disk and from the HF hub? Currently each subset is stored in a jsonl file
### Steps to reproduce the bug
```
from datasets import load_dataset
for subset in ['ka-ml', 'br-sr', 'bg-br', 'kk-lv', 'br-sk', 'br-fi', 'eu-ze_zh', 'kk-nl', 'kk-vi', 'ja-kk', 'br-sv', 'kk-zh_cn', 'kk-ms', 'br-et', 'br-hu', 'eo-kk', 'br-tr', 'ko-tl', 'te-zh_tw', 'br-hr', 'br-nl', 'ka-si', 'br-cs', 'br-is', 'br-ro', 'br-de', 'et-kk', 'fr-hy', 'br-no', 'is-ko', 'br-da', 'br-en', 'eo-lt', 'is-ze_zh', 'eu-ko', 'br-it', 'br-id', 'eu-zh_cn', 'is-ja', 'br-sl', 'br-gl', 'br-pt_br', 'br-es', 'br-pt', 'is-th', 'fa-is', 'br-ca', 'eu-ka', 'is-zh_cn', 'eu-ur', 'id-kk', 'br-sq', 'eu-ja', 'uk-ur', 'is-zh_tw', 'ka-ko', 'eu-zh_tw', 'eu-th', 'eu-is', 'is-tl', 'br-eo', 'eo-ze_zh', 'eu-te', 'ar-kk', 'eo-lv', 'ko-ze_zh', 'ml-ze_zh', 'is-lt', 'br-fr', 'ko-te', 'kk-sl', 'eu-fa', 'eo-ko', 'ka-ze_en', 'eo-eu', 'ta-zh_tw', 'eu-lv', 'ko-lv', 'lt-tl', 'eu-si', 'hy-ru', 'ar-is', 'eu-lt', 'eu-tl', 'eu-uk', 'ka-ze_zh', 'si-ze_zh', 'el-is', 'bn-is', 'ko-ze_en', 'eo-si', 'cs-kk', 'is-uk', 'eu-ze_en', 'ta-ze_zh', 'is-pl', 'is-mk', 'eu-ta', 'ko-lt', 'is-lv', 'fa-ko', 'bn-ko', 'hi-is', 'bn-ze_zh', 'bn-eu', 'bn-ja', 'is-ml', 'eu-ru', 'ko-ta', 'is-vi', 'ja-tl', 'eu-mk', 'eu-he', 'ka-zh_tw', 'ka-zh_cn', 'si-tl', 'is-kk', 'eu-fi', 'fi-ko', 'is-ur', 'ka-th', 'ko-ur', 'eo-ja', 'he-is', 'is-tr', 'ka-ur', 'et-ko', 'eu-vi', 'is-sk', 'gl-is', 'fr-is', 'is-sq', 'hu-is', 'fr-kk', 'eu-sq', 'is-ru', 'ja-ka', 'fi-tl', 'ka-lv', 'fi-is', 'is-si', 'ar-ko', 'ko-sl', 'ar-eu', 'ko-si', 'bg-is', 'eu-hu', 'ko-sv', 'bn-hu', 'kk-ro', 'eu-hi', 'ka-ms', 'ko-th', 'ko-sr', 'ko-mk', 'fi-kk', 'ka-vi', 'eu-ml', 'ko-ml', 'de-ko', 'fa-ze_zh', 'eu-sk', 'is-sl', 'et-is', 'eo-is', 'is-sr', 'is-ze_en', 'kk-pt_br', 'hr-hy', 'kk-pl', 'ja-ta', 'is-ms', 'hi-ze_en', 'is-ro', 'ko-zh_cn', 'el-eu', 'ka-pl', 'ka-sq', 'eu-sl', 'fa-ka', 'ko-no', 'si-ze_en', 'ko-uk', 'ja-ze_zh', 'hu-ko', 'kk-no', 'eu-pl', 'is-pt_br', 'bn-lv', 'tl-zh_cn', 'is-nl', 'he-ko', 'ko-sq', 'ta-th', 'lt-ta', 'da-ko', 'ca-is', 'is-ta', 'bn-fi', 'ja-ml', 'lv-si', 'eu-sv', 'ja-te', 'bn-ur', 'bn-ca', 'bs-ko', 'bs-is', 'eu-sr', 'ko-vi', 'ko-zh_tw', 'et-tl', 'kk-tr', 'eo-vi', 'is-it', 'ja-ko', 'eo-et', 'id-is', 'bn-et', 'bs-eu', 'bn-lt', 'tl-uk', 'bn-zh_tw', 'da-eu', 'el-ko', 'no-tl', 'ko-sk', 'is-pt', 'hu-kk', 'si-zh_tw', 'si-te', 'ka-ru', 'lt-ml', 'af-ja', 'bg-eu', 'eo-th', 'cs-is', 'pl-ze_zh', 'el-kk', 'kk-sv', 'ka-nl', 'ko-pl', 'bg-ko', 'ka-pt_br', 'et-eu', 'tl-zh_tw', 'ka-pt', 'id-ko', 'fi-ze_zh', 'he-kk', 'ka-tr']:
load_dataset('loicmagne/open-subtitles-250-bitext-mining', subset)
```
### Expected behavior
Faster loading?
### Environment info
Copy-and-paste the text below in your GitHub issue.
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-27-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.2
- `fsspec` version: 2023.5.0
It's the fastest way I think :)
Alternatively you can download the dataset repository locally using [huggingface_hub](https://huggingface.co/docs/huggingface_hub/guides/download) (either via CLI or in python) and load the subsets one by one locally using a for loop as you were doing before (just pass the directory path to load_dataset instead of the dataset_id). | [
-0.3364805281162262,
-0.5688940286636353,
-0.09787166118621826,
0.5463959574699402,
-0.1324806660413742,
-0.007739365100860596,
0.07414510101079941,
0.28510570526123047,
0.3549295663833618,
0.07160203158855438,
-0.4010574221611023,
0.19697675108909607,
0.0950620174407959,
0.097148947417736... |
https://github.com/huggingface/datasets/issues/6798 | `DatasetBuilder._split_generators` incomplete type annotation | There is also the [`MockDownloadManager`](https://github.com/JonasLoos/datasets/blob/main/src/datasets/download/mock_download_manager.py#L33), which seems like it might get passed here too. However, to me, it doesn't really seem relevant to the users of the datasets library, so I would just ignore it. What do you think, @mariosasko? | ### Describe the bug
The [`DatasetBuilder._split_generators`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/builder.py#L1449) function has currently the following signature:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: DownloadManager):
...
```
However, the `dl_manager` argument can also be of type [`StreamingDownloadManager`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/download/streaming_download_manager.py#L962), which has different functionality. For example, the `download` function doesn't download, but rather just returns the given url(s).
I suggest changing the function signature to:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: Union[DownloadManager, StreamingDownloadManager]):
...
```
and also adjust the docstring accordingly.
I would like to create a Pull Request to fix this, and have the following questions:
* Are there also other options than `DownloadManager`, and `StreamingDownloadManager`?
* Should this also be changed in other functions?
### Steps to reproduce the bug
Minimal example to print the different class names:
```python
import tempfile
from datasets import load_dataset
example = b'''
from datasets import GeneratorBasedBuilder, DatasetInfo, Features, Value, SplitGenerator
class Test(GeneratorBasedBuilder):
def _info(self):
return DatasetInfo(features=Features({"x": Value("int64")}))
def _split_generators(self, dl_manager):
print(type(dl_manager))
return [SplitGenerator('test')]
def _generate_examples(self):
yield 0, {'x': 42}
'''
with tempfile.NamedTemporaryFile(suffix='.py') as f:
f.write(example)
f.flush()
load_dataset(f.name, streaming=False)
load_dataset(f.name, streaming=True)
```
### Expected behavior
complete type annotations
### Environment info
/ | 40 | `DatasetBuilder._split_generators` incomplete type annotation
### Describe the bug
The [`DatasetBuilder._split_generators`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/builder.py#L1449) function has currently the following signature:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: DownloadManager):
...
```
However, the `dl_manager` argument can also be of type [`StreamingDownloadManager`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/download/streaming_download_manager.py#L962), which has different functionality. For example, the `download` function doesn't download, but rather just returns the given url(s).
I suggest changing the function signature to:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: Union[DownloadManager, StreamingDownloadManager]):
...
```
and also adjust the docstring accordingly.
I would like to create a Pull Request to fix this, and have the following questions:
* Are there also other options than `DownloadManager`, and `StreamingDownloadManager`?
* Should this also be changed in other functions?
### Steps to reproduce the bug
Minimal example to print the different class names:
```python
import tempfile
from datasets import load_dataset
example = b'''
from datasets import GeneratorBasedBuilder, DatasetInfo, Features, Value, SplitGenerator
class Test(GeneratorBasedBuilder):
def _info(self):
return DatasetInfo(features=Features({"x": Value("int64")}))
def _split_generators(self, dl_manager):
print(type(dl_manager))
return [SplitGenerator('test')]
def _generate_examples(self):
yield 0, {'x': 42}
'''
with tempfile.NamedTemporaryFile(suffix='.py') as f:
f.write(example)
f.flush()
load_dataset(f.name, streaming=False)
load_dataset(f.name, streaming=True)
```
### Expected behavior
complete type annotations
### Environment info
/
There is also the [`MockDownloadManager`](https://github.com/JonasLoos/datasets/blob/main/src/datasets/download/mock_download_manager.py#L33), which seems like it might get passed here too. However, to me, it doesn't really seem relevant to the users of the datasets library, so I would just ignore it. What do you think, @mariosasko? | [
-0.263654887676239,
0.07787889242172241,
0.03328349068760872,
0.39026403427124023,
0.226485475897789,
-0.071620874106884,
0.2927022874355316,
0.14622260630130768,
-0.057447630912065506,
-0.030006587505340576,
0.024113286286592484,
0.2629048526287079,
-0.156215637922287,
0.0744917243719101,... |
https://github.com/huggingface/datasets/issues/6798 | `DatasetBuilder._split_generators` incomplete type annotation | The API (`dummy_data` CLI command ) that uses the `MockDownloadManager` has been deprecated, so ignoring it sounds good! | ### Describe the bug
The [`DatasetBuilder._split_generators`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/builder.py#L1449) function has currently the following signature:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: DownloadManager):
...
```
However, the `dl_manager` argument can also be of type [`StreamingDownloadManager`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/download/streaming_download_manager.py#L962), which has different functionality. For example, the `download` function doesn't download, but rather just returns the given url(s).
I suggest changing the function signature to:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: Union[DownloadManager, StreamingDownloadManager]):
...
```
and also adjust the docstring accordingly.
I would like to create a Pull Request to fix this, and have the following questions:
* Are there also other options than `DownloadManager`, and `StreamingDownloadManager`?
* Should this also be changed in other functions?
### Steps to reproduce the bug
Minimal example to print the different class names:
```python
import tempfile
from datasets import load_dataset
example = b'''
from datasets import GeneratorBasedBuilder, DatasetInfo, Features, Value, SplitGenerator
class Test(GeneratorBasedBuilder):
def _info(self):
return DatasetInfo(features=Features({"x": Value("int64")}))
def _split_generators(self, dl_manager):
print(type(dl_manager))
return [SplitGenerator('test')]
def _generate_examples(self):
yield 0, {'x': 42}
'''
with tempfile.NamedTemporaryFile(suffix='.py') as f:
f.write(example)
f.flush()
load_dataset(f.name, streaming=False)
load_dataset(f.name, streaming=True)
```
### Expected behavior
complete type annotations
### Environment info
/ | 18 | `DatasetBuilder._split_generators` incomplete type annotation
### Describe the bug
The [`DatasetBuilder._split_generators`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/builder.py#L1449) function has currently the following signature:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: DownloadManager):
...
```
However, the `dl_manager` argument can also be of type [`StreamingDownloadManager`](https://github.com/huggingface/datasets/blob/0f27d7b77c73412cfc50b24354bfd7a3e838202f/src/datasets/download/streaming_download_manager.py#L962), which has different functionality. For example, the `download` function doesn't download, but rather just returns the given url(s).
I suggest changing the function signature to:
```python
class DatasetBuilder:
def _split_generators(self, dl_manager: Union[DownloadManager, StreamingDownloadManager]):
...
```
and also adjust the docstring accordingly.
I would like to create a Pull Request to fix this, and have the following questions:
* Are there also other options than `DownloadManager`, and `StreamingDownloadManager`?
* Should this also be changed in other functions?
### Steps to reproduce the bug
Minimal example to print the different class names:
```python
import tempfile
from datasets import load_dataset
example = b'''
from datasets import GeneratorBasedBuilder, DatasetInfo, Features, Value, SplitGenerator
class Test(GeneratorBasedBuilder):
def _info(self):
return DatasetInfo(features=Features({"x": Value("int64")}))
def _split_generators(self, dl_manager):
print(type(dl_manager))
return [SplitGenerator('test')]
def _generate_examples(self):
yield 0, {'x': 42}
'''
with tempfile.NamedTemporaryFile(suffix='.py') as f:
f.write(example)
f.flush()
load_dataset(f.name, streaming=False)
load_dataset(f.name, streaming=True)
```
### Expected behavior
complete type annotations
### Environment info
/
The API (`dummy_data` CLI command ) that uses the `MockDownloadManager` has been deprecated, so ignoring it sounds good! | [
-0.263654887676239,
0.07787889242172241,
0.03328349068760872,
0.39026403427124023,
0.226485475897789,
-0.071620874106884,
0.2927022874355316,
0.14622260630130768,
-0.057447630912065506,
-0.030006587505340576,
0.024113286286592484,
0.2629048526287079,
-0.156215637922287,
0.0744917243719101,... |
https://github.com/huggingface/datasets/issues/6796 | CI is broken due to hf-internal-testing/dataset_with_script | Finally:
- the initial issue seems it was temporary
- there is a different issue now: https://github.com/huggingface/datasets/actions/runs/8627153993/job/23646584590?pr=6797
```
FAILED tests/test_load.py::ModuleFactoryTest::test_HubDatasetModuleFactoryWithParquetExport - datasets.utils._dataset_viewer.DatasetViewerError: No exported Parquet files available.
FAILED tests/test_load.py::ModuleFactoryTest::test_HubDatasetModuleFactoryWithParquetExport_errors_on_wrong_sha - datasets.utils._dataset_viewer.DatasetViewerError: No exported Parquet files available.
FAILED tests/test_load.py::test_load_dataset_builder_for_community_dataset_with_script - AssertionError: assert 'dataset_with_script' == 'parquet'
- parquet
+ dataset_with_script
```
Maybe related to `hf-internal-testing/dataset_with_script` dataset: https://huggingface.co/datasets/hf-internal-testing/dataset_with_script | CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
``` | 55 | CI is broken due to hf-internal-testing/dataset_with_script
CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
```
Finally:
- the initial issue seems it was temporary
- there is a different issue now: https://github.com/huggingface/datasets/actions/runs/8627153993/job/23646584590?pr=6797
```
FAILED tests/test_load.py::ModuleFactoryTest::test_HubDatasetModuleFactoryWithParquetExport - datasets.utils._dataset_viewer.DatasetViewerError: No exported Parquet files available.
FAILED tests/test_load.py::ModuleFactoryTest::test_HubDatasetModuleFactoryWithParquetExport_errors_on_wrong_sha - datasets.utils._dataset_viewer.DatasetViewerError: No exported Parquet files available.
