html_url stringlengths 51 51 | comments stringlengths 67 24.7k | title stringlengths 6 280 | body stringlengths 51 36.2k | comment_length int64 16 1.45k | text stringlengths 190 38.3k | embeddings list |
|---|---|---|---|---|---|---|
https://github.com/huggingface/datasets/issues/3909 | Thanks!
And sorry for posting this problem in what turned on to be an unrelated thread.
I rewrote the code, and the model works. The WER is 0.137 however, so I'm not sure if I have missed a step. I will look further into that at a later point. The transcriptions look good through manual inspection.
The rewritt... | Error loading file audio when downloading the Common Voice dataset directly from the Hub | ## Describe the bug
When loading the Common_Voice dataset, by downloading it directly from the Hugging Face hub, some files can not be opened.
## Steps to reproduce the bug
```python
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2... | 130 | Error loading file audio when downloading the Common Voice dataset directly from the Hub
## Describe the bug
When loading the Common_Voice dataset, by downloading it directly from the Hugging Face hub, some files can not be opened.
## Steps to reproduce the bug
```python
import torch
import torchaudio
from d... | [
-0.2770319879055023,
-0.5579100847244263,
0.09589679539203644,
0.43422484397888184,
0.09727855026721954,
0.027911260724067688,
0.15836185216903687,
0.3387337327003479,
0.221358060836792,
0.15522468090057373,
-0.6184943914413452,
0.18397507071495056,
-0.052941083908081055,
-0.34789967536926... |
https://github.com/huggingface/datasets/issues/3909 | I think the issue comes from the fact that you set `batched=False` while `map_to_pred` still returns a list of strings for "transcription". You can fix it by adding `[0]` at the end of this line to get the string:
```python
batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True)[0]
``` | Error loading file audio when downloading the Common Voice dataset directly from the Hub | ## Describe the bug
When loading the Common_Voice dataset, by downloading it directly from the Hugging Face hub, some files can not be opened.
## Steps to reproduce the bug
```python
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2... | 45 | Error loading file audio when downloading the Common Voice dataset directly from the Hub
## Describe the bug
When loading the Common_Voice dataset, by downloading it directly from the Hugging Face hub, some files can not be opened.
## Steps to reproduce the bug
```python
import torch
import torchaudio
from d... | [
-0.2770319879055023,
-0.5579100847244263,
0.09589679539203644,
0.43422484397888184,
0.09727855026721954,
0.027911260724067688,
0.15836185216903687,
0.3387337327003479,
0.221358060836792,
0.15522468090057373,
-0.6184943914413452,
0.18397507071495056,
-0.052941083908081055,
-0.34789967536926... |
https://github.com/huggingface/datasets/issues/3909 | We no longer use `torchaudio` for decoding MP3 files, and the problem with model cards has been addressed, so I'm closing this issue. | Error loading file audio when downloading the Common Voice dataset directly from the Hub | ## Describe the bug
When loading the Common_Voice dataset, by downloading it directly from the Hugging Face hub, some files can not be opened.
## Steps to reproduce the bug
```python
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2... | 23 | Error loading file audio when downloading the Common Voice dataset directly from the Hub
## Describe the bug
When loading the Common_Voice dataset, by downloading it directly from the Hugging Face hub, some files can not be opened.
## Steps to reproduce the bug
```python
import torch
import torchaudio
from d... | [
-0.2770319879055023,
-0.5579100847244263,
0.09589679539203644,
0.43422484397888184,
0.09727855026721954,
0.027911260724067688,
0.15836185216903687,
0.3387337327003479,
0.221358060836792,
0.15522468090057373,
-0.6184943914413452,
0.18397507071495056,
-0.052941083908081055,
-0.34789967536926... |
https://github.com/huggingface/datasets/issues/3906 | Hi @kolk, thanks for reporting.
Indeed, Google Drive service recently changed their service and we had to add a fix to our library to cope with that change:
- #3787
We just made patch release last week: 1.18.4 https://github.com/huggingface/datasets/releases/tag/1.18.4
Please, feel free to update your local ... | NonMatchingChecksumError on Spider dataset | ## Describe the bug
Failure to generate dataset ```spider``` because of checksums error for dataset source files.
## Steps to reproduce the bug
```
from datasets import load_dataset
spider = load_dataset("spider")
```
## Expected results
Checksums should match for files from url ['https://drive.google.com... | 60 | NonMatchingChecksumError on Spider dataset
## Describe the bug
Failure to generate dataset ```spider``` because of checksums error for dataset source files.
## Steps to reproduce the bug
```
from datasets import load_dataset
spider = load_dataset("spider")
```
## Expected results
Checksums should match... | [
-0.16796563565731049,
0.025489553809165955,
-0.08636097609996796,
0.29547420144081116,
0.3215497136116028,
-0.11661061644554138,
0.13645067811012268,
0.4342500567436218,
0.1861148476600647,
0.25885438919067383,
-0.26600900292396545,
0.01770975813269615,
-0.036636218428611755,
0.10005322843... |
https://github.com/huggingface/datasets/issues/3904 | Thanks for reporting, @omarespejel.
I'm sorry but I can't reproduce the issue: the loading of the dataset works perfecto for me and I can reach the data URL: https://data.deepai.org/conll2003.zip
Might it be due to a temporary problem in the data owner site (https://data.deepai.org/) that is fixed now?
Could you... | CONLL2003 Dataset not available | ## Describe the bug
[CONLL2003](https://huggingface.co/datasets/conll2003) Dataset can no longer reach 'https://data.deepai.org/conll2003.zip'
## Steps to reproduce the bug
```python
from datasets impo... | 61 | CONLL2003 Dataset not available
## Describe the bug
[CONLL2003](https://huggingface.co/datasets/conll2003) Dataset can no longer reach 'https://data.deepai.org/conll2003.zip'
## Steps to reproduce the ... | [
-0.19115188717842102,
0.015069171786308289,
-0.04346461966633797,
0.33827441930770874,
0.19229693710803986,
0.003913938999176025,
0.2465219497680664,
0.11542646586894989,
-0.3758273720741272,
0.14859795570373535,
-0.0861063078045845,
0.24523372948169708,
0.4535979628562927,
0.1852292269468... |
https://github.com/huggingface/datasets/issues/3904 | I am getting the same issue. I use google colab with CPU.
The code I used is exactly the same as described above.
```
from datasets import load_dataset
dataset = load_dataset("conll2003")
```
The produced error:
 Dataset can no longer reach 'https://data.deepai.org/conll2003.zip'
## Steps to reproduce the bug
```python
from datasets impo... | 193 | CONLL2003 Dataset not available
## Describe the bug
[CONLL2003](https://huggingface.co/datasets/conll2003) Dataset can no longer reach 'https://data.deepai.org/conll2003.zip'
## Steps to reproduce the ... | [
-0.23122775554656982,
0.10169973969459534,
0.019431695342063904,
0.3545541763305664,
0.08920589834451675,
0.0359596312046051,
0.2744438946247101,
0.08043457567691803,
-0.29682260751724243,
0.26328158378601074,
-0.06898228824138641,
0.32954832911491394,
0.4328652024269104,
0.256467491388320... |
https://github.com/huggingface/datasets/issues/3902 | Update: `"python3 -c "from from datasets import Dataset, DatasetDict"` works, but not if I import without the `python3 -c` | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils' | ## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeError Traceback (most recent call last)
<ipython-input-37-c8c... | 19 | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils'
## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeE... | [
-0.3171517252922058,
-0.09776317328214645,
-0.060437314212322235,
0.2508160173892975,
0.5410328507423401,
0.0515427440404892,
0.23068232834339142,
0.17841292917728424,
0.14518390595912933,
-0.04065912961959839,
-0.0914040058851242,
0.2070273905992508,
-0.08499455451965332,
0.04220127314329... |
https://github.com/huggingface/datasets/issues/3902 | Hi @arunasank, thanks for reporting.
It seems that this can be caused because you are using an old version of `fsspec`: the reason why it works if you run `python3` seems to be that `python3` runs in a Python virtual env (with an updated version of `fsspec`); whereas the error arises when you run the import from oth... | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils' | ## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeError Traceback (most recent call last)
<ipython-input-37-c8c... | 88 | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils'
## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeE... | [
-0.3171517252922058,
-0.09776317328214645,
-0.060437314212322235,
0.2508160173892975,
0.5410328507423401,
0.0515427440404892,
0.23068232834339142,
0.17841292917728424,
0.14518390595912933,
-0.04065912961959839,
-0.0914040058851242,
0.2070273905992508,
-0.08499455451965332,
0.04220127314329... |
https://github.com/huggingface/datasets/issues/3902 | from lightgbm import LGBMModel,LGBMClassifier, plot_importance
after importing lib getting (partially initialized module 'fsspec' has no attribute 'utils' (most likely due to a circular import) error, can help me | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils' | ## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeError Traceback (most recent call last)
<ipython-input-37-c8c... | 28 | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils'
## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeE... | [
-0.3171517252922058,
-0.09776317328214645,
-0.060437314212322235,
0.2508160173892975,
0.5410328507423401,
0.0515427440404892,
0.23068232834339142,
0.17841292917728424,
0.14518390595912933,
-0.04065912961959839,
-0.0914040058851242,
0.2070273905992508,
-0.08499455451965332,
0.04220127314329... |
https://github.com/huggingface/datasets/issues/3902 | @deepakmahtha I think you are not using `datasets`: this is the GitHub repository of Hugging Face Datasets.
If you are using `lightgbm`, you should report the issue to their repository instead.
Anyway, we have proposed a possible fix just in a comment above: to update fsspec.
https://github.com/huggingface/datas... | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils' | ## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeError Traceback (most recent call last)
<ipython-input-37-c8c... | 47 | Can't import datasets: partially initialized module 'fsspec' has no attribute 'utils'
## Describe the bug
Unable to import datasets
## Steps to reproduce the bug
```python
from datasets import Dataset, DatasetDict
```
## Expected results
The import works without errors
## Actual results
```
AttributeE... | [
-0.3171517252922058,
-0.09776317328214645,
-0.060437314212322235,
0.2508160173892975,
0.5410328507423401,
0.0515427440404892,
0.23068232834339142,
0.17841292917728424,
0.14518390595912933,
-0.04065912961959839,
-0.0914040058851242,
0.2070273905992508,
-0.08499455451965332,
0.04220127314329... |
https://github.com/huggingface/datasets/issues/3901 | It seems to have been fixed:
<img width="1534" alt="Capture d’écran 2022-04-12 à 14 10 07" src="https://user-images.githubusercontent.com/1676121/162959599-6b7fef7c-8411-4e03-8f00-90040a658079.png">
| Dataset viewer issue for IndicParaphrase- the preview doesn't show | ## Dataset viewer issue for '*IndicParaphrase*'
**Link:** *[IndicParaphrase](https://huggingface.co/datasets/ai4bharat/IndicParaphrase/viewer/hi/validation)*
*The preview of the dataset doesn't come up.
The error on the console is:
Status code: 400
Exception: FileNotFoundError
Message: [Errno 2] ... | 16 | Dataset viewer issue for IndicParaphrase- the preview doesn't show
## Dataset viewer issue for '*IndicParaphrase*'
**Link:** *[IndicParaphrase](https://huggingface.co/datasets/ai4bharat/IndicParaphrase/viewer/hi/validation)*
*The preview of the dataset doesn't come up.
The error on the console is:
Status cod... | [
-0.23629122972488403,
0.2801050841808319,
0.005715500563383102,
0.17628026008605957,
-0.04072480648756027,
0.15107932686805725,
0.22422856092453003,
0.252490371465683,
0.014741919934749603,
0.10672345757484436,
0.10246384143829346,
0.1366589516401291,
-0.024567164480686188,
-0.212866038084... |
https://github.com/huggingface/datasets/issues/3896 | `datasets` 1.18.4 fixes the issue when you load the dataset with `load_dataset`.
