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/5856 | Hi! You can avoid this error by using the preprocessed version:
```python
import datasets
ds = datasets.load_dataset('natural_questions')
```
PS: Once we finish https://github.com/huggingface/datasets/pull/5364, this error will no longer be a problem. | Error loading natural_questions | ### Describe the bug
When try to load natural_questions through datasets == 2.12.0 with python == 3.8.9:
```python
import datasets
datasets.load_dataset('natural_questions',beam_runner='DirectRunner')
```
It failed with following info:
`pyarrow.lib.ArrowNotImplementedError: Nested data conversions not impl... | 31 | Error loading natural_questions
### Describe the bug
When try to load natural_questions through datasets == 2.12.0 with python == 3.8.9:
```python
import datasets
datasets.load_dataset('natural_questions',beam_runner='DirectRunner')
```
It failed with following info:
`pyarrow.lib.ArrowNotImplementedError... | [
-0.373348593711853,
0.09020346403121948,
-0.024216623976826668,
0.4655776023864746,
0.35770848393440247,
-0.21396303176879883,
0.1596500426530838,
0.511631965637207,
0.04409634321928024,
0.017488248646259308,
-0.09487833082675934,
0.6436846256256104,
-0.23180031776428223,
-0.04579784348607... |
https://github.com/huggingface/datasets/issues/5856 | > Hi! You can avoid this error by using the preprocessed version:
>
> ```python
> import datasets
> ds = datasets.load_dataset('natural_questions')
> ```
>
> PS: Once we finish #5364, this error will no longer be a problem.
Thanks, wish #5364 finish early | Error loading natural_questions | ### Describe the bug
When try to load natural_questions through datasets == 2.12.0 with python == 3.8.9:
```python
import datasets
datasets.load_dataset('natural_questions',beam_runner='DirectRunner')
```
It failed with following info:
`pyarrow.lib.ArrowNotImplementedError: Nested data conversions not impl... | 44 | Error loading natural_questions
### Describe the bug
When try to load natural_questions through datasets == 2.12.0 with python == 3.8.9:
```python
import datasets
datasets.load_dataset('natural_questions',beam_runner='DirectRunner')
```
It failed with following info:
`pyarrow.lib.ArrowNotImplementedError... | [
-0.373348593711853,
0.09020346403121948,
-0.024216623976826668,
0.4655776023864746,
0.35770848393440247,
-0.21396303176879883,
0.1596500426530838,
0.511631965637207,
0.04409634321928024,
0.017488248646259308,
-0.09487833082675934,
0.6436846256256104,
-0.23180031776428223,
-0.04579784348607... |
https://github.com/huggingface/datasets/issues/5855 | Cc @amyeroberts @Rocketknight1
Indded I think it's because it does something like this under the hood when there's no multiprocessing:
```python
tf_dataset = tf_dataset.shuffle(len(dataset))
```
PS: with multiprocessing it appears to be different:
```python
indices = np.arange(len(dataset))
if shuffle:... | `to_tf_dataset` consumes too much memory | ### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle behavior. The [source code](https://github.... | 41 | `to_tf_dataset` consumes too much memory
### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle ... | [
-0.15822795033454895,
-0.23407281935214996,
0.09227368235588074,
0.3770810663700104,
0.5104161500930786,
0.20725011825561523,
-0.034461960196495056,
0.27628540992736816,
-0.2560502588748932,
0.33825477957725525,
-0.024285780265927315,
0.16308927536010742,
-0.44833457469940186,
-0.052459832... |
https://github.com/huggingface/datasets/issues/5855 | Hi @massquantity, the dataset being shuffled there is not the full dataset. If you look at [the line above](https://github.com/huggingface/datasets/blob/main/src/datasets/utils/tf_utils.py#L182), the dataset is actually just a single indices array at that point, and that array is the only thing that gets fully loaded i... | `to_tf_dataset` consumes too much memory | ### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle behavior. The [source code](https://github.... | 134 | `to_tf_dataset` consumes too much memory
### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle ... | [
-0.15822795033454895,
-0.23407281935214996,
0.09227368235588074,
0.3770810663700104,
0.5104161500930786,
0.20725011825561523,
-0.034461960196495056,
0.27628540992736816,
-0.2560502588748932,
0.33825477957725525,
-0.024285780265927315,
0.16308927536010742,
-0.44833457469940186,
-0.052459832... |
https://github.com/huggingface/datasets/issues/5855 | Hi @Rocketknight1, you're right and I also noticed that only indices are used in shuffling. My data has shape (50000000, 10), but really the problem doesn't relate to a specific dataset. Simply running the following code costs me 10GB of memory.