FAILED tests/test_load.py::test_load_dataset_builder_for_community_dataset_with_script - AssertionError: assert 'dataset_with_script' == 'parquet'
- parquet
+ dataset_with_script
```
Maybe related to `hf-internal-testing/dataset_with_script` dataset: https://huggingface.co/datasets/hf-internal-testing/dataset_with_script | [
-0.4997487962245941,
0.004399195313453674,
0.029181867837905884,
-0.011295028030872345,
0.2258736789226532,
-0.15919221937656403,
0.3541249632835388,
0.2953513562679291,
0.08836826682090759,
0.2188534140586853,
-0.030934633687138557,
0.23483261466026306,
0.007308386266231537,
0.60864353179... |
https://github.com/huggingface/datasets/issues/6796 | CI is broken due to hf-internal-testing/dataset_with_script | This URL: https://datasets-server.huggingface.co/parquet?dataset=hf-internal-testing/dataset_with_script
raises:
> {"error":"The dataset viewer doesn't support this dataset because it runs arbitrary python code. Please open a discussion in the discussion tab if you think this is an error and tag @lhoestq and @severo."}
Was there a recent change on the Hub enforcing this behavior? | CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
``` | 49 | CI is broken due to hf-internal-testing/dataset_with_script
CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
```
This URL: https://datasets-server.huggingface.co/parquet?dataset=hf-internal-testing/dataset_with_script
raises:
> {"error":"The dataset viewer doesn't support this dataset because it runs arbitrary python code. Please open a discussion in the discussion tab if you think this is an error and tag @lhoestq and @severo."}
Was there a recent change on the Hub enforcing this behavior? | [
-0.471706748008728,
-0.15549853444099426,
0.026893705129623413,
-0.13122083246707916,
0.05236941576004028,
-0.14564239978790283,
0.4287434220314026,
0.19071753323078156,
0.3088235557079315,
0.30294346809387207,
-0.1492748111486435,
0.2744884192943573,
-0.031457897275686264,
0.5667695999145... |
https://github.com/huggingface/datasets/issues/6796 | CI is broken due to hf-internal-testing/dataset_with_script | OK, I just saw this PR:
- https://github.com/huggingface/dataset-viewer/pull/2689
Once merged and deployed, it should fix the issue. | CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
``` | 17 | CI is broken due to hf-internal-testing/dataset_with_script
CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
```
OK, I just saw this PR:
- https://github.com/huggingface/dataset-viewer/pull/2689
Once merged and deployed, it should fix the issue. | [
-0.5324453711509705,
-0.08422461897134781,
0.017368055880069733,
-0.07246415317058563,
0.2465011328458786,
-0.11196476221084595,
0.3603888154029846,
0.4335522949695587,
0.15042458474636078,
0.33519166707992554,
-0.18170683085918427,
0.1648714393377304,
0.08324290812015533,
0.51768571138381... |
https://github.com/huggingface/datasets/issues/6796 | CI is broken due to hf-internal-testing/dataset_with_script | Once the script-dataset has been allowed in the dataset-viewer, we should fix our test to make the CI pass.
I am addressing this. | CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
``` | 23 | CI is broken due to hf-internal-testing/dataset_with_script
CI is broken for test_load_dataset_distributed_with_script. See: https://github.com/huggingface/datasets/actions/runs/8614926216/job/23609378127
```
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[None] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0c741de3b0>)
FAILED tests/test_load.py::test_load_dataset_distributed_with_script[force_redownload] - assert False
+ where False = all(<generator object test_load_dataset_distributed_with_script.<locals>.<genexpr> at 0x7f0be45f6ea0>)
```
Once the script-dataset has been allowed in the dataset-viewer, we should fix our test to make the CI pass.
I am addressing this. | [
-0.5685020089149475,
-0.028953053057193756,
-0.04315004497766495,
-0.09804169833660126,
0.19765886664390564,
-0.18587946891784668,
0.3800262212753296,
0.3463384211063385,
0.21350279450416565,
0.31035852432250977,
-0.10303186625242233,
0.1566164195537567,
0.13097599148750305,
0.592073440551... |
https://github.com/huggingface/datasets/issues/6791 | `add_faiss_index` raises ValueError: not enough values to unpack (expected 2, got 1) | I realized I was passing a string column to this instead of float. Is it possible to add a warning or error to prevent users from falsely believing there's a bug? | ### Describe the bug
Calling `add_faiss_index` on a `Dataset` with a column argument raises a ValueError. The following is the trace
```python
214 def replacement_add(self, x):
215 """Adds vectors to the index.
216 The index must be trained before vectors can be added to it.
217 The vectors are implicitly numbered in sequence. When `n` vectors are
(...)
224 `dtype` must be float32.
225 """
--> 227 n, d = x.shape
228 assert d == self.d
229 x = np.ascontiguousarray(x, dtype='float32')
ValueError: not enough values to unpack (expected 2, got 1)
```
### Steps to reproduce the bug
1. Load any dataset like `ds = datasets.load_dataset("wikimedia/wikipedia", "20231101.en")["train"]`
2. Add an FAISS index on any column `ds.add_faiss_index('title')`
### Expected behavior
The index should be created
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-26-generic-x86_64-with-glibc2.35
- Python version: 3.9.19
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
- `faiss-cpu` version: 1.8.0 | 31 | `add_faiss_index` raises ValueError: not enough values to unpack (expected 2, got 1)
### Describe the bug
Calling `add_faiss_index` on a `Dataset` with a column argument raises a ValueError. The following is the trace
```python
214 def replacement_add(self, x):
215 """Adds vectors to the index.
216 The index must be trained before vectors can be added to it.
217 The vectors are implicitly numbered in sequence. When `n` vectors are
(...)
224 `dtype` must be float32.
225 """
--> 227 n, d = x.shape
228 assert d == self.d
229 x = np.ascontiguousarray(x, dtype='float32')
ValueError: not enough values to unpack (expected 2, got 1)
```
### Steps to reproduce the bug
1. Load any dataset like `ds = datasets.load_dataset("wikimedia/wikipedia", "20231101.en")["train"]`
2. Add an FAISS index on any column `ds.add_faiss_index('title')`
### Expected behavior
The index should be created
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-26-generic-x86_64-with-glibc2.35
- Python version: 3.9.19
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
- `faiss-cpu` version: 1.8.0
I realized I was passing a string column to this instead of float. Is it possible to add a warning or error to prevent users from falsely believing there's a bug? | [
-0.024560395628213882,
-0.14324170351028442,
0.00876142829656601,
0.12546207010746002,
0.2912639379501343,
0.2699686586856842,
0.5877211689949036,
0.2749163508415222,
0.23181891441345215,
0.37231361865997314,
0.3932422995567322,
0.23768606781959534,
0.05539057403802872,
-0.2544329166412353... |
https://github.com/huggingface/datasets/issues/6791 | `add_faiss_index` raises ValueError: not enough values to unpack (expected 2, got 1) | Hello!
I agree that we could add some safeguards around the type of `ds[column]`. At least for FAISS, we need the column to be made of embeddings as FAISS doesn't perform the embeddings itself.
I can propose a PR sometime this week. | ### Describe the bug
Calling `add_faiss_index` on a `Dataset` with a column argument raises a ValueError. The following is the trace
```python
214 def replacement_add(self, x):
215 """Adds vectors to the index.
216 The index must be trained before vectors can be added to it.
217 The vectors are implicitly numbered in sequence. When `n` vectors are
(...)
224 `dtype` must be float32.
225 """
--> 227 n, d = x.shape
228 assert d == self.d
229 x = np.ascontiguousarray(x, dtype='float32')
ValueError: not enough values to unpack (expected 2, got 1)
```
### Steps to reproduce the bug
1. Load any dataset like `ds = datasets.load_dataset("wikimedia/wikipedia", "20231101.en")["train"]`
2. Add an FAISS index on any column `ds.add_faiss_index('title')`
### Expected behavior
The index should be created
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-26-generic-x86_64-with-glibc2.35
- Python version: 3.9.19
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
- `faiss-cpu` version: 1.8.0 | 42 | `add_faiss_index` raises ValueError: not enough values to unpack (expected 2, got 1)
### Describe the bug
Calling `add_faiss_index` on a `Dataset` with a column argument raises a ValueError. The following is the trace
```python
214 def replacement_add(self, x):
215 """Adds vectors to the index.
216 The index must be trained before vectors can be added to it.
217 The vectors are implicitly numbered in sequence. When `n` vectors are
(...)
224 `dtype` must be float32.
225 """
--> 227 n, d = x.shape
228 assert d == self.d
229 x = np.ascontiguousarray(x, dtype='float32')
ValueError: not enough values to unpack (expected 2, got 1)
```
### Steps to reproduce the bug
1. Load any dataset like `ds = datasets.load_dataset("wikimedia/wikipedia", "20231101.en")["train"]`
2. Add an FAISS index on any column `ds.add_faiss_index('title')`
### Expected behavior
The index should be created
### Environment info
- `datasets` version: 2.18.0
- Platform: Linux-6.5.0-26-generic-x86_64-with-glibc2.35
- Python version: 3.9.19
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
- `faiss-cpu` version: 1.8.0
Hello!
I agree that we could add some safeguards around the type of `ds[column]`. At least for FAISS, we need the column to be made of embeddings as FAISS doesn't perform the embeddings itself.
I can propose a PR sometime this week. | [
0.06022581085562706,
-0.1333373785018921,
0.051341500133275986,
0.2815895974636078,
0.27468815445899963,
0.2707841396331787,
0.6786123514175415,
0.29134684801101685,
0.43802809715270996,
0.4102709889411926,
0.198484405875206,
0.3571662902832031,
0.15970808267593384,
-0.052347730845212936,
... |
https://github.com/huggingface/datasets/issues/6789 | Issue with map | Default `writer_batch_size `is set to 1000 (see [map](https://huggingface.co/docs/datasets/v2.16.1/en/package_reference/main_classes#datasets.Dataset.map)).
The "tmp1335llua" is probably the temp file it creates while writing to disk.
Maybe try lowering the `writer_batch_size`.
For multi-processing you should probably pass the `processor `as an argument (with e.g. partial) to the function or create it inside so that the sub-processes have access to it and maybe add `if __name__ == "__main__"` (not sure that's necessary?).
| ### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0 | 66 | Issue with map
### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
Default `writer_batch_size `is set to 1000 (see [map](https://huggingface.co/docs/datasets/v2.16.1/en/package_reference/main_classes#datasets.Dataset.map)).
The "tmp1335llua" is probably the temp file it creates while writing to disk.
Maybe try lowering the `writer_batch_size`.
For multi-processing you should probably pass the `processor `as an argument (with e.g. partial) to the function or create it inside so that the sub-processes have access to it and maybe add `if __name__ == "__main__"` (not sure that's necessary?).
| [
-0.22016489505767822,
-0.1372167468070984,
-0.12085117399692535,
0.2797888517379761,
0.2651703357696533,
0.06365399807691574,
0.048693347722291946,
0.5038878321647644,
0.03614354506134987,
0.47659629583358765,
0.32433271408081055,
0.49251845479011536,
-0.1788398027420044,
-0.16865730285644... |
https://github.com/huggingface/datasets/issues/6789 | Issue with map | Hi @Modexus,
Thank you very much for the help! Yep after playing around with map, I managed to get the parallel processing to work by implementing it like you suggested.
Regarding the temp files, it seems like the temp files just keep growing in size as the map continues. Eventually, once map finishes, the temp files are deleted, but they are instead saved as cache .arrow files. These cache files are absolutely gigantic (~ 30-50x the size of the initial dataset!).
After playing around with the `prepare_dataset()` function above, it seems this issue is caused by the following line in the function, where the log-Mel spectrogram of the audio is calculated:
`# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"],
sampling_rate=audio["sampling_rate"]).input_features[0]
`
When I remove this line, the final cache files are approximately the same size as the initial dataset.
Can I check whether this is expected behavior with the whisper feature extractor? I cant imagine the spectrograms are that large!
Thank you so much for the help! | ### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0 | 171 | Issue with map
### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
Hi @Modexus,
Thank you very much for the help! Yep after playing around with map, I managed to get the parallel processing to work by implementing it like you suggested.
Regarding the temp files, it seems like the temp files just keep growing in size as the map continues. Eventually, once map finishes, the temp files are deleted, but they are instead saved as cache .arrow files. These cache files are absolutely gigantic (~ 30-50x the size of the initial dataset!).
After playing around with the `prepare_dataset()` function above, it seems this issue is caused by the following line in the function, where the log-Mel spectrogram of the audio is calculated:
`# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"],
sampling_rate=audio["sampling_rate"]).input_features[0]
`
When I remove this line, the final cache files are approximately the same size as the initial dataset.
Can I check whether this is expected behavior with the whisper feature extractor? I cant imagine the spectrograms are that large!
Thank you so much for the help! | [
-0.22016489505767822,
-0.1372167468070984,
-0.12085117399692535,
0.2797888517379761,
0.2651703357696533,
0.06365399807691574,
0.048693347722291946,
0.5038878321647644,
0.03614354506134987,
0.47659629583358765,
0.32433271408081055,
0.49251845479011536,
-0.1788398027420044,
-0.16865730285644... |
https://github.com/huggingface/datasets/issues/6789 | Issue with map | I'm having a similar issue with the spectrographs taking up an incredibly large amount of space. (i.e. 100GB for 3GB of audio). Is this really normal behavior? | ### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0 | 27 | Issue with map
### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
I'm having a similar issue with the spectrographs taking up an incredibly large amount of space. (i.e. 100GB for 3GB of audio). Is this really normal behavior? | [
-0.22016489505767822,
-0.1372167468070984,
-0.12085117399692535,
0.2797888517379761,
0.2651703357696533,
0.06365399807691574,
0.048693347722291946,
0.5038878321647644,
0.03614354506134987,
0.47659629583358765,
0.32433271408081055,
0.49251845479011536,
-0.1788398027420044,
-0.16865730285644... |
https://github.com/huggingface/datasets/issues/6789 | Issue with map | Upon taking a look at the hex contents of the mapped dataset files I found that the overwhelming majority of the data contained within them was duplicated junk similar to this. I'm not very familiar with the inner workings of AI but I have to assume this is an inefficient way of storing data at best and a bug at worst.
| ### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0 | 62 | Issue with map
### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
Upon taking a look at the hex contents of the mapped dataset files I found that the overwhelming majority of the data contained within them was duplicated junk similar to this. I'm not very familiar with the inner workings of AI but I have to assume this is an inefficient way of storing data at best and a bug at worst.
| [
-0.22016489505767822,
-0.1372167468070984,
-0.12085117399692535,
0.2797888517379761,
0.2651703357696533,
0.06365399807691574,
0.048693347722291946,
0.5038878321647644,
0.03614354506134987,
0.47659629583358765,
0.32433271408081055,
0.49251845479011536,
-0.1788398027420044,
-0.16865730285644... |
https://github.com/huggingface/datasets/issues/6789 | Issue with map | Same problem, dataset.map takes long time to process 12GB raw audio data and create 200GB cache file. Is there any method can run process(map) during train, instead current run
once and save cache file ? | ### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0 | 35 | Issue with map
### Describe the bug
Map has been taking extremely long to preprocess my data.
It seems to process 1000 examples (which it does really fast in about 10 seconds), then it hangs for a good 1-2 minutes, before it moves on to the next batch of 1000 examples.
It also keeps eating up my hard drive space for some reason by creating a file named tmp1335llua that is over 300GB.
Trying to set num_proc to be >1 also gives me the following error: NameError: name 'processor' is not defined
Please advise on how I could optimise this?