When loading in streaming mode, the fix is indeed on https://github.com/huggingface/datasets/pull/3843 which will be merged soon :) | Missing google file for `multi_news` dataset | ## Dataset viewer issue for '*multi_news*'
**Link:** https://huggingface.co/datasets/multi_news
```
Server error
Status code: 400
Exception: FileNotFoundError
Message: https://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C/multi-news-original/train.src
```
Am I the ... | 29 | Missing google file for `multi_news` dataset
## Dataset viewer issue for '*multi_news*'
**Link:** https://huggingface.co/datasets/multi_news
```
Server error
Status code: 400
Exception: FileNotFoundError
Message: https://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C/... | [
-0.2699742913246155,
-0.045839667320251465,
0.049731962382793427,
0.1420218050479889,
0.12764476239681244,
0.24816235899925232,
0.19342267513275146,
0.2745968699455261,
-0.02155642956495285,
0.1579298973083496,
0.08163691312074661,
0.07922878861427307,
-0.3635447025299072,
0.33495387434959... |
https://github.com/huggingface/datasets/issues/3896 | That is. The PR #3843 was just opened a bit later we had made our 1.18.4 patch release...
Once merged, that will fix this issue. | Missing google file for `multi_news` dataset | ## Dataset viewer issue for '*multi_news*'
**Link:** https://huggingface.co/datasets/multi_news
```
Server error
Status code: 400
Exception: FileNotFoundError
Message: https://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C/multi-news-original/train.src
```
Am I the ... | 25 | Missing google file for `multi_news` dataset
## Dataset viewer issue for '*multi_news*'
**Link:** https://huggingface.co/datasets/multi_news
```
Server error
Status code: 400
Exception: FileNotFoundError
Message: https://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C/... | [
-0.18722456693649292,
-0.004876613616943359,
0.035689812153577805,
0.18284493684768677,
0.06084601581096649,
0.2791559100151062,
0.23320133984088898,
0.2045084685087204,
-0.036696743220090866,
0.17451825737953186,
0.16457796096801758,
0.019153684377670288,
-0.36171260476112366,
0.257810950... |
https://github.com/huggingface/datasets/issues/3896 | OK. Should fix the viewer for 50 datasets
<img width="148" alt="Capture d’écran 2022-03-14 à 11 51 02" src="https://user-images.githubusercontent.com/1676121/158157853-6c544a47-2d6d-4ac4-964a-6f10951ec36b.png">
| Missing google file for `multi_news` dataset | ## Dataset viewer issue for '*multi_news*'
**Link:** https://huggingface.co/datasets/multi_news
```
Server error
Status code: 400
Exception: FileNotFoundError
Message: https://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C/multi-news-original/train.src
```
Am I the ... | 18 | Missing google file for `multi_news` dataset
## Dataset viewer issue for '*multi_news*'
**Link:** https://huggingface.co/datasets/multi_news
```
Server error
Status code: 400
Exception: FileNotFoundError
Message: https://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C/... | [
-0.19339728355407715,
0.1477288007736206,
0.01884012669324875,
0.2966786026954651,
0.05062606558203697,
0.3109288811683655,
0.29413989186286926,
0.18849103152751923,
-0.0731482282280922,
0.12405822426080704,
0.14978766441345215,
-0.08645810186862946,
-0.34820035099983215,
0.235973745584487... |
https://github.com/huggingface/datasets/issues/3889 | Hi ! A pull request is open to fix the dataset, we'll release a patch soon with a new release of `datasets` :) | Cannot load beans dataset (Couldn't reach the dataset) | ## Describe the bug
The beans dataset is unavailable to download.
## Steps to reproduce the bug
```python
from datasets import load_dataset
ds = load_dataset('beans')
```
## Expected results
The dataset would be downloaded with no issue.
## Actual results
```
ConnectionError: Couldn't reach https://s... | 23 | Cannot load beans dataset (Couldn't reach the dataset)
## Describe the bug
The beans dataset is unavailable to download.
## Steps to reproduce the bug
```python
from datasets import load_dataset
ds = load_dataset('beans')
```
## Expected results
The dataset would be downloaded with no issue.
## Actual ... | [
-0.4721035063266754,
0.2937532365322113,
-0.16326218843460083,
0.4070338308811188,
0.3040819466114044,
0.002488464117050171,
0.1432952880859375,
0.007078586146235466,
-0.04815211519598961,
0.28478312492370605,
-0.06385957449674606,
0.36830934882164,
-0.01640760339796543,
0.3610323071479797... |
https://github.com/huggingface/datasets/issues/3888 | @lhoestq so in order to address what’s not completed in this issue, do you think it makes sense to add a param `features` to `IterableDataset.map` so that the output features right after the `map` are defined there? | IterableDataset columns and feature types | Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's inside your dataset without having to manua... | 37 | IterableDataset columns and feature types
Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's... | [
-0.3458293080329895,
-0.03658224642276764,
-0.0944572389125824,
-0.19148167967796326,
0.3538818061351776,
-0.06071500480175018,
0.4520319104194641,
0.1834230273962021,
0.18687254190444946,
0.12598161399364471,
0.20759445428848267,
0.18288588523864746,
-0.293923944234848,
0.3534747958183288... |
https://github.com/huggingface/datasets/issues/3888 | @lhoestq cool then if you agree I can work on that! I’ll also update the docs accordingly once done, thanks! | IterableDataset columns and feature types | Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's inside your dataset without having to manua... | 20 | IterableDataset columns and feature types
Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's... | [
-0.4446369707584381,
-0.07640551030635834,
-0.1001342236995697,
-0.022534746676683426,
0.2964310944080353,
0.06714220345020294,
0.3907925486564636,
0.19237571954727173,
0.33650103211402893,
0.2634657621383667,
0.03578311949968338,
0.13149875402450562,
-0.23470690846443176,
0.41665521264076... |
https://github.com/huggingface/datasets/issues/3888 | I've already started with a PR as a draft @lhoestq, should we also try to look for a way to explicitly request pre-fetching right after a map operation is applied, so that the features are inferred if the user says explicitly so? Thanks! | IterableDataset columns and feature types | Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's inside your dataset without having to manua... | 43 | IterableDataset columns and feature types
Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's... | [
-0.3794460892677307,
-0.11217371374368668,
-0.048046305775642395,
-0.0016449689865112305,
0.240922212600708,
-0.03550286591053009,
0.32073602080345154,
0.16656188666820526,
0.32377660274505615,
0.2732170522212982,
0.13899248838424683,
0.012926261872053146,
-0.15598149597644806,
0.365395247... |
https://github.com/huggingface/datasets/issues/3888 | > should we also try to look for a way to explicitly request pre-fetching right after a map operation is applied, so that the features are inferred if the user says explicitly so?
Right now one can use `ds = ds._resolve_features()` do to so. It can be used after `map` or `load_dataset` if the features are not known.... | IterableDataset columns and feature types | Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's inside your dataset without having to manua... | 66 | IterableDataset columns and feature types
Right now, an IterableDataset (e.g. when streaming a dataset) doesn't require to know the list of columns it contains, nor their types: `my_iterable_dataset.features` may be `None`
However it's often interesting to know the column types and types. This helps knowing what's... | [
-0.3393740952014923,
-0.12804003059864044,
-0.05538404360413551,
-0.0033453013747930527,
0.22570408880710602,
-0.06992124021053314,
0.36740702390670776,
0.17384161055088043,
0.3320353925228119,
0.29572010040283203,
0.07677540928125381,
0.048531435430049896,
-0.15181124210357666,
0.35509335... |
https://github.com/huggingface/datasets/issues/3881 | Hi @INF800,
Please note that the `imagefolder` feature enhancement was just recently merged to our master branch (https://github.com/huggingface/datasets/commit/207be676bffe9d164740a41a883af6125edef135), but has not yet been released.
We are planning to make the 2.0 release of our library in the coming days and t... | How to use Image folder | Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError Trace... | 140 | How to use Image folder
Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError ... | [
-0.2858949303627014,
0.07919466495513916,
-0.16414381563663483,
0.5192500352859497,
0.39786967635154724,
0.19424109160900116,
0.3166288435459137,
0.14290878176689148,
0.045584723353385925,
0.1703680455684662,
0.07837354391813278,
-0.013808656483888626,
-0.20191241800785065,
0.0991081073880... |
https://github.com/huggingface/datasets/issues/3881 | Hey @albertvillanova. Does this load entire dataset in memory? Because I am facing huge trouble with loading very big datasets (OOM errors) | How to use Image folder | Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError Trace... | 22 | How to use Image folder
Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError ... | [
-0.2858949303627014,
0.07919466495513916,
-0.16414381563663483,
0.5192500352859497,
0.39786967635154724,
0.19424109160900116,
0.3166288435459137,
0.14290878176689148,
0.045584723353385925,
0.1703680455684662,
0.07837354391813278,
-0.013808656483888626,
-0.20191241800785065,
0.0991081073880... |
https://github.com/huggingface/datasets/issues/3881 | Can you provide the error stack trace? The loader only stores the `data_files` dict, which can get big after globbing. Then, the OOM error would mean you don't have enough memory to keep all the paths to the image files. You can circumvent this by generating an archive and loading the dataset from there. Maybe we can o... | How to use Image folder | Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError Trace... | 73 | How to use Image folder
Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError ... | [
-0.2858949303627014,
0.07919466495513916,
-0.16414381563663483,
0.5192500352859497,
0.39786967635154724,
0.19424109160900116,
0.3166288435459137,
0.14290878176689148,
0.045584723353385925,
0.1703680455684662,
0.07837354391813278,
-0.013808656483888626,
-0.20191241800785065,
0.0991081073880... |
https://github.com/huggingface/datasets/issues/3881 | Hey, memory error is resolved. It was fluke.
But there is another issue. Currently `load_dataset("imagefolder", data_dir="./path/to/train",)` takes only `train` as arg to `split` parameter.
I am creating vaildation dataset using
```
ds_valid = datasets.DatasetDict(valid=load_dataset("imagefolder", data_dir=".... | How to use Image folder | Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError Trace... | 36 | How to use Image folder
Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError ... | [
-0.2858949303627014,
0.07919466495513916,
-0.16414381563663483,
0.5192500352859497,
0.39786967635154724,
0.19424109160900116,
0.3166288435459137,
0.14290878176689148,
0.045584723353385925,
0.1703680455684662,
0.07837354391813278,
-0.013808656483888626,
-0.20191241800785065,
0.0991081073880... |
https://github.com/huggingface/datasets/issues/3881 | `data_dir="path/to/folder"` is a shorthand syntax fox `data_files={"train": "path/to/folder/**"}`, so use `data_files` in that case instead:
```python
ds = load_dataset("imagefolder", data_files={"train": "path/to/train/**", "test": "path/to/test/**", "valid": "path/to/valid/**"})
``` | How to use Image folder | Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError Trace... | 26 | How to use Image folder
Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError ... | [
-0.2858949303627014,
0.07919466495513916,
-0.16414381563663483,
0.5192500352859497,
0.39786967635154724,
0.19424109160900116,
0.3166288435459137,
0.14290878176689148,
0.045584723353385925,
0.1703680455684662,
0.07837354391813278,
-0.013808656483888626,
-0.20191241800785065,
0.0991081073880... |
https://github.com/huggingface/datasets/issues/3881 | And there was another issue. I loaded black and white images (jpeg file). Using load dataset. It reads it as PIL jpeg data format. But instead of converting it into 3 channel tensor, input to collator function is coming as a single channel tensor. | How to use Image folder | Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError Trace... | 44 | How to use Image folder
Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError ... | [
-0.2858949303627014,
0.07919466495513916,
-0.16414381563663483,
0.5192500352859497,
0.39786967635154724,
0.19424109160900116,
0.3166288435459137,
0.14290878176689148,
0.045584723353385925,
0.1703680455684662,
0.07837354391813278,
-0.013808656483888626,
-0.20191241800785065,
0.0991081073880... |
https://github.com/huggingface/datasets/issues/3881 | We don't apply any additional preprocessing on top of `PIL.Image.open(image_file)`, so you need to do the conversion yourself:
```python
def to_rgb(batch):
batch["image"] = [img.convert("RGB") for img in batch["image"]]
return batch
ds_rgb = ds.map(to_rgb, batched=True)
```
Please use our Forum for... | How to use Image folder | Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError Trace... | 47 | How to use Image folder
Ran this code
```
load_dataset("imagefolder", data_dir="./my-dataset")
```
`https://raw.githubusercontent.com/huggingface/datasets/master/datasets/imagefolder/imagefolder.py` missing
```
---------------------------------------------------------------------------
FileNotFoundError ... | [
-0.2858949303627014,
0.07919466495513916,
-0.16414381563663483,
0.5192500352859497,
0.39786967635154724,
0.19424109160900116,
0.3166288435459137,
0.14290878176689148,
0.045584723353385925,
0.1703680455684662,
0.07837354391813278,
-0.013808656483888626,
-0.20191241800785065,
0.0991081073880... |
https://github.com/huggingface/datasets/issues/3872 | yes but is there any way you could try pushing with `git` command line directly instead of `push_to_hub`? | HTTP error 504 Server Error: Gateway Time-out | I am trying to push a large dataset(450000+) records with the help of `push_to_hub()`
While pushing, it gives some error like this.