```python
from datasets import Dataset
def gen():
for i in ran... | `to_tf_dataset` consumes too much memory | ### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle behavior. The [source code](https://github.... | 160 | `to_tf_dataset` consumes too much memory
### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle ... | [
-0.15822795033454895,
-0.23407281935214996,
0.09227368235588074,
0.3770810663700104,
0.5104161500930786,
0.20725011825561523,
-0.034461960196495056,
0.27628540992736816,
-0.2560502588748932,
0.33825477957725525,
-0.024285780265927315,
0.16308927536010742,
-0.44833457469940186,
-0.052459832... |
https://github.com/huggingface/datasets/issues/5855 | Thanks for that reproduction script - I've confirmed the same issue is occurring for me. Investigating it now! | `to_tf_dataset` consumes too much memory | ### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle behavior. The [source code](https://github.... | 18 | `to_tf_dataset` consumes too much memory
### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle ... | [
-0.15822795033454895,
-0.23407281935214996,
0.09227368235588074,
0.3770810663700104,
0.5104161500930786,
0.20725011825561523,
-0.034461960196495056,
0.27628540992736816,
-0.2560502588748932,
0.33825477957725525,
-0.024285780265927315,
0.16308927536010742,
-0.44833457469940186,
-0.052459832... |
https://github.com/huggingface/datasets/issues/5855 | Update: The memory usage is occurring in creation of the index and shuffle buffer. You can reproduce it very simply with:
```python
import tensorflow as tf
indices = tf.range(50_000_000, dtype=tf.int64)
dataset = tf.data.Dataset.from_tensor_slices(indices)
dataset = dataset.shuffle(len(dataset))
print(next(iter... | `to_tf_dataset` consumes too much memory | ### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle behavior. The [source code](https://github.... | 99 | `to_tf_dataset` consumes too much memory
### Describe the bug
Hi, I'm using `to_tf_dataset` to convert a _large_ dataset to `tf.data.Dataset`. I observed that the data loading *before* training took a lot of time and memory, even with `batch_size=1`.
After some digging, i believe the reason lies in the shuffle ... | [
-0.15822795033454895,
-0.23407281935214996,
0.09227368235588074,
0.3770810663700104,
0.5104161500930786,
0.20725011825561523,
-0.034461960196495056,
0.27628540992736816,
-0.2560502588748932,
0.33825477957725525,
-0.024285780265927315,
0.16308927536010742,
-0.44833457469940186,
-0.052459832... |
https://github.com/huggingface/datasets/issues/5854 | Hi! `audiofolder` requires `datasets>=2.5.0`, so please update the `datasets`' installation (`pip install -U datasets`) in the environment (and restart the env for the update to take effect) to resolve the issue. | Can not load audiofolder dataset on kaggle | ### Describe the bug
It's crash log:
FileNotFoundError: Couldn't find a dataset script at /kaggle/working/audiofolder/audiofolder.py or any data file in the same directory. Couldn't find 'audiofolder' on the Hugging Face Hub either: FileNotFoundError: Couldn't find file at https://raw.githubusercontent.com/huggingfac... | 31 | Can not load audiofolder dataset on kaggle
### Describe the bug
It's crash log:
FileNotFoundError: Couldn't find a dataset script at /kaggle/working/audiofolder/audiofolder.py or any data file in the same directory. Couldn't find 'audiofolder' on the Hugging Face Hub either: FileNotFoundError: Couldn't find file at... | [
-0.33059805631637573,
-0.3415958285331726,
0.04055991768836975,
0.6100748777389526,
0.37537503242492676,
-0.13818277418613434,
0.2699607312679291,
-0.027829112485051155,
0.33419615030288696,
0.1420486867427826,
-0.3253091275691986,
0.5460795760154724,
-0.33598366379737854,
0.38824459910392... |
https://github.com/huggingface/datasets/issues/5854 | > Hi! `audiofolder` requires `datasets>=2.5.0`, so please update the `datasets`' installation (`pip install -U datasets`) in the environment to resolve the issue.
I don't think it is a problem of the version. It runs ok on colab or local machine. Only on kaggle will has this bug. | Can not load audiofolder dataset on kaggle | ### Describe the bug
It's crash log:
FileNotFoundError: Couldn't find a dataset script at /kaggle/working/audiofolder/audiofolder.py or any data file in the same directory. Couldn't find 'audiofolder' on the Hugging Face Hub either: FileNotFoundError: Couldn't find file at https://raw.githubusercontent.com/huggingfac... | 47 | Can not load audiofolder dataset on kaggle
### Describe the bug
It's crash log:
FileNotFoundError: Couldn't find a dataset script at /kaggle/working/audiofolder/audiofolder.py or any data file in the same directory. Couldn't find 'audiofolder' on the Hugging Face Hub either: FileNotFoundError: Couldn't find file at... | [
-0.32287678122520447,
-0.287451833486557,
0.04319688305258751,
0.5809459686279297,
0.32738155126571655,
-0.14868700504302979,
0.29481619596481323,
-0.014694088138639927,
0.3119141459465027,
0.16970306634902954,
-0.3131583333015442,
0.5365290641784668,
-0.352512001991272,
0.3962965309619903... |
https://github.com/huggingface/datasets/issues/5854 | Based on your dataset info, the installed version is `2.1.0`, which does not include `audiofolder`.