### Steps to reproduce the bug
In general, I have been using map as per normal. Here is a snippet of my code:
````
########################### DATASET LOADING AND PREP #########################
def load_custom_dataset(split):
ds = []
if split == 'train':
for dset in args.train_datasets:
ds.append(load_from_disk(dset))
if split == 'test':
for dset in args.test_datasets:
ds.append(load_from_disk(dset))
ds_to_return = concatenate_datasets(ds)
ds_to_return = ds_to_return.shuffle(seed=22)
return ds_to_return
def prepare_dataset(batch):
# load and (possibly) resample audio data to 16kHz
audio = batch["audio"]
# compute log-Mel input features from input audio array
batch["input_features"] = processor.feature_extractor(audio["array"], sampling_rate=audio["sampling_rate"]).input_features[0]
# compute input length of audio sample in seconds
batch["input_length"] = len(audio["array"]) / audio["sampling_rate"]
# optional pre-processing steps
transcription = batch["sentence"]
if do_lower_case:
transcription = transcription.lower()
if do_remove_punctuation:
transcription = normalizer(transcription).strip()
# encode target text to label ids
batch["labels"] = processor.tokenizer(transcription).input_ids
return batch
print('DATASET PREPARATION IN PROGRESS...')
# case 3: combine_and_shuffle is true, only train provided
# load train datasets
train_set = load_custom_dataset('train')
# split dataset
raw_dataset = DatasetDict()
raw_dataset = train_set.train_test_split(test_size = args.test_size, shuffle=True, seed=42)
raw_dataset = raw_dataset.cast_column("audio", Audio(sampling_rate=args.sampling_rate))
print("Before Map:")
print(raw_dataset)
raw_dataset = raw_dataset.map(prepare_dataset, num_proc=1)
print("After Map:")
print(raw_dataset)
````
### Expected behavior
Based on the speed at which map is processing examples, I would expect a 5-6 hours completion for all mapping
However, because it hangs every 1000 examples, I instead roughly estimate it would take about 40 hours!
Moreover, i cant even finish the map because it keeps exponentially eating up my hard drive space
### Environment info
- `datasets` version: 2.18.0
- Platform: Windows-10-10.0.22631-SP0
- Python version: 3.10.14
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.2.0
Same problem, dataset.map takes long time to process 12GB raw audio data and create 200GB cache file. Is there any method can run process(map) during train, instead current run
once and save cache file ? | [
-0.22016489505767822,
-0.1372167468070984,
-0.12085117399692535,
0.2797888517379761,
0.2651703357696533,
0.06365399807691574,
0.048693347722291946,
0.5038878321647644,
0.03614354506134987,
0.47659629583358765,
0.32433271408081055,
0.49251845479011536,
-0.1788398027420044,
-0.16865730285644... |
https://github.com/huggingface/datasets/issues/6788 | A Question About the Map Function | All data is saved in the arrow format on disk.
If you return a tensor it gets converted to arrow before saving to disk when using map.
To get a tensor when you access data elements you can use `dataset.set_format("pt")`.
Note that this just changes how the data is loaded, not how it is stored. | ### Describe the bug
Hello,
I have a question regarding the map function in the Hugging Face datasets.
The situation is as follows: when I load a jsonl file using load_dataset(..., streaming=False), and then utilize the map function to process it, I specify that the returned example should be of type Torch.tensor. However, I noticed that after applying the map function, the datatype automatically changes to List, which leads to errors in my program.
I attempted to use load_dataset(..., streaming=True), and this issue no longer occurs. I'm not entirely clear on why this happens. Could you please provide some insights into this?
### Steps to reproduce the bug
1.dataset = load_dataset(xxx, streaming = False)
2. dataset.map(function), function will return torch.Tensor.
3. you will find the format of data in dataset is List.
### Expected behavior
I expected to receieve the format of data is torch.Tensor.
### Environment info
2.18.0 | 55 | A Question About the Map Function
### Describe the bug
Hello,
I have a question regarding the map function in the Hugging Face datasets.
The situation is as follows: when I load a jsonl file using load_dataset(..., streaming=False), and then utilize the map function to process it, I specify that the returned example should be of type Torch.tensor. However, I noticed that after applying the map function, the datatype automatically changes to List, which leads to errors in my program.
I attempted to use load_dataset(..., streaming=True), and this issue no longer occurs. I'm not entirely clear on why this happens. Could you please provide some insights into this?
### Steps to reproduce the bug
1.dataset = load_dataset(xxx, streaming = False)
2. dataset.map(function), function will return torch.Tensor.
3. you will find the format of data in dataset is List.
### Expected behavior
I expected to receieve the format of data is torch.Tensor.
### Environment info
2.18.0
All data is saved in the arrow format on disk.
If you return a tensor it gets converted to arrow before saving to disk when using map.
To get a tensor when you access data elements you can use `dataset.set_format("pt")`.
Note that this just changes how the data is loaded, not how it is stored. | [
-0.07809469848871231,
-0.5261644721031189,
-0.015050448477268219,
0.2572656571865082,
0.17012904584407806,
0.15750515460968018,
0.288708359003067,
0.19567933678627014,
0.6957500576972961,
0.004863850772380829,
0.029340626671910286,
0.5183698534965515,
-0.04254040867090225,
-0.0635768175125... |
https://github.com/huggingface/datasets/issues/6788 | A Question About the Map Function | > All data is saved in the arrow format on disk. If you return a tensor it gets converted to arrow before saving to disk when using map.
>
> To get a tensor when you access data elements you can use `dataset.set_format("pt")`. Note that this just changes how the data is loaded, not how it is stored.
Thank you very much for your explanation, I understand what you mean now. So you're saying that when streaming=True, there's no need to convert it to the arrow format and save it to disk. But if we directly load all formats and then convert them into the arrow format after passing through the map function, it will convert torch.Tensor into a List. I see. | ### Describe the bug
Hello,
I have a question regarding the map function in the Hugging Face datasets.
The situation is as follows: when I load a jsonl file using load_dataset(..., streaming=False), and then utilize the map function to process it, I specify that the returned example should be of type Torch.tensor. However, I noticed that after applying the map function, the datatype automatically changes to List, which leads to errors in my program.
I attempted to use load_dataset(..., streaming=True), and this issue no longer occurs. I'm not entirely clear on why this happens. Could you please provide some insights into this?
### Steps to reproduce the bug
1.dataset = load_dataset(xxx, streaming = False)
2. dataset.map(function), function will return torch.Tensor.
3. you will find the format of data in dataset is List.
### Expected behavior
I expected to receieve the format of data is torch.Tensor.
### Environment info
2.18.0 | 122 | A Question About the Map Function
### Describe the bug
Hello,
I have a question regarding the map function in the Hugging Face datasets.
The situation is as follows: when I load a jsonl file using load_dataset(..., streaming=False), and then utilize the map function to process it, I specify that the returned example should be of type Torch.tensor. However, I noticed that after applying the map function, the datatype automatically changes to List, which leads to errors in my program.
I attempted to use load_dataset(..., streaming=True), and this issue no longer occurs. I'm not entirely clear on why this happens. Could you please provide some insights into this?
### Steps to reproduce the bug
1.dataset = load_dataset(xxx, streaming = False)
2. dataset.map(function), function will return torch.Tensor.
3. you will find the format of data in dataset is List.
### Expected behavior
I expected to receieve the format of data is torch.Tensor.
### Environment info
2.18.0
> All data is saved in the arrow format on disk. If you return a tensor it gets converted to arrow before saving to disk when using map.
>
> To get a tensor when you access data elements you can use `dataset.set_format("pt")`. Note that this just changes how the data is loaded, not how it is stored.
Thank you very much for your explanation, I understand what you mean now. So you're saying that when streaming=True, there's no need to convert it to the arrow format and save it to disk. But if we directly load all formats and then convert them into the arrow format after passing through the map function, it will convert torch.Tensor into a List. I see. | [
-0.09028027206659317,
-0.5382764339447021,
-0.02304200269281864,
0.21255333721637726,
0.1866646707057953,
0.12995660305023193,
0.27998730540275574,
0.1794217824935913,
0.6446641683578491,
0.0019021108746528625,
0.01215516310185194,
0.4794826805591583,
-0.048781126737594604,
-0.055350478738... |
https://github.com/huggingface/datasets/issues/6787 | TimeoutError in map | From my current understanding, this timeout is only used when we need to get the results.
One of:
1. All tasks are done
2. One worker died
Your function should work fine and it's definitely a bug if it doesn't. | ### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version | 40 | TimeoutError in map
### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version
From my current understanding, this timeout is only used when we need to get the results.
One of:
1. All tasks are done
2. One worker died
Your function should work fine and it's definitely a bug if it doesn't. | [
-0.08302028477191925,
-0.3202267289161682,
-0.12161970138549805,
-0.14132612943649292,
-0.08951155841350555,
-0.1312093585729599,
0.2986030876636505,
0.05766239017248154,
0.26414912939071655,
0.21171939373016357,
0.671402096748352,
0.30782267451286316,
0.020487602800130844,
-0.191791042685... |
https://github.com/huggingface/datasets/issues/6787 | TimeoutError in map | When one of the `map`'s worker processes crashes, the linked code re-raises an error from the crash and returns it to the caller.
If your question is how to limit the time of long-running tasks/worker processes, such functionality doesn't exist in `datasets` (yet), which means you need to implement it yourself.
E.g., you can implement it using the built-in `signal` module like this:
```python
import time
import signal
from contextlib import contextmanager
from datasets import Dataset
@contextmanager
def max_exec_time(t):
def raise_timeout_handler(signum, frame):
raise TimeoutError
orig_handler = signal.getsignal(signal.SIGALRM)
signal.signal(signal.SIGALRM, raise_timeout_handler)
try:
signal.alarm(t)
yield
finally:
signal.alarm(0)
signal.signal(signal.SIGALRM, orig_handler)
def worker(example, rank):
try:
with max_exec_time(20): # 20 sec execution limit
if rank % 2 == 0:
time.sleep(50) # simulate a long-running task
example["a"] = 100
except TimeoutError:
example["a"] = None # Or return empty batches here in the "batched" mode
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker, num_proc=2, with_rank=True)
print(data[0])
``` | ### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version | 152 | TimeoutError in map
### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version
When one of the `map`'s worker processes crashes, the linked code re-raises an error from the crash and returns it to the caller.
If your question is how to limit the time of long-running tasks/worker processes, such functionality doesn't exist in `datasets` (yet), which means you need to implement it yourself.
E.g., you can implement it using the built-in `signal` module like this:
```python
import time
import signal
from contextlib import contextmanager
from datasets import Dataset
@contextmanager
def max_exec_time(t):
def raise_timeout_handler(signum, frame):
raise TimeoutError
orig_handler = signal.getsignal(signal.SIGALRM)
signal.signal(signal.SIGALRM, raise_timeout_handler)
try:
signal.alarm(t)
yield
finally:
signal.alarm(0)
signal.signal(signal.SIGALRM, orig_handler)
def worker(example, rank):
try:
with max_exec_time(20): # 20 sec execution limit
if rank % 2 == 0:
time.sleep(50) # simulate a long-running task
example["a"] = 100
except TimeoutError:
example["a"] = None # Or return empty batches here in the "batched" mode
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker, num_proc=2, with_rank=True)
print(data[0])
``` | [
-0.4587858021259308,
-0.22682136297225952,
-0.1437147706747055,
-0.18995685875415802,
-0.07908246666193008,
-0.12760818004608154,
0.34103143215179443,
0.14496548473834991,
0.329811155796051,
0.21523816883563995,
0.48750123381614685,
0.39861777424812317,
-0.20390483736991882,
-0.08528787642... |
https://github.com/huggingface/datasets/issues/6787 | TimeoutError in map | > From my current understanding, this timeout is only used when we need to get the results.
>
> One of:
>
> 1. All tasks are done
> 2. One worker died
>
> Your function should work fine and it's definitely a bug if it doesn't.
thanks for responding! can you reproduce the stuck with the above example code? | ### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version | 61 | TimeoutError in map
### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version
> From my current understanding, this timeout is only used when we need to get the results.
>
> One of:
>
> 1. All tasks are done
> 2. One worker died
>
> Your function should work fine and it's definitely a bug if it doesn't.
thanks for responding! can you reproduce the stuck with the above example code? | [
-0.05328620970249176,
-0.32660120725631714,
-0.1180211752653122,
-0.11633157730102539,
-0.06923969835042953,
-0.09220893681049347,
0.2910594046115875,
0.06648428738117218,
0.2722051441669464,
0.21976447105407715,
0.6478757262229919,
0.3284038007259369,
0.0023593828082084656,
-0.15714636445... |
https://github.com/huggingface/datasets/issues/6787 | TimeoutError in map | > When one of the `map`'s worker processes crashes, the linked code re-raises an error from the crash and returns it to the caller.
>
> If your question is how to limit the time of long-running tasks/worker processes, such functionality doesn't exist in `datasets` (yet), which means you need to implement it yourself.
>
> E.g., you can implement it using the built-in `signal` module like this:
>
> ```python
> import time
> import signal
> from contextlib import contextmanager
>
> from datasets import Dataset
>
>
> @contextmanager
> def max_exec_time(t):
> def raise_timeout_handler(signum, frame):
> raise TimeoutError
>
> orig_handler = signal.getsignal(signal.SIGALRM)
> signal.signal(signal.SIGALRM, raise_timeout_handler)
> try:
> signal.alarm(t)
> yield
> finally:
> signal.alarm(0)
> signal.signal(signal.SIGALRM, orig_handler)
>
>
> def worker(example, rank):
> try:
> with max_exec_time(20): # 20 sec execution limit
> if rank % 2 == 0:
> time.sleep(50) # simulate a long-running task
> example["a"] = 100
> except TimeoutError:
> example["a"] = None # Or return empty batches here in the "batched" mode
> return example
>
> data = Dataset.from_list([{"a": 1}, {"a": 2}])
> data = data.map(worker, num_proc=2, with_rank=True)
> print(data[0])
> ```
thanks for responding! However, I don't think we should use `signal` in the context of multiprocessing since sometimes it will crash one process and raise the following error
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L664 | ### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version | 224 | TimeoutError in map
### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version
> When one of the `map`'s worker processes crashes, the linked code re-raises an error from the crash and returns it to the caller.
>
> If your question is how to limit the time of long-running tasks/worker processes, such functionality doesn't exist in `datasets` (yet), which means you need to implement it yourself.
>
> E.g., you can implement it using the built-in `signal` module like this:
>
> ```python
> import time
> import signal
> from contextlib import contextmanager
>
> from datasets import Dataset
>
>
> @contextmanager
> def max_exec_time(t):
> def raise_timeout_handler(signum, frame):
> raise TimeoutError
>
> orig_handler = signal.getsignal(signal.SIGALRM)
> signal.signal(signal.SIGALRM, raise_timeout_handler)
> try:
> signal.alarm(t)
> yield
> finally:
> signal.alarm(0)
> signal.signal(signal.SIGALRM, orig_handler)
>
>
> def worker(example, rank):
> try:
> with max_exec_time(20): # 20 sec execution limit
> if rank % 2 == 0:
> time.sleep(50) # simulate a long-running task
> example["a"] = 100
> except TimeoutError:
> example["a"] = None # Or return empty batches here in the "batched" mode
> return example
>
> data = Dataset.from_list([{"a": 1}, {"a": 2}])
> data = data.map(worker, num_proc=2, with_rank=True)
> print(data[0])
> ```
thanks for responding! However, I don't think we should use `signal` in the context of multiprocessing since sometimes it will crash one process and raise the following error
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L664 | [
-0.41645127534866333,
-0.24910253286361694,
-0.13754624128341675,
-0.16449712216854095,
-0.06373206526041031,
-0.1262134611606598,
0.3471662402153015,
0.15491218864917755,
0.32826536893844604,
0.21145561337471008,
0.4979971647262573,
0.4093160927295685,
-0.17273324728012085,
-0.10743519663... |
https://github.com/huggingface/datasets/issues/6787 | TimeoutError in map | > thanks for responding! However, I don't think we should use signal in the context of multiprocessing since sometimes it will crash one process and raise the following error
The above code has `try/except` to catch the error from the handler. Or do you get an error other than `TimeoutError`? | ### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version | 50 | TimeoutError in map
### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version
> thanks for responding! However, I don't think we should use signal in the context of multiprocessing since sometimes it will crash one process and raise the following error
The above code has `try/except` to catch the error from the handler. Or do you get an error other than `TimeoutError`? | [
-0.3375905156135559,
-0.3138829469680786,
-0.09935571253299713,
-0.12534309923648834,
-0.15953339636325836,
-0.12933114171028137,
0.42860448360443115,
-0.010126423090696335,
0.24521991610527039,
0.2920992970466614,
0.506821870803833,
0.37501272559165955,
-0.16985999047756195,
-0.1655199974... |
https://github.com/huggingface/datasets/issues/6787 | TimeoutError in map | > > thanks for responding! However, I don't think we should use signal in the context of multiprocessing since sometimes it will crash one process and raise the following error
>
> The above code has `try/except` to catch the error from the handler. Or do you get an error other than `TimeoutError`?
yup, it will raise the RuntimeError: https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L667C19-L670C22
```
raise RuntimeError(
"One of the subprocesses has abruptly died during map operation."