```
Traceback (most recent call last):
File "data_split_speech.py", line 159, in <module>
data_new_2.push_to_hub("user-name/dataset-name",private=True)
File "/opt/conda/lib... | 18 | HTTP error 504 Server Error: Gateway Time-out
I am trying to push a large dataset(450000+) records with the help of `push_to_hub()`
While pushing, it gives some error like this.
```
Traceback (most recent call last):
File "data_split_speech.py", line 159, in <module>
data_new_2.push_to_hub("user-name/dat... | [
-0.4594692587852478,
-0.34232935309410095,
-0.06529292464256287,
0.1371452808380127,
0.14353936910629272,
-0.05058155208826065,
-0.01949688233435154,
0.4111798405647278,
-0.3377400040626526,
-0.2412683516740799,
0.04786679893732071,
0.10722547769546509,
0.16047760844230652,
0.4615727663040... |
https://github.com/huggingface/datasets/issues/3872 | Okay. I didnt saved the dataset to my local machine. So, I processed the dataset and pushed it directly to the hub. I think I should try saving those dataset to my local machine by `save_to_disk` and then push it with git command line | HTTP error 504 Server Error: Gateway Time-out | I am trying to push a large dataset(450000+) records with the help of `push_to_hub()`
While pushing, it gives some error like this.
```
Traceback (most recent call last):
File "data_split_speech.py", line 159, in <module>
data_new_2.push_to_hub("user-name/dataset-name",private=True)
File "/opt/conda/lib... | 44 | HTTP error 504 Server Error: Gateway Time-out
I am trying to push a large dataset(450000+) records with the help of `push_to_hub()`
While pushing, it gives some error like this.
```
Traceback (most recent call last):
File "data_split_speech.py", line 159, in <module>
data_new_2.push_to_hub("user-name/dat... | [
-0.3648715019226074,
-0.41700369119644165,
-0.048948489129543304,
0.2769414186477661,
0.09378406405448914,
0.045876339077949524,
-0.06348749995231628,
0.31184715032577515,
-0.36931079626083374,
-0.25901493430137634,
-0.020844314247369766,
0.03183073177933693,
0.1456531286239624,
0.47873413... |
https://github.com/huggingface/datasets/issues/3872 | `push_to_hub` is the preferred way of uploading a dataset to the Hub, which can then be reloaded with `load_dataset`. Feel free to try again and see if the server is working as expected now. Maybe we can add a retry mechanism in the meantime to workaround 504 errors.
Regarding `save_to_disk`, this must only be used ... | HTTP error 504 Server Error: Gateway Time-out | I am trying to push a large dataset(450000+) records with the help of `push_to_hub()`
While pushing, it gives some error like this.
```
Traceback (most recent call last):
File "data_split_speech.py", line 159, in <module>
data_new_2.push_to_hub("user-name/dataset-name",private=True)
File "/opt/conda/lib... | 93 | HTTP error 504 Server Error: Gateway Time-out
I am trying to push a large dataset(450000+) records with the help of `push_to_hub()`
While pushing, it gives some error like this.
```
Traceback (most recent call last):
File "data_split_speech.py", line 159, in <module>
data_new_2.push_to_hub("user-name/dat... | [
-0.3772861361503601,
-0.44991645216941833,
-0.052841074764728546,
0.19174335896968842,
0.07236990332603455,
-0.020948342978954315,
-0.04381465166807175,
0.3406636714935303,
-0.31234318017959595,
-0.29271405935287476,
0.04636349529027939,
0.10223065316677094,
0.2157941609621048,
0.492305785... |
https://github.com/huggingface/datasets/issues/3869 | Hi @omarespejel! I think the philosophy for `datasets` issues is to create concrete issues with proposals to add a specific, individual dataset rather than umbrella issues for things such as datasets for a language, since we could end up with hundreds of issues (one per language). I see NILC - USP has many datasets, I ... | Making the Hub the place for datasets in Portuguese | Let's make Hugging Face Datasets the central hub for datasets in Portuguese :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the Portuguese speaking community.
What are some datasets in Portuguese worth ... | 114 | Making the Hub the place for datasets in Portuguese
Let's make Hugging Face Datasets the central hub for datasets in Portuguese :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the Portuguese speaking co... | [
-0.13159149885177612,
-0.16614000499248505,
-0.10737138986587524,
0.14891739189624786,
0.1300487220287323,
0.07393716275691986,
-0.06666788458824158,
0.13297052681446075,
-0.01103149726986885,
0.12169499695301056,
-0.16813749074935913,
-0.1752084493637085,
-0.2794049084186554,
0.3597373664... |
https://github.com/huggingface/datasets/issues/3861 | To follow up on this: the cased and uncased versions actually contain different content, and the cased one is easier since it contains a Summary of the Invention in the input.
See the paper describing the issue here:
https://aclanthology.org/2022.gem-1.34/ | big_patent cased version | Hi! I am interested in working with the big_patent dataset.
In Tensorflow, there are a number of versions of the dataset:
- 1.0.0 : lower cased tokenized words
- 2.0.0 : Update to use cased raw strings
- 2.1.2 (default): Fix update to cased raw strings.
The version in the huggingface `datasets` library is th... | 39 | big_patent cased version
Hi! I am interested in working with the big_patent dataset.
In Tensorflow, there are a number of versions of the dataset:
- 1.0.0 : lower cased tokenized words
- 2.0.0 : Update to use cased raw strings
- 2.1.2 (default): Fix update to cased raw strings.
The version in the huggingfa... | [
-0.2963559031486511,
-0.3385939598083496,
-0.03500346839427948,
0.17709405720233917,
0.016487713903188705,
0.10124140977859497,
0.14578723907470703,
0.24699369072914124,
0.022616222500801086,
0.0014171674847602844,
0.06292086839675903,
-0.2668644189834595,
-0.32085585594177246,
0.248805761... |
https://github.com/huggingface/datasets/issues/3861 | Thanks for proposing the addition of the cased version of this dataset and for pinging again recently.
I have just merged a PR that adds the cased version: https://huggingface.co/datasets/big_patent/discussions/3
The cased version (2.1.2) is the default one:
```python
ds = load_dataset("big_patent", "all")
```... | big_patent cased version | Hi! I am interested in working with the big_patent dataset.
In Tensorflow, there are a number of versions of the dataset:
- 1.0.0 : lower cased tokenized words
- 2.0.0 : Update to use cased raw strings
- 2.1.2 (default): Fix update to cased raw strings.
The version in the huggingface `datasets` library is th... | 68 | big_patent cased version
Hi! I am interested in working with the big_patent dataset.
In Tensorflow, there are a number of versions of the dataset:
- 1.0.0 : lower cased tokenized words
- 2.0.0 : Update to use cased raw strings
- 2.1.2 (default): Fix update to cased raw strings.
The version in the huggingfa... | [
-0.291733980178833,
-0.29248517751693726,
-0.03984261304140091,
0.06837628781795502,
0.019045963883399963,
0.16050125658512115,
0.15711264312267303,
0.25118014216423035,
-0.08451376110315323,
-0.03758114576339722,
-0.0021421127021312714,
-0.1731712818145752,
-0.32748374342918396,
0.2096539... |
https://github.com/huggingface/datasets/issues/3859 | Hi @slvcsl, thanks for reporting.
Yesterday we just made a patch release of our `datasets` library that fixes this issue: version 1.18.4.
https://pypi.org/project/datasets/#history
Please, feel free to update `datasets` library to the latest version:
```shell
pip install -U datasets
```
And then you should ... | Unable to dowload big_patent (FileNotFoundError) | ## Describe the bug
I am trying to download some splits of the big_patent dataset, using the following code:
`ds = load_dataset("big_patent", "g", split="validation", download_mode="force_redownload")
`
However, this leads to a FileNotFoundError.
FileNotFoundError Traceback (most recent... | 115 | Unable to dowload big_patent (FileNotFoundError)
## Describe the bug
I am trying to download some splits of the big_patent dataset, using the following code:
`ds = load_dataset("big_patent", "g", split="validation", download_mode="force_redownload")
`
However, this leads to a FileNotFoundError.
FileNotFoundE... | [
-0.4091827869415283,
-0.12928354740142822,
0.06419161707162857,
0.30275535583496094,
0.050707608461380005,
0.025050222873687744,
0.42450934648513794,
0.5200849175453186,
0.5669476389884949,
-0.12240174412727356,
-0.27950868010520935,
0.014819733798503876,
-0.012764351442456245,
0.091613344... |
https://github.com/huggingface/datasets/issues/3857 | I agree using `glob.glob` alone is bad practice because it's not deterministic. Using `sorted` is a nice solution.
Note that the `xglob` function you are referring to in the `streaming_download_manager.py` code just extends `glob.glob` for URLs - we don't change its behavior. That's why it has no `sorted()` | Order of dataset changes due to glob.glob. | ## Describe the bug
After discussion with @lhoestq, just want to mention here that `glob.glob(...)` should always be used in combination with `sorted(...)` to make sure the list of files returned by `glob.glob(...)` doesn't change depending on the OS system.
There are currently multiple datasets that use `glob.g... | 48 | Order of dataset changes due to glob.glob.
## Describe the bug
After discussion with @lhoestq, just want to mention here that `glob.glob(...)` should always be used in combination with `sorted(...)` to make sure the list of files returned by `glob.glob(...)` doesn't change depending on the OS system.
There are... | [
-0.0024278834462165833,
-0.3010297417640686,
-0.11895685642957687,
0.10309386998414993,
0.18960227072238922,
-0.30356675386428833,
0.24733737111091614,
-0.09922150522470474,
0.09327240288257599,
0.15001609921455383,
-0.2698245346546173,
-0.02342502400279045,
0.03565174713730812,
0.08285888... |
https://github.com/huggingface/datasets/issues/3855 | We raise the error “ FileNotFoundError: can’t find the dataset” mainly to follow best practice in security (otherwise users could be able to guess what private repositories users/orgs may have)
We can indeed reformulate this and add the "If this is a private repository,..." part ! | Bad error message when loading private dataset | ## Describe the bug
A pretty common behavior of an interaction between the Hub and datasets is the following.
An organization adds a dataset in private mode and wants to load it afterward.
```python
from transformers import load_dataset
ds = load_dataset("NewT5/dummy_data", "dummy")
```
This command th... | 46 | Bad error message when loading private dataset
## Describe the bug
A pretty common behavior of an interaction between the Hub and datasets is the following.
An organization adds a dataset in private mode and wants to load it afterward.
```python
from transformers import load_dataset
ds = load_dataset("Ne... | [
-0.1339561939239502,
-0.02722834050655365,
0.07090628892183304,
0.5178712606430054,
0.14277499914169312,
0.016936257481575012,
0.5494260787963867,
0.1765478253364563,
0.21004341542720795,
0.14558136463165283,
-0.4457780420780182,
0.06750939041376114,
-0.030568299815058708,
-0.0527618974447... |
https://github.com/huggingface/datasets/issues/3854 | Hi @amanjaiswal777,
First note that the dataset you are trying to load is deprecated: it was the Common Voice dataset release as of Dec 2020.
Currently, Common Voice dataset releases are directly hosted on the Hub, under the Mozilla Foundation organization: https://huggingface.co/mozilla-foundation
For example... | load only England English dataset from common voice english dataset | training_data = load_dataset("common_voice", "en",split='train[:250]+validation[:250]')
testing_data = load_dataset("common_voice", "en", split="test[:200]")
I'm trying to load only 8% of the English common voice data with accent == "England English." Can somebody assist me with this?
**Typical Voice Accent Prop... | 172 | load only England English dataset from common voice english dataset
training_data = load_dataset("common_voice", "en",split='train[:250]+validation[:250]')
testing_data = load_dataset("common_voice", "en", split="test[:200]")
I'm trying to load only 8% of the English common voice data with accent == "England Engl... | [
-0.16265305876731873,
0.006645873188972473,
-0.06801293790340424,
-0.07248746603727341,
0.19177360832691193,
0.14461180567741394,
0.5200456380844116,
0.2401534616947174,
0.014608286321163177,
0.08605844527482986,
-0.4945831894874573,
-0.26028913259506226,
0.2044367641210556,
-0.15615965425... |
https://github.com/huggingface/datasets/issues/3851 | Hi @lemoner20, thanks for reporting.
I'm sorry but I cannot reproduce your problem:
```python
In [1]: from datasets import load_dataset, load_metric, Audio
...: raw_datasets = load_dataset("superb", "ks", split="train")
...: print(raw_datasets[0]["audio"])
Downloading builder script: 30.2kB [00:00, 13.0MB... | Load audio dataset error | ## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb", "ks", split="train")
prin... | 178 | Load audio dataset error
## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb",... | [
-0.37698894739151,
-0.1154063269495964,
-0.030820276588201523,
0.39622485637664795,
0.45797035098075867,
0.06998489052057266,
0.29024243354797363,
0.31508970260620117,
-0.0530281588435173,
0.174897700548172,
-0.5356049537658691,
0.31720149517059326,
-0.29740631580352783,
0.0095627456903457... |
https://github.com/huggingface/datasets/issues/3851 | @albertvillanova Thanks for your reply. The environment info below
## Environment info
- `datasets` version: 1.18.3
- Platform: Linux-4.19.91-007.ali4000.alios7.x86_64-x86_64-with-debian-buster-sid
- Python version: 3.6.12
- PyArrow version: 6.0.1 | Load audio dataset error | ## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb", "ks", split="train")
prin... | 27 | Load audio dataset error
## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb",... | [
-0.37698894739151,
-0.1154063269495964,
-0.030820276588201523,
0.39622485637664795,
0.45797035098075867,
0.06998489052057266,
0.29024243354797363,
0.31508970260620117,
-0.0530281588435173,
0.174897700548172,
-0.5356049537658691,
0.31720149517059326,
-0.29740631580352783,
0.0095627456903457... |
https://github.com/huggingface/datasets/issues/3851 | Thanks @lemoner20,
I cannot reproduce your issue in datasets version 1.18.3 either.
Maybe redownloading the data file may work if you had already cached this dataset previously. Could you please try passing "force_redownload"?
```python
raw_datasets = load_dataset("superb", "ks", split="train", download_mode="f... | Load audio dataset error | ## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb", "ks", split="train")
prin... | 40 | Load audio dataset error
## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb",... | [
-0.37698894739151,
-0.1154063269495964,
-0.030820276588201523,
0.39622485637664795,
0.45797035098075867,
0.06998489052057266,
0.29024243354797363,
0.31508970260620117,
-0.0530281588435173,
0.174897700548172,
-0.5356049537658691,
0.31720149517059326,
-0.29740631580352783,
0.0095627456903457... |
https://github.com/huggingface/datasets/issues/3851 | @albertvillanova, you can actually reproduce the error if you reach the cell `common_voice_train[0]["path"]` of this [notebook](https://colab.research.google.com/github/patrickvonplaten/notebooks/blob/master/Fine_Tune_XLSR_Wav2Vec2_on_Turkish_ASR_with_%F0%9F%A4%97_Transformers.ipynb#scrollTo=_0kRndSvqaKk). Error gets s... | Load audio dataset error | ## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb", "ks", split="train")
prin... | 29 | Load audio dataset error
## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb",... | [
-0.37698894739151,
-0.1154063269495964,
-0.030820276588201523,
0.39622485637664795,
0.45797035098075867,
0.06998489052057266,
0.29024243354797363,
0.31508970260620117,
-0.0530281588435173,
0.174897700548172,
-0.5356049537658691,
0.31720149517059326,
-0.29740631580352783,
0.0095627456903457... |
https://github.com/huggingface/datasets/issues/3851 | @jvel07, thanks for reporting and finding a solution.
Maybe we could tell @patrickvonplaten about the version pinning issue in his notebook. | Load audio dataset error | ## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb", "ks", split="train")
prin... | 21 | Load audio dataset error
## Load audio dataset error
Hi, when I load audio dataset following https://huggingface.co/docs/datasets/audio_process and https://github.com/huggingface/datasets/tree/master/datasets/superb,
```
from datasets import load_dataset, load_metric, Audio
raw_datasets = load_dataset("superb",... | [
-0.37698894739151,
-0.1154063269495964,
-0.030820276588201523,
0.39622485637664795,
0.45797035098075867,
0.06998489052057266,
0.29024243354797363,
0.31508970260620117,
-0.0530281588435173,
0.174897700548172,
-0.5356049537658691,
0.31720149517059326,
-0.29740631580352783,
0.0095627456903457... |
https://github.com/huggingface/datasets/issues/3848 | Hi @jxmorris12, thanks for reporting.
The objective of `verify_checksums` is to check that both checksums are equal. Therefore if one is None and the other is non-None, they are not equal, and the function accordingly raises a NonMatchingChecksumError. That behavior is expected.
The question is: how did you gener... | NonMatchingChecksumError when checksum is None | I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': 'efb4cbd3aa4a87bfaffc310ae951981cc0a36c6c71c64... | 157 | NonMatchingChecksumError when checksum is None
I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum... | [
-0.23207403719425201,
0.005118116736412048,
-0.06307563185691833,
0.011417253874242306,
0.03937027230858803,
-0.001827731728553772,
0.0637269914150238,
0.2139839082956314,
0.22390520572662354,
0.24077877402305603,
0.20892900228500366,
-0.1715061515569687,
-0.17938417196273804,
-0.168630659... |
https://github.com/huggingface/datasets/issues/3848 | Thanks @albertvillanova!
That's fine. I did run that command when I was adding a new dataset. Maybe because the command crashed in the middle, the checksum wasn't stored properly. I don't know where the bug is happening. But either (i) `verify_checksums` should properly handle this edge case, where the passed checks... | NonMatchingChecksumError when checksum is None | I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': 'efb4cbd3aa4a87bfaffc310ae951981cc0a36c6c71c64... | 133 | NonMatchingChecksumError when checksum is None
I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum... | [
-0.23207403719425201,
0.005118116736412048,
-0.06307563185691833,
0.011417253874242306,
0.03937027230858803,
-0.001827731728553772,
0.0637269914150238,
0.2139839082956314,
0.22390520572662354,
0.24077877402305603,
0.20892900228500366,
-0.1715061515569687,
-0.17938417196273804,
-0.168630659... |
https://github.com/huggingface/datasets/issues/3848 | Hi @jxmorris12,
Definitely, your `dataset_infos.json` was corrupted (and wrongly contains expected None checksum).
While we further investigate how this can happen and fix it, feel free to delete your `dataset_infos.json` file and recreate it with:
```shell
datasets-cli test <your-dataset-folder> --save_infos ... | NonMatchingChecksumError when checksum is None | I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': 'efb4cbd3aa4a87bfaffc310ae951981cc0a36c6c71c64... | 77 | NonMatchingChecksumError when checksum is None
I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum... | [
-0.23207403719425201,
0.005118116736412048,
-0.06307563185691833,
0.011417253874242306,
0.03937027230858803,
-0.001827731728553772,
0.0637269914150238,
0.2139839082956314,
0.22390520572662354,
0.24077877402305603,
0.20892900228500366,
-0.1715061515569687,
-0.17938417196273804,
-0.168630659... |
https://github.com/huggingface/datasets/issues/3848 | At a higher level, also note that we are preparing the release of `datasets` version 2.0, and some docs are being updated...
In order to add a dataset, I think the most updated instructions are in our official documentation pages: https://huggingface.co/docs/datasets/share | NonMatchingChecksumError when checksum is None | I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': 'efb4cbd3aa4a87bfaffc310ae951981cc0a36c6c71c64... | 41 | NonMatchingChecksumError when checksum is None
I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum... | [
-0.23207403719425201,
0.005118116736412048,
-0.06307563185691833,
0.011417253874242306,
0.03937027230858803,
-0.001827731728553772,
0.0637269914150238,
0.2139839082956314,
0.22390520572662354,
0.24077877402305603,
0.20892900228500366,
-0.1715061515569687,
-0.17938417196273804,
-0.168630659... |
https://github.com/huggingface/datasets/issues/3848 | Hi @jxmorris12, we have discovered the bug why `None` checksums wrongly appeared when generating the `dataset_infos.json` file:
- #3892
The fix will be accessible once this PR merged. And we are planning to do our 2.0 release today.
We are also working on updating all our docs for our release today. | NonMatchingChecksumError when checksum is None | I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': 'efb4cbd3aa4a87bfaffc310ae951981cc0a36c6c71c64... | 51 | NonMatchingChecksumError when checksum is None
I ran into the following error when adding a new dataset:
```bash
expected_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum': None, 'num_bytes': 40662}}
recorded_checksums = {'https://adversarialglue.github.io/dataset/dev.zip': {'checksum... | [
-0.23207403719425201,
0.005118116736412048,
-0.06307563185691833,
0.011417253874242306,
0.03937027230858803,
-0.001827731728553772,
0.0637269914150238,
0.2139839082956314,
0.22390520572662354,
0.24077877402305603,
0.20892900228500366,
-0.1715061515569687,
-0.17938417196273804,
-0.168630659... |
https://github.com/huggingface/datasets/issues/3847 | <s>I think this is because the tokenizer is stateful and because the order in which the splits are processed is not deterministic. Because of that, the hash of the tokenizer may change for certain splits, which causes issues with caching.
To fix this we can try making the order of the splits deterministic for map.</... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 55 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.19358354806900024,
0.25765305757522583,
0.060329534113407135,
0.14393001794815063,
0.08865343034267426,
-0.012036867439746857,
0.23727351427078247,
0.23071344196796417,
-0.1739625632762909,
-0.09237626194953918,
0.09441772848367691,
0.41973960399627686,
0.14737117290496826,
-0.498831421... |
https://github.com/huggingface/datasets/issues/3847 | Actually this is not because of the order of the splits, but most likely because the tokenizer used to process the second split is in a state that has been modified by the first split.
Therefore after reloading the first split from the cache, then the second split can't be reloaded since the tokenizer hasn't seen th... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 81 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.1986139714717865,
0.2466835081577301,
0.06692977249622345,
0.11836960166692734,
0.08003692328929901,
0.012468613684177399,
0.27197974920272827,
0.2143706977367401,
-0.1935911476612091,
-0.10805371403694153,
0.08862224221229553,
0.41674286127090454,
0.1809748411178589,
-0.537752151489257... |
https://github.com/huggingface/datasets/issues/3847 | Sorry didn't have the bandwidth to take care of this yet - will re-assign when I'm diving into it again ! | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 21 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.14964628219604492,
0.3147253096103668,
0.07166846841573715,
0.1290515810251236,
0.157375767827034,
0.04769377410411835,
0.24011585116386414,
0.2392692267894745,
-0.17129305005073547,
-0.11870883405208588,
0.11492938548326492,
0.4237600564956665,
0.13289280235767365,
-0.54347163438797,
... |
https://github.com/huggingface/datasets/issues/3847 | I had this issue with `run_speech_recognition_ctc.py` for wa2vec2.0 fine-tuning. I made a small change and the hash for the function (which includes tokenisation) is now the same before and after pre-porocessing. With the hash being the same, the caching works as intended.