"To debug the error, disable multiprocessing."
)
``` | ### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version | 81 | TimeoutError in map
### Describe the bug
```python
from datasets import Dataset
def worker(example):
while True:
continue
example['a'] = 100
return example
data = Dataset.from_list([{"a": 1}, {"a": 2}])
data = data.map(worker)
print(data[0])
```
I'm implementing a worker function whose runtime will depend on specific examples (e.g., while most examples take 0.01s in worker, several examples may take 50s).
Therefore, I would like to know how the current implementation will handle those subprocesses that require a long (e.g., >= 5min) or even infinite time.
I notice that the current implementation set a timeout of 0.05 second
https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L674
However, this example code still gets stuck.
### Steps to reproduce the bug
run the example above
### Expected behavior
I want to set a default worker to handle these timeout cases, instead of getting stuck
### Environment info
main branch version
> > thanks for responding! However, I don't think we should use signal in the context of multiprocessing since sometimes it will crash one process and raise the following error
>
> The above code has `try/except` to catch the error from the handler. Or do you get an error other than `TimeoutError`?
yup, it will raise the RuntimeError: https://github.com/huggingface/datasets/blob/c3ddb1ef00334a6f973679a51e783905fbc9ef0b/src/datasets/utils/py_utils.py#L667C19-L670C22
```
raise RuntimeError(
"One of the subprocesses has abruptly died during map operation."
"To debug the error, disable multiprocessing."
)
``` | [
-0.34850868582725525,
-0.31877267360687256,
-0.0801568329334259,
-0.10421864688396454,
-0.13138437271118164,
-0.15662577748298645,
0.3912196457386017,
-0.020958110690116882,
0.14485031366348267,
0.3016989827156067,
0.496257483959198,
0.3947380483150482,
-0.14380799233913422,
-0.14730437099... |
https://github.com/huggingface/datasets/issues/6783 | AttributeError: module 'numpy' has no attribute 'object'. in Kaggle Notebook | Hi! You can fix this by updating the `datasets` package with `pip install -U datasets` and restarting the notebook.
| ### Describe the bug
# problem
I can't resample audio dataset in Kaggle Notebook. It looks like some code in `datasets` library use aliases that were deprecated in NumPy 1.20.
## code for resampling
```
from datasets import load_dataset, Audio
from transformers import AutoFeatureExtractor
from transformers import AutoModelForAudioClassification, TrainingArguments, Trainer
minds = load_dataset("PolyAI/minds14", name="en-US", split="train")
feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/wav2vec2-base")
def preprocess_function(examples):
audio_arrays = [x["array"] for x in examples["audio"]]
inputs = feature_extractor(
audio_arrays, sampling_rate=feature_extractor.sampling_rate, max_length=16000, truncation=True
)
return inputs
dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
```
## the error I got
<details>
<summary>Click to expand</summary>
```
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
Cell In[20], line 1
----> 1 dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
2 dataset
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:1955, in Dataset.map(self, function, with_indices, with_rank, 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)
1952 disable_tqdm = not logging.is_progress_bar_enabled()
1954 if num_proc is None or num_proc == 1:
-> 1955 return self._map_single(
1956 function=function,
1957 with_indices=with_indices,
1958 with_rank=with_rank,
1959 input_columns=input_columns,
1960 batched=batched,
1961 batch_size=batch_size,
1962 drop_last_batch=drop_last_batch,
1963 remove_columns=remove_columns,
1964 keep_in_memory=keep_in_memory,
1965 load_from_cache_file=load_from_cache_file,
1966 cache_file_name=cache_file_name,
1967 writer_batch_size=writer_batch_size,
1968 features=features,
1969 disable_nullable=disable_nullable,
1970 fn_kwargs=fn_kwargs,
1971 new_fingerprint=new_fingerprint,
1972 disable_tqdm=disable_tqdm,
1973 desc=desc,
1974 )
1975 else:
1977 def format_cache_file_name(cache_file_name, rank):
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:520, in transmit_tasks.<locals>.wrapper(*args, **kwargs)
518 self: "Dataset" = kwargs.pop("self")
519 # apply actual function
--> 520 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
521 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
522 for dataset in datasets:
523 # Remove task templates if a column mapping of the template is no longer valid
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:487, in transmit_format.<locals>.wrapper(*args, **kwargs)
480 self_format = {
481 "type": self._format_type,
482 "format_kwargs": self._format_kwargs,
483 "columns": self._format_columns,
484 "output_all_columns": self._output_all_columns,
485 }
486 # apply actual function
--> 487 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
488 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
489 # re-apply format to the output
File /opt/conda/lib/python3.10/site-packages/datasets/fingerprint.py:458, in fingerprint_transform.<locals>._fingerprint.<locals>.wrapper(*args, **kwargs)
452 kwargs[fingerprint_name] = update_fingerprint(
453 self._fingerprint, transform, kwargs_for_fingerprint
454 )
456 # Call actual function
--> 458 out = func(self, *args, **kwargs)
460 # Update fingerprint of in-place transforms + update in-place history of transforms
462 if inplace: # update after calling func so that the fingerprint doesn't change if the function fails
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:2356, in Dataset._map_single(self, function, with_indices, with_rank, 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)
2354 writer.write_table(batch)
2355 else:
-> 2356 writer.write_batch(batch)
2357 if update_data and writer is not None:
2358 writer.finalize() # close_stream=bool(buf_writer is None)) # We only close if we are writing in a file
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:507, in ArrowWriter.write_batch(self, batch_examples, writer_batch_size)
505 col_try_type = try_features[col] if try_features is not None and col in try_features else None
506 typed_sequence = OptimizedTypedSequence(batch_examples[col], type=col_type, try_type=col_try_type, col=col)
--> 507 arrays.append(pa.array(typed_sequence))
508 inferred_features[col] = typed_sequence.get_inferred_type()
509 schema = inferred_features.arrow_schema if self.pa_writer is None else self.schema
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:236, in pyarrow.lib.array()
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:110, in pyarrow.lib._handle_arrow_array_protocol()
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:184, in TypedSequence.__arrow_array__(self, type)
182 out = numpy_to_pyarrow_listarray(data)
183 elif isinstance(data, list) and data and isinstance(first_non_null_value(data)[1], np.ndarray):
--> 184 out = list_of_np_array_to_pyarrow_listarray(data)
185 else:
186 trying_cast_to_python_objects = True
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1174, in list_of_np_array_to_pyarrow_listarray(l_arr, type)
1172 """Build a PyArrow ListArray from a possibly nested list of NumPy arrays"""
1173 if len(l_arr) > 0:
-> 1174 return list_of_pa_arrays_to_pyarrow_listarray(
1175 [numpy_to_pyarrow_listarray(arr, type=type) if arr is not None else None for arr in l_arr]
1176 )
1177 else:
1178 return pa.array([], type=type)
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1163, in list_of_pa_arrays_to_pyarrow_listarray(l_arr)
1160 null_indices = [i for i, arr in enumerate(l_arr) if arr is None]
1161 l_arr = [arr for arr in l_arr if arr is not None]
1162 offsets = np.cumsum(
-> 1163 [0] + [len(arr) for arr in l_arr], dtype=np.object
1164 ) # convert to dtype object to allow None insertion
1165 offsets = np.insert(offsets, null_indices, None)
1166 offsets = pa.array(offsets, type=pa.int32())
File /opt/conda/lib/python3.10/site-packages/numpy/__init__.py:324, in __getattr__(attr)
319 warnings.warn(
320 f"In the future `np.{attr}` will be defined as the "
321 "corresponding NumPy scalar.", FutureWarning, stacklevel=2)
323 if attr in __former_attrs__:
--> 324 raise AttributeError(__former_attrs__[attr])
326 if attr == 'testing':
327 import numpy.testing as testing
AttributeError: module 'numpy' has no attribute 'object'.
`np.object` was a deprecated alias for the builtin `object`. To avoid this error in existing code, use `object` by itself. Doing this will not modify any behavior and is safe.
The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at:
https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
```
</details>
### Steps to reproduce the bug
Run above code in Kaggle Notebook.
### Expected behavior
I can resample audio data without fail.
### Environment info
- `datasets` version: 2.1.0
- Platform: Linux-5.15.133+-x86_64-with-glibc2.31
- Python version: 3.10.13
- PyArrow version: 11.0.0
- Pandas version: 2.2.1 | 19 | AttributeError: module 'numpy' has no attribute 'object'. in Kaggle Notebook
### Describe the bug
# problem
I can't resample audio dataset in Kaggle Notebook. It looks like some code in `datasets` library use aliases that were deprecated in NumPy 1.20.
## code for resampling
```
from datasets import load_dataset, Audio
from transformers import AutoFeatureExtractor
from transformers import AutoModelForAudioClassification, TrainingArguments, Trainer
minds = load_dataset("PolyAI/minds14", name="en-US", split="train")
feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/wav2vec2-base")
def preprocess_function(examples):
audio_arrays = [x["array"] for x in examples["audio"]]
inputs = feature_extractor(
audio_arrays, sampling_rate=feature_extractor.sampling_rate, max_length=16000, truncation=True
)
return inputs
dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
```
## the error I got
<details>
<summary>Click to expand</summary>
```
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
Cell In[20], line 1
----> 1 dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
2 dataset
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:1955, in Dataset.map(self, function, with_indices, with_rank, 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)
1952 disable_tqdm = not logging.is_progress_bar_enabled()
1954 if num_proc is None or num_proc == 1:
-> 1955 return self._map_single(
1956 function=function,
1957 with_indices=with_indices,
1958 with_rank=with_rank,
1959 input_columns=input_columns,
1960 batched=batched,
1961 batch_size=batch_size,
1962 drop_last_batch=drop_last_batch,
1963 remove_columns=remove_columns,
1964 keep_in_memory=keep_in_memory,
1965 load_from_cache_file=load_from_cache_file,
1966 cache_file_name=cache_file_name,
1967 writer_batch_size=writer_batch_size,
1968 features=features,
1969 disable_nullable=disable_nullable,
1970 fn_kwargs=fn_kwargs,
1971 new_fingerprint=new_fingerprint,
1972 disable_tqdm=disable_tqdm,
1973 desc=desc,
1974 )
1975 else:
1977 def format_cache_file_name(cache_file_name, rank):
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:520, in transmit_tasks.<locals>.wrapper(*args, **kwargs)
518 self: "Dataset" = kwargs.pop("self")
519 # apply actual function
--> 520 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
521 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
522 for dataset in datasets:
523 # Remove task templates if a column mapping of the template is no longer valid
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:487, in transmit_format.<locals>.wrapper(*args, **kwargs)
480 self_format = {
481 "type": self._format_type,
482 "format_kwargs": self._format_kwargs,
483 "columns": self._format_columns,
484 "output_all_columns": self._output_all_columns,
485 }
486 # apply actual function
--> 487 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
488 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
489 # re-apply format to the output
File /opt/conda/lib/python3.10/site-packages/datasets/fingerprint.py:458, in fingerprint_transform.<locals>._fingerprint.<locals>.wrapper(*args, **kwargs)
452 kwargs[fingerprint_name] = update_fingerprint(
453 self._fingerprint, transform, kwargs_for_fingerprint
454 )
456 # Call actual function
--> 458 out = func(self, *args, **kwargs)
460 # Update fingerprint of in-place transforms + update in-place history of transforms
462 if inplace: # update after calling func so that the fingerprint doesn't change if the function fails
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:2356, in Dataset._map_single(self, function, with_indices, with_rank, 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)
2354 writer.write_table(batch)
2355 else:
-> 2356 writer.write_batch(batch)
2357 if update_data and writer is not None:
2358 writer.finalize() # close_stream=bool(buf_writer is None)) # We only close if we are writing in a file
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:507, in ArrowWriter.write_batch(self, batch_examples, writer_batch_size)
505 col_try_type = try_features[col] if try_features is not None and col in try_features else None
506 typed_sequence = OptimizedTypedSequence(batch_examples[col], type=col_type, try_type=col_try_type, col=col)
--> 507 arrays.append(pa.array(typed_sequence))
508 inferred_features[col] = typed_sequence.get_inferred_type()
509 schema = inferred_features.arrow_schema if self.pa_writer is None else self.schema
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:236, in pyarrow.lib.array()
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:110, in pyarrow.lib._handle_arrow_array_protocol()
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:184, in TypedSequence.__arrow_array__(self, type)
182 out = numpy_to_pyarrow_listarray(data)
183 elif isinstance(data, list) and data and isinstance(first_non_null_value(data)[1], np.ndarray):
--> 184 out = list_of_np_array_to_pyarrow_listarray(data)
185 else:
186 trying_cast_to_python_objects = True
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1174, in list_of_np_array_to_pyarrow_listarray(l_arr, type)
1172 """Build a PyArrow ListArray from a possibly nested list of NumPy arrays"""
1173 if len(l_arr) > 0:
-> 1174 return list_of_pa_arrays_to_pyarrow_listarray(
1175 [numpy_to_pyarrow_listarray(arr, type=type) if arr is not None else None for arr in l_arr]
1176 )
1177 else:
1178 return pa.array([], type=type)
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1163, in list_of_pa_arrays_to_pyarrow_listarray(l_arr)
1160 null_indices = [i for i, arr in enumerate(l_arr) if arr is None]
1161 l_arr = [arr for arr in l_arr if arr is not None]
1162 offsets = np.cumsum(
-> 1163 [0] + [len(arr) for arr in l_arr], dtype=np.object
1164 ) # convert to dtype object to allow None insertion
1165 offsets = np.insert(offsets, null_indices, None)
1166 offsets = pa.array(offsets, type=pa.int32())
File /opt/conda/lib/python3.10/site-packages/numpy/__init__.py:324, in __getattr__(attr)
319 warnings.warn(
320 f"In the future `np.{attr}` will be defined as the "
321 "corresponding NumPy scalar.", FutureWarning, stacklevel=2)
323 if attr in __former_attrs__:
--> 324 raise AttributeError(__former_attrs__[attr])
326 if attr == 'testing':
327 import numpy.testing as testing
AttributeError: module 'numpy' has no attribute 'object'.
`np.object` was a deprecated alias for the builtin `object`. To avoid this error in existing code, use `object` by itself. Doing this will not modify any behavior and is safe.
The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at:
https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
```
</details>
### Steps to reproduce the bug
Run above code in Kaggle Notebook.
### Expected behavior
I can resample audio data without fail.