Before:
```
def prepare_dataset(batch... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 159 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.11391603946685791,
0.2947026193141937,
0.10052897036075592,
0.1007809042930603,
0.14382940530776978,
0.009961709380149841,
0.17227628827095032,
0.29016420245170593,
-0.25690656900405884,
-0.16485454142093658,
0.13269025087356567,
0.40187811851501465,
0.11924254149198532,
-0.561412036418... |
https://github.com/huggingface/datasets/issues/3847 | Not sure why the second one would work and not the first one - they're basically the same with respect to hashing. In both cases the function is hashed recursively, and therefore the feature_extractor and the tokenizer are hashed the same way.
With which tokenizer or feature extractor are you experiencing this behav... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 83 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.10074012726545334,
0.2938990592956543,
0.06404461711645126,
0.13115236163139343,
0.13234394788742065,
0.02284032106399536,
0.23811785876750946,
0.22757990658283234,
-0.20204676687717438,
-0.08689344674348831,
0.13989681005477905,
0.42688220739364624,
0.09699411690235138,
-0.532903373241... |
https://github.com/huggingface/datasets/issues/3847 | Thanks ! Hopefully this can be useful to others, and also to better understand and improve hashing/caching | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 17 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.12665748596191406,
0.2988384962081909,
0.08312654495239258,
0.17981979250907898,
0.16854317486286163,
0.06762245297431946,
0.2584797441959381,
0.23254257440567017,
-0.20233355462551117,
-0.14918699860572815,
0.13579913973808289,
0.3831775188446045,
0.13598911464214325,
-0.54830324649810... |
https://github.com/huggingface/datasets/issues/3847 | `tokenizer.save_pretrained(training_args.output_dir)` produces a different tokenizer hash when loaded on restart of the script. When I was debugging before I was terminating the script prior to this command, then rerunning.
I compared the tokenizer items on the first and second runs, there are two different items:
... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 321 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.16589900851249695,
0.3236856162548065,
0.08280570805072784,
0.12946438789367676,
0.18572531640529633,
0.03222183883190155,
0.29114100337028503,
0.24498701095581055,
-0.19679784774780273,
-0.09392358362674713,
0.12795346975326538,
0.39861413836479187,
0.07593581080436707,
-0.557339012622... |
https://github.com/huggingface/datasets/issues/3847 | Thanks for investigating ! Does that mean that `save_pretrained`() produces non-deterministic tokenizers on disk ? Or is it `from_pretrained()` which is not deterministic given the same files on disk ?
I think one way to fix this would be to make save/from_pretrained deterministic, or make the pickling of `transform... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 70 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.10411102324724197,
0.20114873349666595,
0.06225219741463661,
0.16448941826820374,
0.20187051594257355,
0.07767505943775177,
0.327997088432312,
0.10047197341918945,
-0.20836487412452698,
-0.1435128152370453,
0.08319960534572601,
0.39779672026634216,
0.03241744637489319,
-0.59894496202468... |
https://github.com/huggingface/datasets/issues/3847 | Late to the party but everything should be deterministic (afaik at least).
But `Trie` is a simple class object, so afaik it's hash function is linked to its `id(self)` so basically where it's stored in memory, so super highly non deterministic. Could that be the issue ? | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 47 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.0841127336025238,
0.23371079564094543,
0.07575037330389023,
0.1667776107788086,
0.19347424805164337,
0.02686731517314911,
0.24100258946418762,
0.21442562341690063,
-0.18492090702056885,
-0.1199599876999855,
0.0748920887708664,
0.47011181712150574,
0.15747776627540588,
-0.564740180969238... |
https://github.com/huggingface/datasets/issues/3847 | > But Trie is a simple class object, so afaik it's hash function is linked to its id(self) so basically where it's stored in memory, so super highly non deterministic. Could that be the issue ?
We're computing the hash of the pickle dump of the class so it should be fine, as long as the pickle dump is deterministic | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 60 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.05189945921301842,
0.20742106437683105,
0.09117482602596283,
0.1883251965045929,
0.22655431926250458,
0.022534437477588654,
0.26121142506599426,
0.21120978891849518,
-0.1954527199268341,
-0.13296276330947876,
0.10679000616073608,
0.4456557035446167,
0.16125687956809998,
-0.5669119954109... |
https://github.com/huggingface/datasets/issues/3847 | I've ported wav2vec2.0 fine-tuning into Optimum-Graphcore which is where I found the issue. The majority of the script was copied from the Transformers version to keep it similar, [here is the tokenizer loading section from the source](https://github.com/huggingface/transformers/blob/f0982682bd6fd0b438dda79ec45f3a8fac8... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 116 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.08007006347179413,
0.258276104927063,
0.10429313033819199,
0.11781194806098938,
0.13655918836593628,
-0.07179176807403564,
0.11799369752407074,
0.23306414484977722,
-0.1783328652381897,
-0.15908551216125488,
0.12981221079826355,
0.41637516021728516,
0.10382114350795746,
-0.5137028694152... |
https://github.com/huggingface/datasets/issues/3847 | Which Python version are you using ?
The trie is basically a big dict of dics, so deterministic nature depends on python version:
https://stackoverflow.com/questions/2053021/is-the-order-of-a-python-dictionary-guaranteed-over-iterations
Maybe the investigation is actually not finding the right culprit though (th... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 63 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.09939133375883102,
0.1641964614391327,
0.05952886492013931,
0.11624377965927124,
0.13546247780323029,
-0.00883089005947113,
0.20630833506584167,
0.2005600482225418,
-0.16751481592655182,
-0.0839846208691597,
0.10367266088724136,
0.48232319951057434,
0.17413364350795746,
-0.5205731391906... |
https://github.com/huggingface/datasets/issues/3847 | Similar issue found on `BartTokenizer`. You can bypass the bug by loading a fresh new tokenizer everytime.
```
dataset = dataset.map(lambda x: tokenize_func(x, BartTokenizer.from_pretrained(xxx)),
num_proc=num_proc, desc='Tokenize')
``` | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 27 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.16687816381454468,
0.33831819891929626,
0.06849215924739838,
0.1348816454410553,
0.16055895388126373,
0.02588513493537903,
0.24182730913162231,
0.23366431891918182,
-0.17249532043933868,
-0.10836057364940643,
0.09585655480623245,
0.4308624267578125,
0.1363021582365036,
-0.56485599279403... |
https://github.com/huggingface/datasets/issues/3847 | I got the same problem while using Wav2Vec2CTCTokenizer in a distributed experiment (many processes), and found that the problem was localized in the serialization (pickle dump) of the field `tokenizer.tokens_trie._tokens` (just a python set). I focussed into the set serialization and found it is not deterministic:
... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 97 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.1142750233411789,
0.3124810457229614,
0.0932392105460167,
0.12759806215763092,
0.13084088265895844,
-0.0025873109698295593,
0.20931121706962585,
0.2543276846408844,
-0.2231585830450058,
-0.14100202918052673,
0.1485624462366104,
0.48311740159988403,
0.12295311689376831,
-0.58149033784866... |
https://github.com/huggingface/datasets/issues/3847 | OK, I assume python's set is just a hash table implementation that uses internally the hash() function. The problem is that python's hash() is not deterministic. I believe that setting the environment variable PYTHONHASHSEED to a fixed value, you can force it to be deterministic. I tried it (file `set_pickle_dump.py`):... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 139 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.1804419755935669,
0.2794873118400574,
0.062471695244312286,
0.12339435517787933,
0.16611146926879883,
0.07879507541656494,
0.2757985591888428,
0.2529333233833313,
-0.17466869950294495,
-0.07347744703292847,
0.14858883619308472,
0.43186789751052856,
0.09926337003707886,
-0.50730848312377... |
https://github.com/huggingface/datasets/issues/3847 | Thanks for reporting. I opened a PR here to propose a fix: https://github.com/huggingface/datasets/pull/6318 and doesn't require setting `PYTHONHASHSEED`
Can you try to install `datasets` from this branch and tell me if it fixes the issue ? | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 36 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.16839349269866943,
0.2789086103439331,
0.07585213333368301,
0.16233444213867188,
0.1720108687877655,
0.05907590687274933,
0.22213901579380035,
0.25853925943374634,
-0.17556549608707428,
-0.14856207370758057,
0.08232194930315018,
0.42024895548820496,
0.16028742492198944,
-0.4802697300910... |
https://github.com/huggingface/datasets/issues/3847 | I patched (*) the file `datasets/utils/py_utils.py` and cache is working propperly now. Thanks!
(*): I am running my experiments inside a docker container that depends on `huggingface/transformers-pytorch-gpu:latest`, so pattched the file instead of rebuilding the container from scratch | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 38 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.15966099500656128,
0.23978807032108307,
0.10841071605682373,
0.19765746593475342,
0.19014573097229004,
0.003737904131412506,
0.2869730293750763,
0.21496987342834473,
-0.14684158563613892,
-0.1517270803451538,
0.053289689123630524,
0.384452223777771,
0.1924055814743042,
-0.56607306003570... |
https://github.com/huggingface/datasets/issues/3847 | I think the Trie()._tokens of PreTrainedTokenizer need to be a sorted set So that the results of `hash_bytes(dumps(tokenizer))` are consistent every time | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 22 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.12439455091953278,
0.25687670707702637,
0.062205202877521515,
0.14954790472984314,
0.18100517988204956,
0.03493094444274902,
0.2350013703107834,
0.23690658807754517,
-0.23388153314590454,
-0.1104135736823082,
0.09626293182373047,
0.42446374893188477,
0.117528036236763,
-0.55228829383850... |
https://github.com/huggingface/datasets/issues/3847 | I believe the issue may be linked to [tokenization_utils.py#L507](https://github.com/huggingface/transformers/blob/main/src/transformers/tokenization_utils.py#L507)，specifically in the line where self.tokens_trie.add(token.content) is called. The function _update_trie appears to modify an unordered set. Consequently, t... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 81 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.13172775506973267,
0.22908565402030945,
0.1099361851811409,
0.16339516639709473,
0.18574506044387817,
0.01777835190296173,
0.2914983034133911,
0.1949392706155777,
-0.211125910282135,
-0.16296198964118958,
0.05725633352994919,
0.38519155979156494,
0.18223664164543152,
-0.570158064365387,... |
https://github.com/huggingface/datasets/issues/3847 | We did a fix that does sorting whenever we hash sets. The fix is available on `main` if you want to try it out. We'll do a new release soon :) | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 31 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.1391856074333191,
0.2630721628665924,
0.06136839836835861,
0.1422164887189865,
0.1564342975616455,
0.05836629867553711,
0.21195973455905914,
0.27680376172065735,
-0.19538521766662598,
-0.09325698018074036,
0.10779307037591934,
0.40425679087638855,
0.12266594171524048,
-0.485643327236175... |
https://github.com/huggingface/datasets/issues/3847 | Is there a documentation chapter that discusses in which cases you should expect your dataset preprocessing to be cached. Including do's and don'ts for the preprocessing functions? I think Datasets team does amazing job at tacking this issue on their side, but it would be great to have some guidelines on the user side ... | Datasets' cache not re-used | ## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tokenizer works perfectly with ca... | 95 | Datasets' cache not re-used
## Describe the bug
For most tokenizers I have tested (e.g. the RoBERTa tokenizer), the data preprocessing cache are not fully reused in the first few runs, although their `.arrow` cache files are in the cache directory.