### Environment info
- `datasets` version: 2.1.0
- Platform: Linux-5.15.133+-x86_64-with-glibc2.31
- Python version: 3.10.13
- PyArrow version: 11.0.0
- Pandas version: 2.2.1
Hi! You can fix this by updating the `datasets` package with `pip install -U datasets` and restarting the notebook.
| [
-0.22040624916553497,
0.02457261085510254,
-0.03590615093708038,
0.00249418243765831,
0.4256819486618042,
-0.05877440422773361,
0.38796424865722656,
0.31887146830558777,
-0.18779487907886505,
0.3352118730545044,
-0.2983803451061249,
0.683189332485199,
-0.2616359293460846,
0.252659529447555... |
https://github.com/huggingface/datasets/issues/6783 | AttributeError: module 'numpy' has no attribute 'object'. in Kaggle Notebook | Kaggle removed the problematic `datasets==2.1.0` pin last week, so I'm closing this issue (now it pre-installs the latest version). | ### Describe the bug
# problem
I can't resample audio dataset in Kaggle Notebook. It looks like some code in `datasets` library use aliases that were deprecated in NumPy 1.20.
## code for resampling
```
from datasets import load_dataset, Audio
from transformers import AutoFeatureExtractor
from transformers import AutoModelForAudioClassification, TrainingArguments, Trainer
minds = load_dataset("PolyAI/minds14", name="en-US", split="train")
feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/wav2vec2-base")
def preprocess_function(examples):
audio_arrays = [x["array"] for x in examples["audio"]]
inputs = feature_extractor(
audio_arrays, sampling_rate=feature_extractor.sampling_rate, max_length=16000, truncation=True
)
return inputs
dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
```
## the error I got
<details>
<summary>Click to expand</summary>
```
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
Cell In[20], line 1
----> 1 dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
2 dataset
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:1955, in Dataset.map(self, function, with_indices, with_rank, 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)
1952 disable_tqdm = not logging.is_progress_bar_enabled()
1954 if num_proc is None or num_proc == 1:
-> 1955 return self._map_single(
1956 function=function,
1957 with_indices=with_indices,
1958 with_rank=with_rank,
1959 input_columns=input_columns,
1960 batched=batched,
1961 batch_size=batch_size,
1962 drop_last_batch=drop_last_batch,
1963 remove_columns=remove_columns,
1964 keep_in_memory=keep_in_memory,
1965 load_from_cache_file=load_from_cache_file,
1966 cache_file_name=cache_file_name,
1967 writer_batch_size=writer_batch_size,
1968 features=features,
1969 disable_nullable=disable_nullable,
1970 fn_kwargs=fn_kwargs,
1971 new_fingerprint=new_fingerprint,
1972 disable_tqdm=disable_tqdm,
1973 desc=desc,
1974 )
1975 else:
1977 def format_cache_file_name(cache_file_name, rank):
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:520, in transmit_tasks.<locals>.wrapper(*args, **kwargs)
518 self: "Dataset" = kwargs.pop("self")
519 # apply actual function
--> 520 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
521 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
522 for dataset in datasets:
523 # Remove task templates if a column mapping of the template is no longer valid
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:487, in transmit_format.<locals>.wrapper(*args, **kwargs)
480 self_format = {
481 "type": self._format_type,
482 "format_kwargs": self._format_kwargs,
483 "columns": self._format_columns,
484 "output_all_columns": self._output_all_columns,
485 }
486 # apply actual function
--> 487 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
488 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
489 # re-apply format to the output
File /opt/conda/lib/python3.10/site-packages/datasets/fingerprint.py:458, in fingerprint_transform.<locals>._fingerprint.<locals>.wrapper(*args, **kwargs)
452 kwargs[fingerprint_name] = update_fingerprint(
453 self._fingerprint, transform, kwargs_for_fingerprint
454 )
456 # Call actual function
--> 458 out = func(self, *args, **kwargs)
460 # Update fingerprint of in-place transforms + update in-place history of transforms
462 if inplace: # update after calling func so that the fingerprint doesn't change if the function fails
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:2356, in Dataset._map_single(self, function, with_indices, with_rank, 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)
2354 writer.write_table(batch)
2355 else:
-> 2356 writer.write_batch(batch)
2357 if update_data and writer is not None:
2358 writer.finalize() # close_stream=bool(buf_writer is None)) # We only close if we are writing in a file
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:507, in ArrowWriter.write_batch(self, batch_examples, writer_batch_size)
505 col_try_type = try_features[col] if try_features is not None and col in try_features else None
506 typed_sequence = OptimizedTypedSequence(batch_examples[col], type=col_type, try_type=col_try_type, col=col)
--> 507 arrays.append(pa.array(typed_sequence))
508 inferred_features[col] = typed_sequence.get_inferred_type()
509 schema = inferred_features.arrow_schema if self.pa_writer is None else self.schema
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:236, in pyarrow.lib.array()
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:110, in pyarrow.lib._handle_arrow_array_protocol()
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:184, in TypedSequence.__arrow_array__(self, type)
182 out = numpy_to_pyarrow_listarray(data)
183 elif isinstance(data, list) and data and isinstance(first_non_null_value(data)[1], np.ndarray):
--> 184 out = list_of_np_array_to_pyarrow_listarray(data)
185 else:
186 trying_cast_to_python_objects = True
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1174, in list_of_np_array_to_pyarrow_listarray(l_arr, type)
1172 """Build a PyArrow ListArray from a possibly nested list of NumPy arrays"""
1173 if len(l_arr) > 0:
-> 1174 return list_of_pa_arrays_to_pyarrow_listarray(
1175 [numpy_to_pyarrow_listarray(arr, type=type) if arr is not None else None for arr in l_arr]
1176 )
1177 else:
1178 return pa.array([], type=type)
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1163, in list_of_pa_arrays_to_pyarrow_listarray(l_arr)
1160 null_indices = [i for i, arr in enumerate(l_arr) if arr is None]
1161 l_arr = [arr for arr in l_arr if arr is not None]
1162 offsets = np.cumsum(
-> 1163 [0] + [len(arr) for arr in l_arr], dtype=np.object
1164 ) # convert to dtype object to allow None insertion
1165 offsets = np.insert(offsets, null_indices, None)
1166 offsets = pa.array(offsets, type=pa.int32())
File /opt/conda/lib/python3.10/site-packages/numpy/__init__.py:324, in __getattr__(attr)
319 warnings.warn(
320 f"In the future `np.{attr}` will be defined as the "
321 "corresponding NumPy scalar.", FutureWarning, stacklevel=2)
323 if attr in __former_attrs__:
--> 324 raise AttributeError(__former_attrs__[attr])
326 if attr == 'testing':
327 import numpy.testing as testing
AttributeError: module 'numpy' has no attribute 'object'.
`np.object` was a deprecated alias for the builtin `object`. To avoid this error in existing code, use `object` by itself. Doing this will not modify any behavior and is safe.
The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at:
https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
```
</details>
### Steps to reproduce the bug
Run above code in Kaggle Notebook.
### Expected behavior
I can resample audio data without fail.
### Environment info
- `datasets` version: 2.1.0
- Platform: Linux-5.15.133+-x86_64-with-glibc2.31
- Python version: 3.10.13
- PyArrow version: 11.0.0
- Pandas version: 2.2.1 | 19 | AttributeError: module 'numpy' has no attribute 'object'. in Kaggle Notebook
### Describe the bug
# problem
I can't resample audio dataset in Kaggle Notebook. It looks like some code in `datasets` library use aliases that were deprecated in NumPy 1.20.
## code for resampling
```
from datasets import load_dataset, Audio
from transformers import AutoFeatureExtractor
from transformers import AutoModelForAudioClassification, TrainingArguments, Trainer
minds = load_dataset("PolyAI/minds14", name="en-US", split="train")
feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/wav2vec2-base")
def preprocess_function(examples):
audio_arrays = [x["array"] for x in examples["audio"]]
inputs = feature_extractor(
audio_arrays, sampling_rate=feature_extractor.sampling_rate, max_length=16000, truncation=True
)
return inputs
dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
```
## the error I got
<details>
<summary>Click to expand</summary>
```
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
Cell In[20], line 1
----> 1 dataset = dataset.map(preprocess_function, remove_columns="audio", batched=True, batch_size=100)
2 dataset
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:1955, in Dataset.map(self, function, with_indices, with_rank, 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)
1952 disable_tqdm = not logging.is_progress_bar_enabled()
1954 if num_proc is None or num_proc == 1:
-> 1955 return self._map_single(
1956 function=function,
1957 with_indices=with_indices,
1958 with_rank=with_rank,
1959 input_columns=input_columns,
1960 batched=batched,
1961 batch_size=batch_size,
1962 drop_last_batch=drop_last_batch,
1963 remove_columns=remove_columns,
1964 keep_in_memory=keep_in_memory,
1965 load_from_cache_file=load_from_cache_file,
1966 cache_file_name=cache_file_name,
1967 writer_batch_size=writer_batch_size,
1968 features=features,
1969 disable_nullable=disable_nullable,
1970 fn_kwargs=fn_kwargs,
1971 new_fingerprint=new_fingerprint,
1972 disable_tqdm=disable_tqdm,
1973 desc=desc,
1974 )
1975 else:
1977 def format_cache_file_name(cache_file_name, rank):
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:520, in transmit_tasks.<locals>.wrapper(*args, **kwargs)
518 self: "Dataset" = kwargs.pop("self")
519 # apply actual function
--> 520 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
521 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
522 for dataset in datasets:
523 # Remove task templates if a column mapping of the template is no longer valid
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:487, in transmit_format.<locals>.wrapper(*args, **kwargs)
480 self_format = {
481 "type": self._format_type,
482 "format_kwargs": self._format_kwargs,
483 "columns": self._format_columns,
484 "output_all_columns": self._output_all_columns,
485 }
486 # apply actual function
--> 487 out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
488 datasets: List["Dataset"] = list(out.values()) if isinstance(out, dict) else [out]
489 # re-apply format to the output
File /opt/conda/lib/python3.10/site-packages/datasets/fingerprint.py:458, in fingerprint_transform.<locals>._fingerprint.<locals>.wrapper(*args, **kwargs)
452 kwargs[fingerprint_name] = update_fingerprint(
453 self._fingerprint, transform, kwargs_for_fingerprint
454 )
456 # Call actual function
--> 458 out = func(self, *args, **kwargs)
460 # Update fingerprint of in-place transforms + update in-place history of transforms
462 if inplace: # update after calling func so that the fingerprint doesn't change if the function fails
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_dataset.py:2356, in Dataset._map_single(self, function, with_indices, with_rank, 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)
2354 writer.write_table(batch)
2355 else:
-> 2356 writer.write_batch(batch)
2357 if update_data and writer is not None:
2358 writer.finalize() # close_stream=bool(buf_writer is None)) # We only close if we are writing in a file
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:507, in ArrowWriter.write_batch(self, batch_examples, writer_batch_size)
505 col_try_type = try_features[col] if try_features is not None and col in try_features else None
506 typed_sequence = OptimizedTypedSequence(batch_examples[col], type=col_type, try_type=col_try_type, col=col)
--> 507 arrays.append(pa.array(typed_sequence))
508 inferred_features[col] = typed_sequence.get_inferred_type()
509 schema = inferred_features.arrow_schema if self.pa_writer is None else self.schema
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:236, in pyarrow.lib.array()
File /opt/conda/lib/python3.10/site-packages/pyarrow/array.pxi:110, in pyarrow.lib._handle_arrow_array_protocol()
File /opt/conda/lib/python3.10/site-packages/datasets/arrow_writer.py:184, in TypedSequence.__arrow_array__(self, type)
182 out = numpy_to_pyarrow_listarray(data)
183 elif isinstance(data, list) and data and isinstance(first_non_null_value(data)[1], np.ndarray):
--> 184 out = list_of_np_array_to_pyarrow_listarray(data)
185 else:
186 trying_cast_to_python_objects = True
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1174, in list_of_np_array_to_pyarrow_listarray(l_arr, type)
1172 """Build a PyArrow ListArray from a possibly nested list of NumPy arrays"""
1173 if len(l_arr) > 0:
-> 1174 return list_of_pa_arrays_to_pyarrow_listarray(
1175 [numpy_to_pyarrow_listarray(arr, type=type) if arr is not None else None for arr in l_arr]
1176 )
1177 else:
1178 return pa.array([], type=type)
File /opt/conda/lib/python3.10/site-packages/datasets/features/features.py:1163, in list_of_pa_arrays_to_pyarrow_listarray(l_arr)
1160 null_indices = [i for i, arr in enumerate(l_arr) if arr is None]
1161 l_arr = [arr for arr in l_arr if arr is not None]
1162 offsets = np.cumsum(
-> 1163 [0] + [len(arr) for arr in l_arr], dtype=np.object
1164 ) # convert to dtype object to allow None insertion
1165 offsets = np.insert(offsets, null_indices, None)
1166 offsets = pa.array(offsets, type=pa.int32())
File /opt/conda/lib/python3.10/site-packages/numpy/__init__.py:324, in __getattr__(attr)
319 warnings.warn(
320 f"In the future `np.{attr}` will be defined as the "
321 "corresponding NumPy scalar.", FutureWarning, stacklevel=2)
323 if attr in __former_attrs__:
--> 324 raise AttributeError(__former_attrs__[attr])
326 if attr == 'testing':
327 import numpy.testing as testing
AttributeError: module 'numpy' has no attribute 'object'.
`np.object` was a deprecated alias for the builtin `object`. To avoid this error in existing code, use `object` by itself. Doing this will not modify any behavior and is safe.
The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at:
https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
```
</details>
### Steps to reproduce the bug
Run above code in Kaggle Notebook.
### Expected behavior
I can resample audio data without fail.
### Environment info
- `datasets` version: 2.1.0
- Platform: Linux-5.15.133+-x86_64-with-glibc2.31
- Python version: 3.10.13
- PyArrow version: 11.0.0
- Pandas version: 2.2.1
Kaggle removed the problematic `datasets==2.1.0` pin last week, so I'm closing this issue (now it pre-installs the latest version). | [
-0.22040624916553497,
0.02457261085510254,
-0.03590615093708038,
0.00249418243765831,
0.4256819486618042,
-0.05877440422773361,
0.38796424865722656,
0.31887146830558777,
-0.18779487907886505,
0.3352118730545044,
-0.2983803451061249,
0.683189332485199,
-0.2616359293460846,
0.252659529447555... |
https://github.com/huggingface/datasets/issues/6782 | Image cast_storage very slow for arrays (e.g. numpy, tensors) | This may be a solution that only changes `cast_storage` of `Image`.
However, I'm not totally sure that the assumptions hold that are made about the `ListArray`.
```python
elif pa.types.is_list(storage.type):
from .features import Array3DExtensionType
def get_shapes(arr):
shape = ()
while isinstance(arr, pa.ListArray):
len_curr = len(arr)
arr = arr.flatten()
len_new = len(arr)
shape = shape + (len_new // len_curr,)
return shape
def get_dtypes(arr):
dtype = storage.type
while hasattr(dtype, "value_type"):
dtype = dtype.value_type
return dtype
arrays = []
for i, is_null in enumerate(storage.is_null()):
if not is_null.as_py():
storage_part = storage.take([i])
shape = get_shapes(storage_part)
dtype = get_dtypes(storage_part)
extension_type = Array3DExtensionType(shape=shape, dtype=str(dtype))
array = pa.ExtensionArray.from_storage(extension_type, storage_part)
arrays.append(array.to_numpy().squeeze(0))
else:
arrays.append(None)
bytes_array = pa.array(
[encode_np_array(arr)["bytes"] if arr is not None else None for arr in arrays],
type=pa.binary(),
)
path_array = pa.array([None] * len(storage), type=pa.string())
storage = pa.StructArray.from_arrays(
[bytes_array, path_array], ["bytes", "path"], mask=bytes_array.is_null()
)
```
(Edited): to handle nulls
Notably this doesn't change anything about the passing through of data or other things, just in the `Image` class.