## Steps to reproduce the bug
Here is a reproducer. The GPT2 tok... | [
-0.07781969755887985,
0.32464808225631714,
0.04743676260113716,
0.08195948600769043,
0.15503299236297607,
0.0566411018371582,
0.24692264199256897,
0.11230770498514175,
-0.23134484887123108,
-0.1770492047071457,
0.1307278722524643,
0.3741540014743805,
0.20596180856227875,
-0.550763905048370... |
https://github.com/huggingface/datasets/issues/3841 | Hi!
This issue stems from `datasets` having `py.typed` defined (see https://github.com/microsoft/pyright/discussions/3764#discussioncomment-3282142) - to avoid it, we would either have to remove `py.typed` (added to be compliant with PEP-561) or export the names with `__all__`/`from .submodule import name as name`.... | Pyright reportPrivateImportUsage when `from datasets import load_dataset` | ## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected results
No complain from Pyright
## Actual results
Pyright complain below... | 90 | Pyright reportPrivateImportUsage when `from datasets import load_dataset`
## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected r... | [
-0.12595169246196747,
-0.28096455335617065,
0.027085840702056885,
0.27159008383750916,
0.3544701635837555,
0.10418292880058289,
0.6413933038711548,
0.19132962822914124,
0.15554407238960266,
-0.24281197786331177,
-0.18240536749362946,
0.5359144806861877,
0.04852690175175667,
0.1362614482641... |
https://github.com/huggingface/datasets/issues/3841 | > (cc @Wauplin since huggingface_hub has the same issue.)
Hmm maybe we have the same issue but I haven't been able to reproduce something similar to `"load_dataset" is not exported from module "datasets"` message (using VSCode+Pylance -that is powered by Pyright). `huggingface_hub` contains a `py.typed` file but th... | Pyright reportPrivateImportUsage when `from datasets import load_dataset` | ## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected results
No complain from Pyright
## Actual results
Pyright complain below... | 115 | Pyright reportPrivateImportUsage when `from datasets import load_dataset`
## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected r... | [
-0.17349320650100708,
-0.3427664637565613,
0.059319525957107544,
0.2731616199016571,
0.253451406955719,
0.03000076860189438,
0.39011701941490173,
0.1951955407857895,
0.19493547081947327,
-0.25562340021133423,
-0.2439197301864624,
0.37836548686027527,
-0.024694841355085373,
0.20143829286098... |
https://github.com/huggingface/datasets/issues/3841 | @Wauplin I only get the error if I use Pyright's CLI tool or the Pyright extension (not sure why, but Pylance also doesn't report this issue on my machine)
> Also the top-level [`__init__.py`](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/__init__.py) is quite different in `hfh` than `... | Pyright reportPrivateImportUsage when `from datasets import load_dataset` | ## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected results
No complain from Pyright
## Actual results
Pyright complain below... | 78 | Pyright reportPrivateImportUsage when `from datasets import load_dataset`
## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected r... | [
-0.08133730292320251,
-0.35310128331184387,
0.10134046524763107,
0.2233797013759613,
0.2092587649822235,
0.08671054989099503,
0.4403260052204132,
0.23700043559074402,
0.2594301104545593,
-0.2501111626625061,
-0.19836151599884033,
0.3156059682369232,
0.060629162937402725,
0.2214425206184387... |
https://github.com/huggingface/datasets/issues/3841 | @mariosasko thank for the tip. I have been able to reproduce the issue as well. I would be up for including a (huge) static `__all__` variable in the `__init__.py` (since the file is already generated automatically in `hfh`) but honestly I don't think it's worth the hassle.
I'll delete the `py.typed` file in `huggi... | Pyright reportPrivateImportUsage when `from datasets import load_dataset` | ## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected results
No complain from Pyright
## Actual results
Pyright complain below... | 66 | Pyright reportPrivateImportUsage when `from datasets import load_dataset`
## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected r... | [
-0.1466846466064453,
-0.39144167304039,
0.07356608659029007,
0.2058163285255432,
0.22391653060913086,
0.08266499638557434,
0.44132813811302185,
0.1992199867963791,
0.24119573831558228,
-0.21934418380260468,
-0.2073211967945099,
0.3753395676612854,
0.08785111457109451,
0.18381953239440918,
... |
https://github.com/huggingface/datasets/issues/3841 | I am getting this error in google colab today:
The code runs just fine too. | Pyright reportPrivateImportUsage when `from datasets import load_dataset` | ## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected results
No complain from Pyright
## Actual results
Pyright complain below... | 16 | Pyright reportPrivateImportUsage when `from datasets import load_dataset`
## Describe the bug
Pyright complains about module not exported.
## Steps to reproduce the bug
Use an editor/IDE with Pyright Language server with default configuration:
```python
from datasets import load_dataset
```
## Expected r... | [
-0.1839921772480011,
-0.12173615396022797,
0.02431561052799225,
0.28692859411239624,
0.21422958374023438,
0.07978426665067673,
0.5122384428977966,
0.08669635653495789,
0.14836618304252625,
-0.18913844227790833,
-0.005789339542388916,
0.7045692205429077,
-0.02210153266787529,
0.187218427658... |
https://github.com/huggingface/datasets/issues/3832 | It will be indeed really great to add support to GNN datasets. Big :+1: for this initiative. | Making Hugging Face the place to go for Graph NNs datasets | Let's make Hugging Face Datasets the central hub for GNN datasets :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the GNN field.
What are some datasets worth integrating into the Hugging Face hub?
In... | 17 | Making Hugging Face the place to go for Graph NNs datasets
Let's make Hugging Face Datasets the central hub for GNN datasets :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the GNN field.
What are som... | [
-0.02342338114976883,
-0.23334729671478271,
-0.10941953212022781,
-0.052362192422151566,
-0.09441505372524261,
-0.043583303689956665,
-0.028115157037973404,
0.10992133617401123,
0.326602578163147,
0.18508872389793396,
-0.13129431009292603,
-0.08443969488143921,
-0.23202446103096008,
0.4511... |
https://github.com/huggingface/datasets/issues/3832 | @napoles-uach identifies the [TUDatasets](https://chrsmrrs.github.io/datasets/) (A collection of benchmark datasets for graph classification and regression).
Added to the Tasks in the initial issue. | Making Hugging Face the place to go for Graph NNs datasets | Let's make Hugging Face Datasets the central hub for GNN datasets :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the GNN field.
What are some datasets worth integrating into the Hugging Face hub?
In... | 22 | Making Hugging Face the place to go for Graph NNs datasets
Let's make Hugging Face Datasets the central hub for GNN datasets :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the GNN field.
What are som... | [
-0.02336067706346512,
-0.24508532881736755,
-0.09793679416179657,
-0.005754740908741951,
-0.10700173676013947,
-0.036405548453330994,
-0.029364079236984253,
0.11817415803670883,
0.3275017738342285,
0.14410167932510376,
-0.1514698565006256,
-0.10237426310777664,
-0.22312134504318237,
0.4196... |
https://github.com/huggingface/datasets/issues/3832 | Great initiative! Let's keep this issue for these 3 datasets, but moving forward maybe let's create a new issue per dataset :rocket: great work @napoles-uach and @omarespejel! | Making Hugging Face the place to go for Graph NNs datasets | Let's make Hugging Face Datasets the central hub for GNN datasets :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the GNN field.
What are some datasets worth integrating into the Hugging Face hub?
In... | 27 | Making Hugging Face the place to go for Graph NNs datasets
Let's make Hugging Face Datasets the central hub for GNN datasets :)
**Motivation**. Datasets are currently quite scattered and an open-source central point such as the Hugging Face Hub would be ideal to support the growth of the GNN field.
What are som... | [
-0.01863967627286911,
-0.23025695979595184,
-0.09310664236545563,
-0.028547536581754684,
-0.12222643196582794,
-0.06372932344675064,
0.005652200430631638,
0.11484649777412415,
0.3288175165653229,
0.16832223534584045,
-0.12983962893486023,
-0.08059713989496231,
-0.23037156462669373,
0.44871... |
https://github.com/huggingface/datasets/issues/3831 | Hi @greenned, this is expected behaviour for `to_tf_dataset`. By default, we drop the smaller 'remainder' batch during training (i.e. when `shuffle=True`). If you really want to keep that batch, you can set `drop_remainder=False` when calling `to_tf_dataset()`. | when using to_tf_dataset with shuffle is true, not all completed batches are made | ## Describe the bug
when converting a dataset to tf_dataset by using to_tf_dataset with shuffle true, the remainder is not converted to one batch
## Steps to reproduce the bug
this is the sample code below
https://colab.research.google.com/drive/1_oRXWsR38ElO1EYF9ayFoCU7Ou1AAej4?usp=sharing
## Expected resul... | 36 | when using to_tf_dataset with shuffle is true, not all completed batches are made
## Describe the bug
when converting a dataset to tf_dataset by using to_tf_dataset with shuffle true, the remainder is not converted to one batch
## Steps to reproduce the bug
this is the sample code below
https://colab.research.g... | [
-0.21016058325767517,
-0.15655343234539032,
0.10085833072662354,
0.10950109362602234,
0.3623254895210266,
0.17205461859703064,
0.09992387145757675,
0.3753470182418823,
-0.4610954225063324,
0.4258151054382324,
-0.1732448786497116,
0.35223084688186646,
0.05077914893627167,
-0.030581936240196... |
https://github.com/huggingface/datasets/issues/3830 | Was able to reproduce the issue on Colab; full logs below.
```
---------------------------------------------------------------------------
NotADirectoryError Traceback (most recent call last)
[<ipython-input-2-39967739ba7f>](https://localhost:8080/#) in <module>()
1 import datasets
... | Got error when load cnn_dailymail dataset | When using datasets.load_dataset method to load cnn_dailymail dataset, got error as below:
- windows os: FileNotFoundError: [WinError 3] 系统找不到指定的路径。: 'D:\\SourceCode\\DataScience\\HuggingFace\\Data\\downloads\\1bc05d24fa6dda2468e83a73cf6dc207226e01e3c48a507ea716dc0421da583b\\cnn\\stories'
- google colab: NotADirec... | 201 | Got error when load cnn_dailymail dataset
When using datasets.load_dataset method to load cnn_dailymail dataset, got error as below:
- windows os: FileNotFoundError: [WinError 3] 系统找不到指定的路径。: 'D:\\SourceCode\\DataScience\\HuggingFace\\Data\\downloads\\1bc05d24fa6dda2468e83a73cf6dc207226e01e3c48a507ea716dc0421da58... | [
-0.309506893157959,
0.20986396074295044,
0.02820286899805069,
0.4526914060115814,
0.27861282229423523,
0.030075743794441223,
0.6285233497619629,
-0.014282329007983208,
0.012505806982517242,
0.29893237352371216,
-0.30066627264022827,
0.2344314306974411,
-0.40582334995269775,
0.1356498897075... |
https://github.com/huggingface/datasets/issues/3830 | Hi @jon-tow, thanks for reporting. And hi @dynamicwebpaige, thanks for your investigation.
This issue was already reported
- #3784
and its root cause is a change in the Google Drive service. See:
- #3786
We have already fixed it. See:
- #3787
We are planning to make a patch release today (indeed, we ... | Got error when load cnn_dailymail dataset | When using datasets.load_dataset method to load cnn_dailymail dataset, got error as below:
- windows os: FileNotFoundError: [WinError 3] 系统找不到指定的路径。: 'D:\\SourceCode\\DataScience\\HuggingFace\\Data\\downloads\\1bc05d24fa6dda2468e83a73cf6dc207226e01e3c48a507ea716dc0421da583b\\cnn\\stories'
- google colab: NotADirec... | 129 | Got error when load cnn_dailymail dataset
When using datasets.load_dataset method to load cnn_dailymail dataset, got error as below:
- windows os: FileNotFoundError: [WinError 3] 系统找不到指定的路径。: 'D:\\SourceCode\\DataScience\\HuggingFace\\Data\\downloads\\1bc05d24fa6dda2468e83a73cf6dc207226e01e3c48a507ea716dc0421da58... | [
-0.13006338477134705,
0.21794839203357697,
0.02283388003706932,
0.3490012288093567,
0.2328822761774063,
0.13096299767494202,
0.5199148654937744,
-0.04545578360557556,
0.015079930424690247,
0.22897358238697052,
-0.12210632115602493,
0.10922284424304962,
-0.3064780533313751,
0.33780029416084... |
https://github.com/huggingface/datasets/issues/3829 | Hi ! Yes this is definitely something we'll explore, since optimizing processing pipelines can be challenging and because performance is key here: we want anyone to be able to play with large-scale datasets more easily.
I think we'll start by documenting the performance of the dataset transforms we provide, and then... | [📄 Docs] Create a `datasets` performance guide. | ## Brief Overview
Downloading, saving, and preprocessing large datasets from the `datasets` library can often result in [performance bottlenecks](https://github.com/huggingface/datasets/issues/3735). These performance snags can be challenging to identify and to debug, especially for users who are less experienced with... | 60 | [📄 Docs] Create a `datasets` performance guide.
## Brief Overview
Downloading, saving, and preprocessing large datasets from the `datasets` library can often result in [performance bottlenecks](https://github.com/huggingface/datasets/issues/3735). These performance snags can be challenging to identify and to debug,... | [
-0.32639992237091064,
-0.06727629154920578,
-0.1491248607635498,
0.07967495173215866,
0.1978183537721634,
0.12434281408786774,
-0.08178788423538208,
0.451526939868927,
-0.2909836173057556,
0.04553689807653427,
-0.042812712490558624,
0.28554555773735046,
-0.32810941338539124,
0.409906089305... |
https://github.com/huggingface/datasets/issues/3828 | Hi @dlwh, thanks for reporting.
Please note, that the source data files for "all" config are different from the other configurations.
The "all" config contains the official Pile data files, from https://mystic.the-eye.eu/public/AI/pile/
All data examples contain a "meta" dict with a single "pile_set_name" key:
... | The Pile's _FEATURE spec seems to be incorrect | ## Describe the bug
If you look at https://huggingface.co/datasets/the_pile/blob/main/the_pile.py:
For "all"
* the pile_set_name is never set for data
* there's actually an id field inside of "meta"
For subcorpora pubmed_central and hacker_news:
* the meta is specified to be a string, but it's actually a di... | 219 | The Pile's _FEATURE spec seems to be incorrect
## Describe the bug
If you look at https://huggingface.co/datasets/the_pile/blob/main/the_pile.py:
For "all"
* the pile_set_name is never set for data
* there's actually an id field inside of "meta"
For subcorpora pubmed_central and hacker_news:
* the meta is... | [
-0.029128439724445343,
-0.00901050865650177,
0.09973619878292084,
0.1450231671333313,
0.32511743903160095,
-0.014572307467460632,
0.5119672417640686,
0.268432080745697,
-0.2205348163843155,
0.12272682785987854,
-0.1169433668255806,
0.30823996663093567,
0.24741780757904053,
0.37088331580162... |
https://github.com/huggingface/datasets/issues/3823 | Hi @jacobbieker, thanks for reporting!
I have transferred this issue to our Hub team and they are investigating it. I keep you informed. | 500 internal server error when trying to open a dataset composed of Zarr stores | ## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet, and I did push two [xarray](https://docs.xarray.dev/en/stable/) Zarr stores of da... | 23 | 500 internal server error when trying to open a dataset composed of Zarr stores
## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet,... | [
-0.39843615889549255,
0.10997380316257477,
0.07565900683403015,
0.26787039637565613,
0.30847302079200745,
-0.037865519523620605,
0.3065188527107239,
0.18530094623565674,
0.1879683881998062,
0.183724045753479,
-0.3976975381374359,
0.25280818343162537,
0.060528356581926346,
0.393114030361175... |
https://github.com/huggingface/datasets/issues/3823 | Hi @jacobbieker, we are investigating this issue on our side and we'll see if we can fix it, but please note that your repo is considered problematic for git. Here are the results of running https://github.com/github/git-sizer on it:
```
Processing blobs: 147448
Processing trees: 27 ... | 500 internal server error when trying to open a dataset composed of Zarr stores | ## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet, and I did push two [xarray](https://docs.xarray.dev/en/stable/) Zarr stores of da... | 142 | 500 internal server error when trying to open a dataset composed of Zarr stores
## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet,... | [
-0.39843615889549255,
0.10997380316257477,
0.07565900683403015,
0.26787039637565613,
0.30847302079200745,
-0.037865519523620605,
0.3065188527107239,
0.18530094623565674,
0.1879683881998062,
0.183724045753479,
-0.3976975381374359,
0.25280818343162537,
0.060528356581926346,
0.393114030361175... |
https://github.com/huggingface/datasets/issues/3823 | Hi, thanks for getting back to me so quick! And yeah, I figured that was probably the problem. I was going to try to delete the repo, but couldn't through the website, so if that's the easiest way to solve it, I can regenerate the dataset in a different format with less tiny files, and you guys can delete the repo as i... | 500 internal server error when trying to open a dataset composed of Zarr stores | ## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet, and I did push two [xarray](https://docs.xarray.dev/en/stable/) Zarr stores of da... | 97 | 500 internal server error when trying to open a dataset composed of Zarr stores
## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet,... | [
-0.39843615889549255,
0.10997380316257477,
0.07565900683403015,
0.26787039637565613,
0.30847302079200745,
-0.037865519523620605,
0.3065188527107239,
0.18530094623565674,
0.1879683881998062,
0.183724045753479,
-0.3976975381374359,
0.25280818343162537,
0.060528356581926346,
0.393114030361175... |
https://github.com/huggingface/datasets/issues/3823 | Hi @jacobbieker,
For future use cases, our Hub team is still pondering whether to limit the maximum number of files per repo to avoid technical issues...
On the meantime, they have made a fix and your dataset is working: https://huggingface.co/datasets/openclimatefix/mrms | 500 internal server error when trying to open a dataset composed of Zarr stores | ## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet, and I did push two [xarray](https://docs.xarray.dev/en/stable/) Zarr stores of da... | 40 | 500 internal server error when trying to open a dataset composed of Zarr stores
## Describe the bug
The dataset [openclimatefix/mrms](https://huggingface.co/datasets/openclimatefix/mrms) gives a 500 server error when trying to open it on the website, or through code.
The dataset doesn't have a loading script yet,... | [
-0.39843615889549255,
0.10997380316257477,
0.07565900683403015,
0.26787039637565613,
0.30847302079200745,
-0.037865519523620605,
0.3065188527107239,
0.18530094623565674,
0.1879683881998062,
0.183724045753479,
-0.3976975381374359,
0.25280818343162537,
0.060528356581926346,
0.393114030361175... |
https://github.com/huggingface/datasets/issues/3822 | Official dataset location : https://icu.ee.ethz.ch/research/datsets.html
In the "Biwi Kinect Head Pose Database" section, I do not find any information regarding "Downloading the dataset." . Do we mail the authors regarding this ?
I found the dataset on Kaggle : [Link](https://www.kaggle.com/kmader/biwi-kinect-head... | Add Biwi Kinect Head Pose Database | ## Adding a Dataset
- **Name:** Biwi Kinect Head Pose Database
- **Description:** Over 15K images of 20 people recorded with a Kinect while turning their heads around freely. For each frame, depth and rgb images are provided, together with ground in the form of the 3D location of the head and its rotation angles.
- ... | 85 | Add Biwi Kinect Head Pose Database
## Adding a Dataset
- **Name:** Biwi Kinect Head Pose Database
- **Description:** Over 15K images of 20 people recorded with a Kinect while turning their heads around freely. For each frame, depth and rgb images are provided, together with ground in the form of the 3D location of ... | [
0.09413466602563858,
0.0514218807220459,
-0.033351831138134,
0.21059153974056244,
-0.37651965022087097,
0.40544652938842773,
0.12584330141544342,
0.17784065008163452,
0.2764802873134613,
0.1559714376926422,
-0.010699216276407242,
-0.17998439073562622,
0.10054226219654083,
0.196102127432823... |
https://github.com/huggingface/datasets/issues/3820 | Hi @jon-tow, thanks for reporting.
This issue was already reported and its root cause is a change in the Google Drive service. See:
- #3786
We have already fixed it. See:
- #3787
We are planning to make a patch release today.
In the meantime, you can get this fix by installing our library from the GitHu... | `pubmed_qa` checksum mismatch | ## Describe the bug
Loading [`pubmed_qa`](https://huggingface.co/datasets/pubmed_qa) results in a mismatched checksum error.
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import datasets
try:
datasets.load_dataset("pubmed_qa", "pqa_labeled")
except Exception as e:
print(e... | 109 | `pubmed_qa` checksum mismatch
## Describe the bug
Loading [`pubmed_qa`](https://huggingface.co/datasets/pubmed_qa) results in a mismatched checksum error.