Seems quite fast:
```bash
Fri Apr 5 17:55:51 2024 restats
63818 function calls (61995 primitive calls) in 0.812 seconds
Ordered by: cumulative time
List reduced from 1051 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
47/1 0.000 0.000 0.810 0.810 {built-in method builtins.exec}
2/1 0.000 0.000 0.810 0.810 <string>:1(<module>)
2/1 0.000 0.000 0.809 0.809 arrow_dataset.py:594(wrapper)
2/1 0.000 0.000 0.809 0.809 arrow_dataset.py:551(wrapper)
2/1 0.000 0.000 0.809 0.809 arrow_dataset.py:2916(map)
3 0.000 0.000 0.807 0.269 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 0.760 0.760 arrow_writer.py:589(finalize)
1 0.000 0.000 0.760 0.760 arrow_writer.py:423(write_examples_on_file)
1 0.000 0.000 0.759 0.759 arrow_writer.py:527(write_batch)
1 0.001 0.001 0.754 0.754 arrow_writer.py:161(__arrow_array__)
2/1 0.000 0.000 0.719 0.719 table.py:1800(wrapper)
1 0.000 0.000 0.719 0.719 table.py:1950(cast_array_to_feature)
1 0.006 0.006 0.718 0.718 image.py:209(cast_storage)
1 0.000 0.000 0.451 0.451 image.py:361(encode_np_array)
1 0.000 0.000 0.444 0.444 image.py:343(image_to_bytes)
1 0.000 0.000 0.413 0.413 Image.py:2376(save)
1 0.000 0.000 0.413 0.413 PngImagePlugin.py:1233(_save)
1 0.000 0.000 0.413 0.413 ImageFile.py:517(_save)
1 0.000 0.000 0.413 0.413 ImageFile.py:545(_encode_tile)
397 0.409 0.001 0.409 0.001 {method 'encode' of 'ImagingEncoder' objects}
``` | Update: see comments below
### Describe the bug
Operations that save an image from a path are very slow.
I believe the reason for this is that the image data (`numpy`) is converted into `pyarrow` format but then back to python using `.pylist()` before being converted to a numpy array again.
`pylist` is already slow but used on a multi-dimensional numpy array such as an image it takes a very long time.
From the trace below we can see that `__arrow_array__` takes a long time.
It is currently also called in `get_inferred_type`, this should be removable #6781 but doesn't change the underyling issue.
The conversion to `pyarrow` and back also leads to the `numpy` array having type `int64` which causes a warning message because the image type excepts `uint8`.
However, originally the `numpy` image array was in `uint8`.
### Steps to reproduce the bug
```python
from PIL import Image
import numpy as np
import datasets
import cProfile
image = Image.fromarray(np.random.randint(0, 255, (2048, 2048, 3), dtype=np.uint8))
image.save("test_image.jpg")
ds = datasets.Dataset.from_dict(
{"image": ["test_image.jpg"]},
features=datasets.Features({"image": datasets.Image(decode=True)}),
)
# load as numpy array, e.g. for further processing with map
# same result as map returning numpy arrays
ds.set_format("numpy")
cProfile.run("ds.map(writer_batch_size=1, load_from_cache_file=False)", "restats")
```
```bash
Fri Apr 5 14:56:17 2024 restats
66817 function calls (64992 primitive calls) in 33.382 seconds
Ordered by: cumulative time
List reduced from 1073 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
46/1 0.000 0.000 33.382 33.382 {built-in method builtins.exec}
1 0.000 0.000 33.382 33.382 <string>:1(<module>)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:594(wrapper)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:551(wrapper)
1 0.000 0.000 33.379 33.379 arrow_dataset.py:2916(map)
4 0.000 0.000 33.327 8.332 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 33.311 33.311 arrow_writer.py:465(write)
2 0.000 0.000 33.311 16.656 arrow_writer.py:423(write_examples_on_file)
1 0.000 0.000 33.311 33.311 arrow_writer.py:527(write_batch)
2 14.484 7.242 33.260 16.630 arrow_writer.py:161(__arrow_array__)
1 0.001 0.001 16.438 16.438 arrow_writer.py:121(get_inferred_type)
1 0.000 0.000 14.398 14.398 threading.py:637(wait)
1 0.000 0.000 14.398 14.398 threading.py:323(wait)
8 14.398 1.800 14.398 1.800 {method 'acquire' of '_thread.lock' objects}
4/2 0.000 0.000 4.337 2.169 table.py:1800(wrapper)
2 0.000 0.000 4.337 2.169 table.py:1950(cast_array_to_feature)
2 0.475 0.238 4.337 2.169 image.py:209(cast_storage)
9 2.583 0.287 2.583 0.287 {built-in method numpy.array}
2 0.000 0.000 1.284 0.642 image.py:319(encode_np_array)
2 0.000 0.000 1.246 0.623 image.py:301(image_to_bytes)
```
### Expected behavior
The `numpy` image data should be passed through as it will be directly consumed by `pillow` to convert it to bytes.
As an example one can replace `list_of_np_array_to_pyarrow_listarray(data)` in `__arrow_array__` with just `out = data` as a test.
We have to change `cast_storage` of the `Image` feature so it handles the passed through data (& if to handle type before)
```python
bytes_array = pa.array(
[encode_np_array(arr)["bytes"] if arr is not None else None for arr in storage],
type=pa.binary(),
)
```
Leading to the following:
```bash
Fri Apr 5 15:44:27 2024 restats
66419 function calls (64595 primitive calls) in 0.937 seconds
Ordered by: cumulative time
List reduced from 1023 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
47/1 0.000 0.000 0.935 0.935 {built-in method builtins.exec}
2/1 0.000 0.000 0.935 0.935 <string>:1(<module>)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:594(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:551(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:2916(map)
4 0.000 0.000 0.933 0.233 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 0.883 0.883 arrow_writer.py:466(write)
2 0.000 0.000 0.883 0.441 arrow_writer.py:424(write_examples_on_file)
1 0.000 0.000 0.882 0.882 arrow_writer.py:528(write_batch)
2 0.000 0.000 0.877 0.439 arrow_writer.py:161(__arrow_array__)
4/2 0.000 0.000 0.877 0.439 table.py:1800(wrapper)
2 0.000 0.000 0.877 0.439 table.py:1950(cast_array_to_feature)
2 0.009 0.005 0.877 0.439 image.py:209(cast_storage)
2 0.000 0.000 0.868 0.434 image.py:335(encode_np_array)
2 0.000 0.000 0.856 0.428 image.py:317(image_to_bytes)
2 0.000 0.000 0.822 0.411 Image.py:2376(save)
2 0.000 0.000 0.822 0.411 PngImagePlugin.py:1233(_save)
2 0.000 0.000 0.822 0.411 ImageFile.py:517(_save)
2 0.000 0.000 0.821 0.411 ImageFile.py:545(_encode_tile)
589 0.803 0.001 0.803 0.001 {method 'encode' of 'ImagingEncoder' objects}
```
This is of course only a test as it passes through all `numpy` arrays irrespective of if they should be an image.
Also I guess `cast_storage` is meant for casting `pyarrow` storage exclusively.
Converting to `pyarrow` array seems like a good solution as it also handles `pytorch` tensors etc., maybe there is a more efficient way to create a PIL image from a `pyarrow` array?
Not sure how this should be handled but I would be happy to help if there is a good solution.
### Environment info
- `datasets` version: 2.18.1.dev0
- Platform: Linux-6.7.11-200.fc39.x86_64-x86_64-with-glibc2.38
- Python version: 3.12.2
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.3.1 | 325 | Image cast_storage very slow for arrays (e.g. numpy, tensors)
Update: see comments below
### Describe the bug
Operations that save an image from a path are very slow.
I believe the reason for this is that the image data (`numpy`) is converted into `pyarrow` format but then back to python using `.pylist()` before being converted to a numpy array again.
`pylist` is already slow but used on a multi-dimensional numpy array such as an image it takes a very long time.
From the trace below we can see that `__arrow_array__` takes a long time.
It is currently also called in `get_inferred_type`, this should be removable #6781 but doesn't change the underyling issue.
The conversion to `pyarrow` and back also leads to the `numpy` array having type `int64` which causes a warning message because the image type excepts `uint8`.
However, originally the `numpy` image array was in `uint8`.
### Steps to reproduce the bug
```python
from PIL import Image
import numpy as np
import datasets
import cProfile
image = Image.fromarray(np.random.randint(0, 255, (2048, 2048, 3), dtype=np.uint8))
image.save("test_image.jpg")
ds = datasets.Dataset.from_dict(
{"image": ["test_image.jpg"]},
features=datasets.Features({"image": datasets.Image(decode=True)}),
)
# load as numpy array, e.g. for further processing with map
# same result as map returning numpy arrays
ds.set_format("numpy")
cProfile.run("ds.map(writer_batch_size=1, load_from_cache_file=False)", "restats")
```
```bash
Fri Apr 5 14:56:17 2024 restats
66817 function calls (64992 primitive calls) in 33.382 seconds
Ordered by: cumulative time
List reduced from 1073 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
46/1 0.000 0.000 33.382 33.382 {built-in method builtins.exec}
1 0.000 0.000 33.382 33.382 <string>:1(<module>)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:594(wrapper)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:551(wrapper)
1 0.000 0.000 33.379 33.379 arrow_dataset.py:2916(map)
4 0.000 0.000 33.327 8.332 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 33.311 33.311 arrow_writer.py:465(write)
2 0.000 0.000 33.311 16.656 arrow_writer.py:423(write_examples_on_file)
1 0.000 0.000 33.311 33.311 arrow_writer.py:527(write_batch)
2 14.484 7.242 33.260 16.630 arrow_writer.py:161(__arrow_array__)
1 0.001 0.001 16.438 16.438 arrow_writer.py:121(get_inferred_type)
1 0.000 0.000 14.398 14.398 threading.py:637(wait)
1 0.000 0.000 14.398 14.398 threading.py:323(wait)
8 14.398 1.800 14.398 1.800 {method 'acquire' of '_thread.lock' objects}
4/2 0.000 0.000 4.337 2.169 table.py:1800(wrapper)
2 0.000 0.000 4.337 2.169 table.py:1950(cast_array_to_feature)
2 0.475 0.238 4.337 2.169 image.py:209(cast_storage)
9 2.583 0.287 2.583 0.287 {built-in method numpy.array}
2 0.000 0.000 1.284 0.642 image.py:319(encode_np_array)
2 0.000 0.000 1.246 0.623 image.py:301(image_to_bytes)
```
### Expected behavior
The `numpy` image data should be passed through as it will be directly consumed by `pillow` to convert it to bytes.
As an example one can replace `list_of_np_array_to_pyarrow_listarray(data)` in `__arrow_array__` with just `out = data` as a test.
We have to change `cast_storage` of the `Image` feature so it handles the passed through data (& if to handle type before)
```python
bytes_array = pa.array(
[encode_np_array(arr)["bytes"] if arr is not None else None for arr in storage],
type=pa.binary(),
)
```
Leading to the following:
```bash
Fri Apr 5 15:44:27 2024 restats
66419 function calls (64595 primitive calls) in 0.937 seconds
Ordered by: cumulative time
List reduced from 1023 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
47/1 0.000 0.000 0.935 0.935 {built-in method builtins.exec}
2/1 0.000 0.000 0.935 0.935 <string>:1(<module>)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:594(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:551(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:2916(map)
4 0.000 0.000 0.933 0.233 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 0.883 0.883 arrow_writer.py:466(write)
2 0.000 0.000 0.883 0.441 arrow_writer.py:424(write_examples_on_file)
1 0.000 0.000 0.882 0.882 arrow_writer.py:528(write_batch)
2 0.000 0.000 0.877 0.439 arrow_writer.py:161(__arrow_array__)
4/2 0.000 0.000 0.877 0.439 table.py:1800(wrapper)
2 0.000 0.000 0.877 0.439 table.py:1950(cast_array_to_feature)
2 0.009 0.005 0.877 0.439 image.py:209(cast_storage)
2 0.000 0.000 0.868 0.434 image.py:335(encode_np_array)
2 0.000 0.000 0.856 0.428 image.py:317(image_to_bytes)
2 0.000 0.000 0.822 0.411 Image.py:2376(save)
2 0.000 0.000 0.822 0.411 PngImagePlugin.py:1233(_save)
2 0.000 0.000 0.822 0.411 ImageFile.py:517(_save)
2 0.000 0.000 0.821 0.411 ImageFile.py:545(_encode_tile)
589 0.803 0.001 0.803 0.001 {method 'encode' of 'ImagingEncoder' objects}
```
This is of course only a test as it passes through all `numpy` arrays irrespective of if they should be an image.
Also I guess `cast_storage` is meant for casting `pyarrow` storage exclusively.
Converting to `pyarrow` array seems like a good solution as it also handles `pytorch` tensors etc., maybe there is a more efficient way to create a PIL image from a `pyarrow` array?
Not sure how this should be handled but I would be happy to help if there is a good solution.
### Environment info
- `datasets` version: 2.18.1.dev0
- Platform: Linux-6.7.11-200.fc39.x86_64-x86_64-with-glibc2.38
- Python version: 3.12.2
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.3.1
This may be a solution that only changes `cast_storage` of `Image`.
However, I'm not totally sure that the assumptions hold that are made about the `ListArray`.
```python
elif pa.types.is_list(storage.type):
from .features import Array3DExtensionType
def get_shapes(arr):
shape = ()
while isinstance(arr, pa.ListArray):
len_curr = len(arr)
arr = arr.flatten()
len_new = len(arr)
shape = shape + (len_new // len_curr,)
return shape
def get_dtypes(arr):
dtype = storage.type
while hasattr(dtype, "value_type"):
dtype = dtype.value_type
return dtype
arrays = []
for i, is_null in enumerate(storage.is_null()):
if not is_null.as_py():
storage_part = storage.take([i])
shape = get_shapes(storage_part)
dtype = get_dtypes(storage_part)
extension_type = Array3DExtensionType(shape=shape, dtype=str(dtype))
array = pa.ExtensionArray.from_storage(extension_type, storage_part)
arrays.append(array.to_numpy().squeeze(0))
else:
arrays.append(None)
bytes_array = pa.array(
[encode_np_array(arr)["bytes"] if arr is not None else None for arr in arrays],
type=pa.binary(),
)
path_array = pa.array([None] * len(storage), type=pa.string())
storage = pa.StructArray.from_arrays(
[bytes_array, path_array], ["bytes", "path"], mask=bytes_array.is_null()
)
```
(Edited): to handle nulls
Notably this doesn't change anything about the passing through of data or other things, just in the `Image` class.