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import datasets
try:
datasets.load_dataset("pubmed_qa", "pqa_labeled")
exc... | [
-0.0026245489716529846,
0.15023401379585266,
-0.09011359512805939,
0.18270398676395416,
0.3038475513458252,
-0.05285906046628952,
0.2118266224861145,
0.332444429397583,
0.03438393771648407,
0.18833187222480774,
-0.07668942213058472,
0.11536199599504471,
0.30355945229530334,
0.0669022276997... |
https://github.com/huggingface/datasets/issues/3818 | Hi, thanks for reporting! We can add a `sources: datasets.Value("string")` feature to the `Features` dict in the `SARI` script to fix this. Would you be interested in submitting a PR? | Support for "sources" parameter in the add() and add_batch() methods in datasets.metric - SARI | **Is your feature request related to a problem? Please describe.**
The methods `add_batch` and `add` from the `Metric` [class](https://github.com/huggingface/datasets/blob/1675ad6a958435b675a849eafa8a7f10fe0f43bc/src/datasets/metric.py) does not work with [SARI](https://github.com/huggingface/datasets/blob/master/metr... | 30 | Support for "sources" parameter in the add() and add_batch() methods in datasets.metric - SARI
**Is your feature request related to a problem? Please describe.**
The methods `add_batch` and `add` from the `Metric` [class](https://github.com/huggingface/datasets/blob/1675ad6a958435b675a849eafa8a7f10fe0f43bc/src/datas... | [
-0.4188568592071533,
0.27242526412010193,
-0.009295731782913208,
-0.10950394719839096,
0.12285403907299042,
-0.07086554914712906,
0.23206773400306702,
0.2151489406824112,
-0.18371284008026123,
0.22034309804439545,
-0.2901640832424164,
0.285544753074646,
0.13469737768173218,
0.0324125885963... |
https://github.com/huggingface/datasets/issues/3818 | Hi Mario,
Thanks for your message. I did try to add `sources` into the `Features` dict using a script for the metric:
```
features=datasets.Features(
{
"sources": datasets.Value("string", id="sequence"),
"predictions": datasets.Value("string", ... | Support for "sources" parameter in the add() and add_batch() methods in datasets.metric - SARI | **Is your feature request related to a problem? Please describe.**
The methods `add_batch` and `add` from the `Metric` [class](https://github.com/huggingface/datasets/blob/1675ad6a958435b675a849eafa8a7f10fe0f43bc/src/datasets/metric.py) does not work with [SARI](https://github.com/huggingface/datasets/blob/master/metr... | 133 | Support for "sources" parameter in the add() and add_batch() methods in datasets.metric - SARI
**Is your feature request related to a problem? Please describe.**
The methods `add_batch` and `add` from the `Metric` [class](https://github.com/huggingface/datasets/blob/1675ad6a958435b675a849eafa8a7f10fe0f43bc/src/datas... | [
-0.4188568592071533,
0.27242526412010193,
-0.009295731782913208,
-0.10950394719839096,
0.12285403907299042,
-0.07086554914712906,
0.23206773400306702,
0.2151489406824112,
-0.18371284008026123,
0.22034309804439545,
-0.2901640832424164,
0.285544753074646,
0.13469737768173218,
0.0324125885963... |
https://github.com/huggingface/datasets/issues/3818 | The `Metric` class has been modified recently to support this use-case, but the `add_batch` + `compute` pattern still doesn't work correctly. I'll open a PR. | Support for "sources" parameter in the add() and add_batch() methods in datasets.metric - SARI | **Is your feature request related to a problem? Please describe.**
The methods `add_batch` and `add` from the `Metric` [class](https://github.com/huggingface/datasets/blob/1675ad6a958435b675a849eafa8a7f10fe0f43bc/src/datasets/metric.py) does not work with [SARI](https://github.com/huggingface/datasets/blob/master/metr... | 25 | Support for "sources" parameter in the add() and add_batch() methods in datasets.metric - SARI
**Is your feature request related to a problem? Please describe.**
The methods `add_batch` and `add` from the `Metric` [class](https://github.com/huggingface/datasets/blob/1675ad6a958435b675a849eafa8a7f10fe0f43bc/src/datas... | [
-0.4188568592071533,
0.27242526412010193,
-0.009295731782913208,
-0.10950394719839096,
0.12285403907299042,
-0.07086554914712906,
0.23206773400306702,
0.2151489406824112,
-0.18371284008026123,
0.22034309804439545,
-0.2901640832424164,
0.285544753074646,
0.13469737768173218,
0.0324125885963... |
https://github.com/huggingface/datasets/issues/3813 | I would like to take this up and give it a shot. Any image specific - dataset guidelines to keep in mind ? Thank you. | Add MetaShift dataset | ## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https://github.com/huggingface/... | 25 | Add MetaShift dataset
## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https:/... | [
-0.08063685894012451,
-0.0981556847691536,
-0.11934756487607956,
0.0553121343255043,
0.307888388633728,
-0.12920722365379333,
0.020914839580655098,
-0.005607111379504204,
0.13249775767326355,
-0.047706447541713715,
0.08887449651956558,
-0.033262018114328384,
-0.24265989661216736,
0.2571820... |
https://github.com/huggingface/datasets/issues/3813 | I've started working on adding this dataset. I require some inputs on the following :
Ref for the initial draft [here](https://github.com/dnaveenr/datasets/blob/add_metashift_dataset/datasets/metashift/metashift.py)
1. The dataset does not have a typical - train/test/val split. What do we do for the _split_generat... | Add MetaShift dataset | ## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https://github.com/huggingface/... | 156 | Add MetaShift dataset
## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https:/... | [
-0.020686857402324677,
-0.16771399974822998,
-0.004767287522554398,
0.0912783294916153,
0.322817325592041,
0.05552732199430466,
0.2001732587814331,
0.340548038482666,
-0.029036439955234528,
0.010151728987693787,
0.04645527899265289,
0.16041412949562073,
-0.11582872271537781,
0.355698525905... |
https://github.com/huggingface/datasets/issues/3813 | Hi ! Thanks for adding this dataset :) Let me answer your questions:
1. in this case you can put everything in the "train" split
2. Yes you can copy the script (provided you also include the MIT license of the code in the file header for example). Though we ideally try to not create new directories nor files when g... | Add MetaShift dataset | ## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https://github.com/huggingface/... | 152 | Add MetaShift dataset
## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https:/... | [
-0.17137476801872253,
-0.08549278974533081,
-0.14299973845481873,
0.12203129380941391,
0.24328134953975677,
0.021914832293987274,
0.5056436061859131,
0.2920890152454376,
0.1803073287010193,
0.1295091211795807,
-0.12590903043746948,
0.31878629326820374,
-0.38148820400238037,
0.3359365463256... |
https://github.com/huggingface/datasets/issues/3813 | Great. This is helpful. Thanks @lhoestq .
Regarding Point 2, I'll try using yield instead of creating the directories and see if its feasible. selected_classes config sounds good. | Add MetaShift dataset | ## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https://github.com/huggingface/... | 28 | Add MetaShift dataset
## Adding a Dataset
- **Name:** MetaShift
- **Description:** collection of 12,868 sets of natural images across 410 classes-
- **Paper:** https://arxiv.org/abs/2202.06523v1
- **Data:** https://github.com/weixin-liang/metashift
Instructions to add a new dataset can be found [here](https:/... | [
-0.38951757550239563,
-0.05094289779663086,
-0.1791783571243286,
0.16719427704811096,
0.3236088454723358,
-0.23887315392494202,
0.24628125131130219,
0.2632826268672943,
0.11790020763874054,
0.3443956673145294,
-0.1489604115486145,
0.1883937120437622,
-0.1833636462688446,
0.2654520273208618... |
https://github.com/huggingface/datasets/issues/3809 | Hi @muelletm, thanks for reporting.
This issue was already reported and its root cause is a change in the Google Drive service. See:
- #3786
We have already fixed it. See:
- #3787
Until our next `datasets` library release, you can get this fix by installing our library from the GitHub master branch:
```sh... | Checksums didn't match for datasets on Google Drive | ## Describe the bug
Datasets hosted on Google Drive do not seem to work right now.
Loading them fails with a checksum error.
## Steps to reproduce the bug
```python
from datasets import load_dataset
for dataset in ["head_qa", "yelp_review_full"]:
try:
load_dataset(dataset)
except Exception as excep... | 103 | Checksums didn't match for datasets on Google Drive
## Describe the bug
Datasets hosted on Google Drive do not seem to work right now.
Loading them fails with a checksum error.
## Steps to reproduce the bug
```python
from datasets import load_dataset
for dataset in ["head_qa", "yelp_review_full"]:
try:
... | [
-0.20049509406089783,
0.13795839250087738,
-0.09350788593292236,
0.385181725025177,
0.1761961579322815,
0.0669630914926529,
0.2867943346500397,
0.2709660530090332,
0.31332075595855713,
0.03521735593676567,
-0.22747769951820374,
-0.13300354778766632,
0.09227846562862396,
0.47497692704200745... |
https://github.com/huggingface/datasets/issues/3808 | Ok - this is still an issue but I believe the root cause is different than I originally thought. I'm now able to get caching to work consistently with the above example as long as I fix the python hash seed `export PYTHONHASHSEED=1234` | Pre-Processing Cache Fails when using a Factory pattern | ## Describe the bug
If you utilize a pre-processing function which is created using a factory pattern, the function hash changes on each run (even if the function is identical) and therefore the data will be reproduced each time.
## Steps to reproduce the bug
```python
def preprocess_function_factory(augmenta... | 43 | Pre-Processing Cache Fails when using a Factory pattern
## Describe the bug
If you utilize a pre-processing function which is created using a factory pattern, the function hash changes on each run (even if the function is identical) and therefore the data will be reproduced each time.
## Steps to reproduce the bu... | [
0.045696601271629333,
0.12486892938613892,
-0.07195926457643509,
0.014907030388712883,
0.19251008331775665,
-0.24582426249980927,
0.30334869027137756,
0.25059089064598083,
-0.17094284296035767,
-0.0935288518667221,
0.38170790672302246,
0.17649048566818237,
-0.057801637798547745,
0.15927414... |
https://github.com/huggingface/datasets/issues/3808 | Hi!
Yes, our hasher should work with decorators. For instance, this dummy example:
```python
def f(arg):
def f1(ex):
return {"a": ex["col1"] + arg}
return f1
```
gives the same hash across different Python sessions (`datasets.fingerprint.Hasher.hash(f("string1")` returns `"408c9059f89dbd6c"` ... | Pre-Processing Cache Fails when using a Factory pattern | ## Describe the bug
If you utilize a pre-processing function which is created using a factory pattern, the function hash changes on each run (even if the function is identical) and therefore the data will be reproduced each time.
## Steps to reproduce the bug
```python
def preprocess_function_factory(augmenta... | 76 | Pre-Processing Cache Fails when using a Factory pattern
## Describe the bug
If you utilize a pre-processing function which is created using a factory pattern, the function hash changes on each run (even if the function is identical) and therefore the data will be reproduced each time.
## Steps to reproduce the bu... | [
-0.0620613768696785,
0.12366876006126404,
-0.09023395925760269,
0.09065454453229904,
0.1542712301015854,
-0.23903822898864746,
0.3725143074989319,
0.19262230396270752,
-0.24231044948101044,
0.10529597103595734,
0.4187677502632141,
0.14369621872901917,
-0.26276248693466187,
0.06271073222160... |
https://github.com/huggingface/datasets/issues/3807 | Hi @afcruzs-ms, thanks for opening this separate issue for your problem.
The root problem in the other issue (#3792) was a change in the service of Google Drive.
But in your case, the `xcopa` dataset is not hosted on Google Drive. Therefore, the root cause should be a different one.
Let me look at it... | NonMatchingChecksumError in xcopa dataset | ## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be loaded correctly.
## Actual results
Fails ... | 55 | NonMatchingChecksumError in xcopa dataset
## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be lo... | [
-0.13779962062835693,
0.15731245279312134,
-0.054184988141059875,
0.11445808410644531,
0.08197665959596634,
-0.018044687807559967,
0.17932552099227905,
0.2524215281009674,
0.20484401285648346,
0.17849290370941162,
0.17943547666072845,
0.10884588956832886,
0.22947905957698822,
0.05968272686... |
https://github.com/huggingface/datasets/issues/3807 | @afcruzs-ms, I'm not able to reproduce the issue you reported:
```python
In [1]: from datasets import load_dataset
...: dataset = load_dataset("xcopa", "it")
Downloading builder script: 5.21kB [00:00, 2.75MB/s] ... | NonMatchingChecksumError in xcopa dataset | ## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be loaded correctly.
## Actual results
Fails ... | 134 | NonMatchingChecksumError in xcopa dataset
## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be lo... | [
-0.1630280315876007,
0.019622087478637695,
-0.03254516050219536,
0.15580055117607117,
0.09577509760856628,
0.0057955533266067505,
-0.0534990094602108,
0.33961278200149536,
0.16124355792999268,
0.14249208569526672,
0.07432848960161209,
0.2626977860927582,
0.24026155471801758,
-0.04657915979... |
https://github.com/huggingface/datasets/issues/3807 | unfortunately, i am having a similar problem with the irc_disentaglement dataset :/
my code:
```
from datasets import load_dataset
dataset = load_dataset("irc_disentangle", download_mode="force_redownload")
```
however, it produces the same error as @afcruzs-ms
```
[38](file:///Library/Frameworks/Python.fr... | NonMatchingChecksumError in xcopa dataset | ## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be loaded correctly.
## Actual results
Fails ... | 205 | NonMatchingChecksumError in xcopa dataset
## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be lo... | [
-0.1971902698278427,
0.13928525149822235,
0.029066242277622223,
0.253673255443573,
0.08791641145944595,
0.008565530180931091,
0.05496474355459213,
0.28888946771621704,
0.3888357877731323,
0.14704246819019318,
-0.014185760170221329,
0.39748436212539673,
0.29787325859069824,
-0.1480469405651... |
https://github.com/huggingface/datasets/issues/3807 | Thanks @labouz for reporting: yes, better opening a new GitHub issue as you did. I'm addressing it:
- #4376 | NonMatchingChecksumError in xcopa dataset | ## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be loaded correctly.
## Actual results
Fails ... | 19 | NonMatchingChecksumError in xcopa dataset
## Describe the bug
Loading the xcopa dataset doesn't work, it fails due to a mismatch in the checksum.
## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("xcopa", "it")
```
## Expected results
The dataset should be lo... | [
-0.1311446726322174,
0.04470427334308624,
-0.02775384485721588,
0.10829128324985504,
0.1141074001789093,
-0.020578093826770782,
-0.01739710196852684,
0.3379514813423157,
0.18836160004138947,
0.12099549174308777,
0.12528783082962036,
0.2861757278442383,
0.2626546025276184,
-0.05870484188199... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.