Seems quite fast:
```bash
Fri Apr 5 17:55:51 2024 restats
63818 function calls (61995 primitive calls) in 0.812 seconds
Ordered by: cumulative time
List reduced from 1051 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
47/1 0.000 0.000 0.810 0.810 {built-in method builtins.exec}
2/1 0.000 0.000 0.810 0.810 <string>:1(<module>)
2/1 0.000 0.000 0.809 0.809 arrow_dataset.py:594(wrapper)
2/1 0.000 0.000 0.809 0.809 arrow_dataset.py:551(wrapper)
2/1 0.000 0.000 0.809 0.809 arrow_dataset.py:2916(map)
3 0.000 0.000 0.807 0.269 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 0.760 0.760 arrow_writer.py:589(finalize)
1 0.000 0.000 0.760 0.760 arrow_writer.py:423(write_examples_on_file)
1 0.000 0.000 0.759 0.759 arrow_writer.py:527(write_batch)
1 0.001 0.001 0.754 0.754 arrow_writer.py:161(__arrow_array__)
2/1 0.000 0.000 0.719 0.719 table.py:1800(wrapper)
1 0.000 0.000 0.719 0.719 table.py:1950(cast_array_to_feature)
1 0.006 0.006 0.718 0.718 image.py:209(cast_storage)
1 0.000 0.000 0.451 0.451 image.py:361(encode_np_array)
1 0.000 0.000 0.444 0.444 image.py:343(image_to_bytes)
1 0.000 0.000 0.413 0.413 Image.py:2376(save)
1 0.000 0.000 0.413 0.413 PngImagePlugin.py:1233(_save)
1 0.000 0.000 0.413 0.413 ImageFile.py:517(_save)
1 0.000 0.000 0.413 0.413 ImageFile.py:545(_encode_tile)
397 0.409 0.001 0.409 0.001 {method 'encode' of 'ImagingEncoder' objects}
``` | [
-0.20979847013950348,
-0.07302135229110718,
-0.0948244035243988,
0.2252599000930786,
0.46149539947509766,
0.0019903481006622314,
0.2881958782672882,
0.21179988980293274,
0.04564521089196205,
0.21190133690834045,
-0.1347082406282425,
0.47250670194625854,
-0.1871844381093979,
-0.567169308662... |
https://github.com/huggingface/datasets/issues/6782 | Image cast_storage very slow for arrays (e.g. numpy, tensors) | This actually applies to all arrays (numpy or tensors like in torch), not only from external files.
```python
import numpy as np
import datasets
ds = datasets.Dataset.from_dict(
{"image": [np.random.randint(0, 255, (2048, 2048, 3), dtype=np.uint8)]},
features=datasets.Features({"image": datasets.Image(decode=True)}),
)
ds.set_format("numpy")
ds = ds.map(load_from_cache_file=False)
``` | Update: see comments below
### Describe the bug
Operations that save an image from a path are very slow.
I believe the reason for this is that the image data (`numpy`) is converted into `pyarrow` format but then back to python using `.pylist()` before being converted to a numpy array again.
`pylist` is already slow but used on a multi-dimensional numpy array such as an image it takes a very long time.
From the trace below we can see that `__arrow_array__` takes a long time.
It is currently also called in `get_inferred_type`, this should be removable #6781 but doesn't change the underyling issue.
The conversion to `pyarrow` and back also leads to the `numpy` array having type `int64` which causes a warning message because the image type excepts `uint8`.
However, originally the `numpy` image array was in `uint8`.
### Steps to reproduce the bug
```python
from PIL import Image
import numpy as np
import datasets
import cProfile
image = Image.fromarray(np.random.randint(0, 255, (2048, 2048, 3), dtype=np.uint8))
image.save("test_image.jpg")
ds = datasets.Dataset.from_dict(
{"image": ["test_image.jpg"]},
features=datasets.Features({"image": datasets.Image(decode=True)}),
)
# load as numpy array, e.g. for further processing with map
# same result as map returning numpy arrays
ds.set_format("numpy")
cProfile.run("ds.map(writer_batch_size=1, load_from_cache_file=False)", "restats")
```
```bash
Fri Apr 5 14:56:17 2024 restats
66817 function calls (64992 primitive calls) in 33.382 seconds
Ordered by: cumulative time
List reduced from 1073 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
46/1 0.000 0.000 33.382 33.382 {built-in method builtins.exec}
1 0.000 0.000 33.382 33.382 <string>:1(<module>)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:594(wrapper)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:551(wrapper)
1 0.000 0.000 33.379 33.379 arrow_dataset.py:2916(map)
4 0.000 0.000 33.327 8.332 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 33.311 33.311 arrow_writer.py:465(write)
2 0.000 0.000 33.311 16.656 arrow_writer.py:423(write_examples_on_file)
1 0.000 0.000 33.311 33.311 arrow_writer.py:527(write_batch)
2 14.484 7.242 33.260 16.630 arrow_writer.py:161(__arrow_array__)
1 0.001 0.001 16.438 16.438 arrow_writer.py:121(get_inferred_type)
1 0.000 0.000 14.398 14.398 threading.py:637(wait)
1 0.000 0.000 14.398 14.398 threading.py:323(wait)
8 14.398 1.800 14.398 1.800 {method 'acquire' of '_thread.lock' objects}
4/2 0.000 0.000 4.337 2.169 table.py:1800(wrapper)
2 0.000 0.000 4.337 2.169 table.py:1950(cast_array_to_feature)
2 0.475 0.238 4.337 2.169 image.py:209(cast_storage)
9 2.583 0.287 2.583 0.287 {built-in method numpy.array}
2 0.000 0.000 1.284 0.642 image.py:319(encode_np_array)
2 0.000 0.000 1.246 0.623 image.py:301(image_to_bytes)
```
### Expected behavior
The `numpy` image data should be passed through as it will be directly consumed by `pillow` to convert it to bytes.
As an example one can replace `list_of_np_array_to_pyarrow_listarray(data)` in `__arrow_array__` with just `out = data` as a test.
We have to change `cast_storage` of the `Image` feature so it handles the passed through data (& if to handle type before)
```python
bytes_array = pa.array(
[encode_np_array(arr)["bytes"] if arr is not None else None for arr in storage],
type=pa.binary(),
)
```
Leading to the following:
```bash
Fri Apr 5 15:44:27 2024 restats
66419 function calls (64595 primitive calls) in 0.937 seconds
Ordered by: cumulative time
List reduced from 1023 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
47/1 0.000 0.000 0.935 0.935 {built-in method builtins.exec}
2/1 0.000 0.000 0.935 0.935 <string>:1(<module>)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:594(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:551(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:2916(map)
4 0.000 0.000 0.933 0.233 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 0.883 0.883 arrow_writer.py:466(write)
2 0.000 0.000 0.883 0.441 arrow_writer.py:424(write_examples_on_file)
1 0.000 0.000 0.882 0.882 arrow_writer.py:528(write_batch)
2 0.000 0.000 0.877 0.439 arrow_writer.py:161(__arrow_array__)
4/2 0.000 0.000 0.877 0.439 table.py:1800(wrapper)
2 0.000 0.000 0.877 0.439 table.py:1950(cast_array_to_feature)
2 0.009 0.005 0.877 0.439 image.py:209(cast_storage)
2 0.000 0.000 0.868 0.434 image.py:335(encode_np_array)
2 0.000 0.000 0.856 0.428 image.py:317(image_to_bytes)
2 0.000 0.000 0.822 0.411 Image.py:2376(save)
2 0.000 0.000 0.822 0.411 PngImagePlugin.py:1233(_save)
2 0.000 0.000 0.822 0.411 ImageFile.py:517(_save)
2 0.000 0.000 0.821 0.411 ImageFile.py:545(_encode_tile)
589 0.803 0.001 0.803 0.001 {method 'encode' of 'ImagingEncoder' objects}
```
This is of course only a test as it passes through all `numpy` arrays irrespective of if they should be an image.
Also I guess `cast_storage` is meant for casting `pyarrow` storage exclusively.
Converting to `pyarrow` array seems like a good solution as it also handles `pytorch` tensors etc., maybe there is a more efficient way to create a PIL image from a `pyarrow` array?
Not sure how this should be handled but I would be happy to help if there is a good solution.
### Environment info
- `datasets` version: 2.18.1.dev0
- Platform: Linux-6.7.11-200.fc39.x86_64-x86_64-with-glibc2.38
- Python version: 3.12.2
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.3.1 | 42 | Image cast_storage very slow for arrays (e.g. numpy, tensors)
Update: see comments below
### Describe the bug
Operations that save an image from a path are very slow.
I believe the reason for this is that the image data (`numpy`) is converted into `pyarrow` format but then back to python using `.pylist()` before being converted to a numpy array again.
`pylist` is already slow but used on a multi-dimensional numpy array such as an image it takes a very long time.
From the trace below we can see that `__arrow_array__` takes a long time.
It is currently also called in `get_inferred_type`, this should be removable #6781 but doesn't change the underyling issue.
The conversion to `pyarrow` and back also leads to the `numpy` array having type `int64` which causes a warning message because the image type excepts `uint8`.
However, originally the `numpy` image array was in `uint8`.
### Steps to reproduce the bug
```python
from PIL import Image
import numpy as np
import datasets
import cProfile
image = Image.fromarray(np.random.randint(0, 255, (2048, 2048, 3), dtype=np.uint8))
image.save("test_image.jpg")
ds = datasets.Dataset.from_dict(
{"image": ["test_image.jpg"]},
features=datasets.Features({"image": datasets.Image(decode=True)}),
)
# load as numpy array, e.g. for further processing with map
# same result as map returning numpy arrays
ds.set_format("numpy")
cProfile.run("ds.map(writer_batch_size=1, load_from_cache_file=False)", "restats")
```
```bash
Fri Apr 5 14:56:17 2024 restats
66817 function calls (64992 primitive calls) in 33.382 seconds
Ordered by: cumulative time
List reduced from 1073 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
46/1 0.000 0.000 33.382 33.382 {built-in method builtins.exec}
1 0.000 0.000 33.382 33.382 <string>:1(<module>)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:594(wrapper)
1 0.000 0.000 33.382 33.382 arrow_dataset.py:551(wrapper)
1 0.000 0.000 33.379 33.379 arrow_dataset.py:2916(map)
4 0.000 0.000 33.327 8.332 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 33.311 33.311 arrow_writer.py:465(write)
2 0.000 0.000 33.311 16.656 arrow_writer.py:423(write_examples_on_file)
1 0.000 0.000 33.311 33.311 arrow_writer.py:527(write_batch)
2 14.484 7.242 33.260 16.630 arrow_writer.py:161(__arrow_array__)
1 0.001 0.001 16.438 16.438 arrow_writer.py:121(get_inferred_type)
1 0.000 0.000 14.398 14.398 threading.py:637(wait)
1 0.000 0.000 14.398 14.398 threading.py:323(wait)
8 14.398 1.800 14.398 1.800 {method 'acquire' of '_thread.lock' objects}
4/2 0.000 0.000 4.337 2.169 table.py:1800(wrapper)
2 0.000 0.000 4.337 2.169 table.py:1950(cast_array_to_feature)
2 0.475 0.238 4.337 2.169 image.py:209(cast_storage)
9 2.583 0.287 2.583 0.287 {built-in method numpy.array}
2 0.000 0.000 1.284 0.642 image.py:319(encode_np_array)
2 0.000 0.000 1.246 0.623 image.py:301(image_to_bytes)
```
### Expected behavior
The `numpy` image data should be passed through as it will be directly consumed by `pillow` to convert it to bytes.
As an example one can replace `list_of_np_array_to_pyarrow_listarray(data)` in `__arrow_array__` with just `out = data` as a test.
We have to change `cast_storage` of the `Image` feature so it handles the passed through data (& if to handle type before)
```python
bytes_array = pa.array(
[encode_np_array(arr)["bytes"] if arr is not None else None for arr in storage],
type=pa.binary(),
)
```
Leading to the following:
```bash
Fri Apr 5 15:44:27 2024 restats
66419 function calls (64595 primitive calls) in 0.937 seconds
Ordered by: cumulative time
List reduced from 1023 to 20 due to restriction <20>
ncalls tottime percall cumtime percall filename:lineno(function)
47/1 0.000 0.000 0.935 0.935 {built-in method builtins.exec}
2/1 0.000 0.000 0.935 0.935 <string>:1(<module>)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:594(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:551(wrapper)
2/1 0.000 0.000 0.934 0.934 arrow_dataset.py:2916(map)
4 0.000 0.000 0.933 0.233 arrow_dataset.py:3277(_map_single)
1 0.000 0.000 0.883 0.883 arrow_writer.py:466(write)
2 0.000 0.000 0.883 0.441 arrow_writer.py:424(write_examples_on_file)
1 0.000 0.000 0.882 0.882 arrow_writer.py:528(write_batch)
2 0.000 0.000 0.877 0.439 arrow_writer.py:161(__arrow_array__)
4/2 0.000 0.000 0.877 0.439 table.py:1800(wrapper)
2 0.000 0.000 0.877 0.439 table.py:1950(cast_array_to_feature)
2 0.009 0.005 0.877 0.439 image.py:209(cast_storage)
2 0.000 0.000 0.868 0.434 image.py:335(encode_np_array)
2 0.000 0.000 0.856 0.428 image.py:317(image_to_bytes)
2 0.000 0.000 0.822 0.411 Image.py:2376(save)
2 0.000 0.000 0.822 0.411 PngImagePlugin.py:1233(_save)
2 0.000 0.000 0.822 0.411 ImageFile.py:517(_save)
2 0.000 0.000 0.821 0.411 ImageFile.py:545(_encode_tile)
589 0.803 0.001 0.803 0.001 {method 'encode' of 'ImagingEncoder' objects}
```
This is of course only a test as it passes through all `numpy` arrays irrespective of if they should be an image.
Also I guess `cast_storage` is meant for casting `pyarrow` storage exclusively.
Converting to `pyarrow` array seems like a good solution as it also handles `pytorch` tensors etc., maybe there is a more efficient way to create a PIL image from a `pyarrow` array?
Not sure how this should be handled but I would be happy to help if there is a good solution.
### Environment info
- `datasets` version: 2.18.1.dev0
- Platform: Linux-6.7.11-200.fc39.x86_64-x86_64-with-glibc2.38
- Python version: 3.12.2
- `huggingface_hub` version: 0.22.2
- PyArrow version: 15.0.2
- Pandas version: 2.2.1
- `fsspec` version: 2024.3.1
This actually applies to all arrays (numpy or tensors like in torch), not only from external files.
```python
import numpy as np
import datasets
ds = datasets.Dataset.from_dict(
{"image": [np.random.randint(0, 255, (2048, 2048, 3), dtype=np.uint8)]},
features=datasets.Features({"image": datasets.Image(decode=True)}),
)
ds.set_format("numpy")
ds = ds.map(load_from_cache_file=False)
``` | [
-0.20979847013950348,
-0.07302135229110718,
-0.0948244035243988,
0.2252599000930786,
0.46149539947509766,
0.0019903481006622314,
0.2881958782672882,
0.21179988980293274,
0.04564521089196205,
0.21190133690834045,
-0.1347082406282425,
0.47250670194625854,
-0.1871844381093979,
-0.567169308662... |
https://github.com/huggingface/datasets/issues/6778 | Dataset.to_csv() missing commas in columns with lists | Hello!
This is due to how pandas write numpy arrays to csv. [Source](https://stackoverflow.com/questions/54753179/to-csv-saves-np-array-as-string-instead-of-as-a-list)
To fix this, you can convert them to list yourselves.
```python
df = ds.to_pandas()
df['int'] = df['int'].apply(lambda arr: list(arr))
df.to_csv(index=False, '../output/temp.csv')
```
I think it would be good if `datasets` would do the conversion itself, but it's a breaking change and I would wait for the greenlight from someone from HF. | ### Describe the bug
The `to_csv()` method does not output commas in lists. So when the Dataset is loaded back in the data structure of the column with a list is not correct.
Here's an example:
Obviously, it's not as trivial as inserting commas in the list, since its a comma-separated file. But hopefully there's a way to export the list in a way that it'll be imported by `load_dataset()` correctly.
### Steps to reproduce the bug
Here's some code to reproduce the bug:
```python
from datasets import Dataset
ds = Dataset.from_dict(
{
"pokemon": ["bulbasaur", "squirtle"],
"type": ["grass", "water"]
}
)
def ascii_to_hex(text):
return [ord(c) for c in text]
ds = ds.map(lambda x: {"int": ascii_to_hex(x['pokemon'])})
ds.to_csv('../output/temp.csv')
```
temp.csv then contains:
```
### Expected behavior
ACTUAL OUTPUT:
```
pokemon,type,int
bulbasaur,grass,[ 98 117 108 98 97 115 97 117 114]
squirtle,water,[115 113 117 105 114 116 108 101]
```
EXPECTED OUTPUT:
```
pokemon,type,int
bulbasaur,grass,[98, 117, 108, 98, 97, 115, 97, 117, 114]
squirtle,water,[115, 113, 117, 105, 114, 116, 108, 101]
```
or probably something more like this since it's a CSV file:
```
pokemon,type,int
bulbasaur,grass,"[98, 117, 108, 98, 97, 115, 97, 117, 114]"
squirtle,water,"[115, 113, 117, 105, 114, 116, 108, 101]"
```
### Environment info
### Package Version
Name: datasets
Version: 2.16.1
### Python
version: 3.10.12
### OS Info
PRETTY_NAME="Ubuntu 22.04.4 LTS"
NAME="Ubuntu"
VERSION_ID="22.04"
VERSION="22.04.4 LTS (Jammy Jellyfish)"
VERSION_CODENAME=jammy
ID=ubuntu
ID_LIKE=debian
...
UBUNTU_CODENAME=jammy | 64 | Dataset.to_csv() missing commas in columns with lists
### Describe the bug
The `to_csv()` method does not output commas in lists. So when the Dataset is loaded back in the data structure of the column with a list is not correct.
Here's an example:
Obviously, it's not as trivial as inserting commas in the list, since its a comma-separated file. But hopefully there's a way to export the list in a way that it'll be imported by `load_dataset()` correctly.
### Steps to reproduce the bug
Here's some code to reproduce the bug:
```python
from datasets import Dataset
ds = Dataset.from_dict(
{
"pokemon": ["bulbasaur", "squirtle"],
"type": ["grass", "water"]
}
)
def ascii_to_hex(text):
return [ord(c) for c in text]
ds = ds.map(lambda x: {"int": ascii_to_hex(x['pokemon'])})
ds.to_csv('../output/temp.csv')
```
temp.csv then contains:
```
### Expected behavior
ACTUAL OUTPUT:
```
pokemon,type,int
bulbasaur,grass,[ 98 117 108 98 97 115 97 117 114]
squirtle,water,[115 113 117 105 114 116 108 101]
```
EXPECTED OUTPUT:
```
pokemon,type,int
bulbasaur,grass,[98, 117, 108, 98, 97, 115, 97, 117, 114]
squirtle,water,[115, 113, 117, 105, 114, 116, 108, 101]
```
or probably something more like this since it's a CSV file:
```
pokemon,type,int
bulbasaur,grass,"[98, 117, 108, 98, 97, 115, 97, 117, 114]"
squirtle,water,"[115, 113, 117, 105, 114, 116, 108, 101]"
```
### Environment info
### Package Version
Name: datasets
Version: 2.16.1
### Python
version: 3.10.12
### OS Info
PRETTY_NAME="Ubuntu 22.04.4 LTS"
NAME="Ubuntu"
VERSION_ID="22.04"
VERSION="22.04.4 LTS (Jammy Jellyfish)"
VERSION_CODENAME=jammy
ID=ubuntu
ID_LIKE=debian
...
UBUNTU_CODENAME=jammy
Hello!
This is due to how pandas write numpy arrays to csv. [Source](https://stackoverflow.com/questions/54753179/to-csv-saves-np-array-as-string-instead-of-as-a-list)
To fix this, you can convert them to list yourselves.
```python
df = ds.to_pandas()
df['int'] = df['int'].apply(lambda arr: list(arr))
df.to_csv(index=False, '../output/temp.csv')
```
I think it would be good if `datasets` would do the conversion itself, but it's a breaking change and I would wait for the greenlight from someone from HF. | [
-0.008384395390748978,
-0.13903416693210602,
-0.12325026094913483,
0.19654807448387146,
0.28234148025512695,
0.42689791321754456,
0.4133239984512329,
0.25605615973472595,
0.48126405477523804,
-0.1573658138513565,
0.12567222118377686,
0.4118981659412384,
0.2057298719882965,
-0.0987256988883... |
https://github.com/huggingface/datasets/issues/6777 | .Jsonl metadata not detected | @mariosasko it says metadata.csv not found
<img width="1150" alt="image" src="https://github.com/huggingface/datasets/assets/81643693/3754980c-6185-4413-88fa-b499bcdd4195">
dataset = load_dataset('/dataset',metadata.csv)
| workspace
|| source code
| dataset
| |-- images
| |-- metadata.csv
| |-- metadata.jsonl
| |-- padded_images
Example of metadata.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
| ### Describe the bug
Hi I have the following directory structure:
|--dataset
| |-- images
| |-- metadata1000.csv
| |-- metadata1000.jsonl
| |-- padded_images
Example of metadata1000.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
.
.
.
I'm trying to use dataset = load_dataset("imagefolder", data_dir='/dataset/', split='train') to load the the dataset, however it is not able to load according to the fields in the metadata1000.jsonl .
please assist to load the data properly
also getting
```
File "/workspace/train_trans_vae.py", line 1089, in <module>
print(get_metadata_patterns('/dataset/'))
File "/opt/conda/lib/python3.10/site-packages/datasets/data_files.py", line 499, in get_metadata_patterns
raise FileNotFoundError(f"The directory at {base_path} doesn't contain any metadata file") from None
FileNotFoundError: The directory at /dataset/ doesn't contain any metadata file
```
when trying
```
from datasets.data_files import get_metadata_patterns
print(get_metadata_patterns('/dataset/'))
```
### Steps to reproduce the bug
dataset Version: 2.18.0
make a similar jsonl and similar directory format
### Expected behavior
creates a dataset object with the column names, caption,image,gaussian_padded_image
### Environment info
dataset Version: 2.18.0 | 113 | .Jsonl metadata not detected
### Describe the bug
Hi I have the following directory structure:
|--dataset
| |-- images
| |-- metadata1000.csv
| |-- metadata1000.jsonl
| |-- padded_images
Example of metadata1000.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
.
.
.
I'm trying to use dataset = load_dataset("imagefolder", data_dir='/dataset/', split='train') to load the the dataset, however it is not able to load according to the fields in the metadata1000.jsonl .
please assist to load the data properly
also getting
```
File "/workspace/train_trans_vae.py", line 1089, in <module>
print(get_metadata_patterns('/dataset/'))
File "/opt/conda/lib/python3.10/site-packages/datasets/data_files.py", line 499, in get_metadata_patterns
raise FileNotFoundError(f"The directory at {base_path} doesn't contain any metadata file") from None
FileNotFoundError: The directory at /dataset/ doesn't contain any metadata file
```
when trying
```
from datasets.data_files import get_metadata_patterns
print(get_metadata_patterns('/dataset/'))
```
### Steps to reproduce the bug
dataset Version: 2.18.0
make a similar jsonl and similar directory format
### Expected behavior
creates a dataset object with the column names, caption,image,gaussian_padded_image
### Environment info
dataset Version: 2.18.0
@mariosasko it says metadata.csv not found
<img width="1150" alt="image" src="https://github.com/huggingface/datasets/assets/81643693/3754980c-6185-4413-88fa-b499bcdd4195">
dataset = load_dataset('/dataset',metadata.csv)
| workspace
|| source code
| dataset
| |-- images
| |-- metadata.csv
| |-- metadata.jsonl
| |-- padded_images
Example of metadata.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
| [
-0.04014730453491211,
0.3537839353084564,
-0.0807504653930664,
0.6944766044616699,
0.15412485599517822,
0.035916633903980255,
0.17984703183174133,
0.3550626337528229,
-0.1673462837934494,
-0.13054195046424866,
0.16933415830135345,
0.49628689885139465,
-0.2693973183631897,
-0.09884150326251... |
https://github.com/huggingface/datasets/issues/6777 | .Jsonl metadata not detected | Loading more than one image per row with `imagefolder` is not supported currently. You can subscribe to https://github.com/huggingface/datasets/issues/5760 to see when it will be.
Instead, you can load the dataset with `Dataset.from_generator`:
```python
import json
from datasets import Dataset, Value, Image, Features
def gen():
with open("./dataset/metadata.jsonl") as f:
for line in f:
line = json.loads(line)
yield {"caption": line["caption"], "image": os.path.join("./dataset", line["image"], "gaussian_padded_image": os.path.join("./dataset", line["gaussian_padded_image"]))}
features = Features({"caption": Value("string"), "image": Image(), "gaussian_padded_image": Image()})
dataset = Dataset.from_generator(gen, features=features)
```
(E.g., if you want to share this dataset on the Hub, you can call `dataset.push_to_hub(...)` afterward) | ### Describe the bug
Hi I have the following directory structure:
|--dataset
| |-- images
| |-- metadata1000.csv
| |-- metadata1000.jsonl
| |-- padded_images
Example of metadata1000.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
.
.
.
I'm trying to use dataset = load_dataset("imagefolder", data_dir='/dataset/', split='train') to load the the dataset, however it is not able to load according to the fields in the metadata1000.jsonl .
please assist to load the data properly
also getting
```
File "/workspace/train_trans_vae.py", line 1089, in <module>
print(get_metadata_patterns('/dataset/'))
File "/opt/conda/lib/python3.10/site-packages/datasets/data_files.py", line 499, in get_metadata_patterns
raise FileNotFoundError(f"The directory at {base_path} doesn't contain any metadata file") from None
FileNotFoundError: The directory at /dataset/ doesn't contain any metadata file
```
when trying
```
from datasets.data_files import get_metadata_patterns
print(get_metadata_patterns('/dataset/'))
```
### Steps to reproduce the bug
dataset Version: 2.18.0
make a similar jsonl and similar directory format
### Expected behavior
creates a dataset object with the column names, caption,image,gaussian_padded_image
### Environment info
dataset Version: 2.18.0 | 93 | .Jsonl metadata not detected
### Describe the bug
Hi I have the following directory structure:
|--dataset
| |-- images
| |-- metadata1000.csv
| |-- metadata1000.jsonl
| |-- padded_images
Example of metadata1000.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
.
.
.
I'm trying to use dataset = load_dataset("imagefolder", data_dir='/dataset/', split='train') to load the the dataset, however it is not able to load according to the fields in the metadata1000.jsonl .
please assist to load the data properly
also getting
```
File "/workspace/train_trans_vae.py", line 1089, in <module>
print(get_metadata_patterns('/dataset/'))
File "/opt/conda/lib/python3.10/site-packages/datasets/data_files.py", line 499, in get_metadata_patterns
raise FileNotFoundError(f"The directory at {base_path} doesn't contain any metadata file") from None
FileNotFoundError: The directory at /dataset/ doesn't contain any metadata file
```
when trying
```
from datasets.data_files import get_metadata_patterns
print(get_metadata_patterns('/dataset/'))
```
### Steps to reproduce the bug
dataset Version: 2.18.0
make a similar jsonl and similar directory format
### Expected behavior
creates a dataset object with the column names, caption,image,gaussian_padded_image
### Environment info
dataset Version: 2.18.0
Loading more than one image per row with `imagefolder` is not supported currently. You can subscribe to https://github.com/huggingface/datasets/issues/5760 to see when it will be.
Instead, you can load the dataset with `Dataset.from_generator`:
```python
import json
from datasets import Dataset, Value, Image, Features
def gen():
with open("./dataset/metadata.jsonl") as f:
for line in f:
line = json.loads(line)
yield {"caption": line["caption"], "image": os.path.join("./dataset", line["image"], "gaussian_padded_image": os.path.join("./dataset", line["gaussian_padded_image"]))}
features = Features({"caption": Value("string"), "image": Image(), "gaussian_padded_image": Image()})
dataset = Dataset.from_generator(gen, features=features)
```
(E.g., if you want to share this dataset on the Hub, you can call `dataset.push_to_hub(...)` afterward) | [
-0.04014730453491211,
0.3537839353084564,
-0.0807504653930664,
0.6944766044616699,
0.15412485599517822,
0.035916633903980255,
0.17984703183174133,
0.3550626337528229,
-0.1673462837934494,
-0.13054195046424866,
0.16933415830135345,
0.49628689885139465,
-0.2693973183631897,
-0.09884150326251... |
https://github.com/huggingface/datasets/issues/6777 | .Jsonl metadata not detected | hi Thanks for sharing this, Actually I was trying with a webdataset format of the data as well and it did'nt work. Could you share how i can create Dataset object from webdataset format of this data? | ### Describe the bug
Hi I have the following directory structure:
|--dataset
| |-- images
| |-- metadata1000.csv
| |-- metadata1000.jsonl
| |-- padded_images
Example of metadata1000.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
.
.
.
I'm trying to use dataset = load_dataset("imagefolder", data_dir='/dataset/', split='train') to load the the dataset, however it is not able to load according to the fields in the metadata1000.jsonl .
please assist to load the data properly
also getting
```
File "/workspace/train_trans_vae.py", line 1089, in <module>
print(get_metadata_patterns('/dataset/'))
File "/opt/conda/lib/python3.10/site-packages/datasets/data_files.py", line 499, in get_metadata_patterns
raise FileNotFoundError(f"The directory at {base_path} doesn't contain any metadata file") from None
FileNotFoundError: The directory at /dataset/ doesn't contain any metadata file
```
when trying
```
from datasets.data_files import get_metadata_patterns
print(get_metadata_patterns('/dataset/'))
```
### Steps to reproduce the bug
dataset Version: 2.18.0
make a similar jsonl and similar directory format
### Expected behavior
creates a dataset object with the column names, caption,image,gaussian_padded_image
### Environment info
dataset Version: 2.18.0 | 37 | .Jsonl metadata not detected
### Describe the bug
Hi I have the following directory structure:
|--dataset
| |-- images
| |-- metadata1000.csv
| |-- metadata1000.jsonl
| |-- padded_images
Example of metadata1000.jsonl file
{"caption": "a drawing depicts a full shot of a black t-shirt with a triangular pattern on the front there is a white label on the left side of the triangle", "image": "images/212734.png", "gaussian_padded_image": "padded_images/p_212734.png"}
{"caption": "an eye-level full shot of a large elephant and a baby elephant standing in a watering hole on the left side is a small elephant with its head turned to the right of dry land, trees, and bushes", "image": "images/212735.png", "gaussian_padded_image": "padded_images/p_212735.png"}
.
.
.
I'm trying to use dataset = load_dataset("imagefolder", data_dir='/dataset/', split='train') to load the the dataset, however it is not able to load according to the fields in the metadata1000.jsonl .
please assist to load the data properly
also getting
```
File "/workspace/train_trans_vae.py", line 1089, in <module>
print(get_metadata_patterns('/dataset/'))
File "/opt/conda/lib/python3.10/site-packages/datasets/data_files.py", line 499, in get_metadata_patterns
raise FileNotFoundError(f"The directory at {base_path} doesn't contain any metadata file") from None
FileNotFoundError: The directory at /dataset/ doesn't contain any metadata file
```
when trying
```
from datasets.data_files import get_metadata_patterns
print(get_metadata_patterns('/dataset/'))
```
### Steps to reproduce the bug
dataset Version: 2.18.0
make a similar jsonl and similar directory format
### Expected behavior
creates a dataset object with the column names, caption,image,gaussian_padded_image
### Environment info
dataset Version: 2.18.0
hi Thanks for sharing this, Actually I was trying with a webdataset format of the data as well and it did'nt work. Could you share how i can create Dataset object from webdataset format of this data? | [
-0.04014730453491211,
0.3537839353084564,
-0.0807504653930664,
0.6944766044616699,
0.15412485599517822,
0.035916633903980255,
0.17984703183174133,
0.3550626337528229,
-0.1673462837934494,
-0.13054195046424866,
0.16933415830135345,
0.49628689885139465,
-0.2693973183631897,
-0.09884150326251... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.