By default, Kaggle preinstalls `datasets` into a new env, but the version it installs is outdated and does not contain newer features such as `audiofolder` | Can not load audiofolder dataset on kaggle | ### Describe the bug
It's crash log:
FileNotFoundError: Couldn't find a dataset script at /kaggle/working/audiofolder/audiofolder.py or any data file in the same directory. Couldn't find 'audiofolder' on the Hugging Face Hub either: FileNotFoundError: Couldn't find file at https://raw.githubusercontent.com/huggingfac... | 40 | Can not load audiofolder dataset on kaggle
### Describe the bug
It's crash log:
FileNotFoundError: Couldn't find a dataset script at /kaggle/working/audiofolder/audiofolder.py or any data file in the same directory. Couldn't find 'audiofolder' on the Hugging Face Hub either: FileNotFoundError: Couldn't find file at... | [
-0.3101239502429962,
-0.32566794753074646,
0.0438629575073719,
0.5551077127456665,
0.32403919100761414,
-0.10100151598453522,
0.24710452556610107,
-0.008645685389637947,
0.31990259885787964,
0.11650456488132477,
-0.3348088562488556,
0.5558660626411438,
-0.33686235547065735,
0.3863137662410... |
https://github.com/huggingface/datasets/issues/5847 | [`KeyDataset`](https://github.com/huggingface/transformers/blob/7f8b909189547944617741d8d3c6c84504701693/src/transformers/pipelines/pt_utils.py#L296) doesn't support iterable datasets, so you either need to implement a version that does (and also indexing nested (translation) fields):
```python
from torch.utils.dat... | Streaming IterableDataset not working with translation pipeline | ### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that IterableDataset has no len().
### Steps to reproduce the bug
CODE:
```
... | 128 | Streaming IterableDataset not working with translation pipeline
### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that Iterable... | [
-0.5097402334213257,
-0.1700969636440277,
-0.03699735552072525,
0.1410585343837738,
0.32595449686050415,
-0.04014071822166443,
0.5023482441902161,
0.08091886341571808,
-0.21698808670043945,
0.3707205653190613,
0.11250735074281693,
-0.12919551134109497,
-0.09642905741930008,
-0.056972090154... |
https://github.com/huggingface/datasets/issues/5847 | @mariosasko The map snippet didn't quite work, but gave me enough of a clue to get it working. The following snippet does work:
```
def en_translation(x):
return {"en":x['translation']['en']}
ks = ds.map(en_translation, remove_columns=['translation'])
test=[]
for x in iter(ks):
test.append(x['en'])
xx=... | Streaming IterableDataset not working with translation pipeline | ### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that IterableDataset has no len().
### Steps to reproduce the bug
CODE:
```
... | 219 | Streaming IterableDataset not working with translation pipeline
### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that Iterable... | [
-0.5097402334213257,
-0.1700969636440277,
-0.03699735552072525,
0.1410585343837738,
0.32595449686050415,
-0.04014071822166443,
0.5023482441902161,
0.08091886341571808,
-0.21698808670043945,
0.3707205653190613,
0.11250735074281693,
-0.12919551134109497,
-0.09642905741930008,
-0.056972090154... |
https://github.com/huggingface/datasets/issues/5847 | So perhaps there's no bug exactly, but I would love to see two things: 1) improve the documentation to better understand what's really getting returned. 2) update the example provided of using transformer pipeline with a dataset to include the oddball case that translation appears to be. | Streaming IterableDataset not working with translation pipeline | ### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that IterableDataset has no len().
### Steps to reproduce the bug
CODE:
```
... | 47 | Streaming IterableDataset not working with translation pipeline
### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that Iterable... | [
-0.5097402334213257,
-0.1700969636440277,
-0.03699735552072525,
0.1410585343837738,
0.32595449686050415,
-0.04014071822166443,
0.5023482441902161,
0.08091886341571808,
-0.21698808670043945,
0.3707205653190613,
0.11250735074281693,
-0.12919551134109497,
-0.09642905741930008,
-0.056972090154... |
https://github.com/huggingface/datasets/issues/5847 | Hi,
for the original snippet, the issue is that `streaming` datasets are not countable (they have no len) and therefore `KeyDataset` cannot work with them ( KeyDataset is a dataset and therefore requires a length).
I modified slightly the original snippet to make it work:
```python
from transformers import pi... | Streaming IterableDataset not working with translation pipeline | ### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that IterableDataset has no len().
### Steps to reproduce the bug
CODE:
```
... | 166 | Streaming IterableDataset not working with translation pipeline
### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that Iterable... | [
-0.5097402334213257,
-0.1700969636440277,
-0.03699735552072525,
0.1410585343837738,
0.32595449686050415,
-0.04014071822166443,
0.5023482441902161,
0.08091886341571808,
-0.21698808670043945,
0.3707205653190613,
0.11250735074281693,
-0.12919551134109497,
-0.09642905741930008,
-0.056972090154... |
https://github.com/huggingface/datasets/issues/5847 | Maybe moving `KeyDataset` later on in the guide and specify it's mostly for streaming then? Or is it also necessary for batch_size>1 (which is what the current doc implies)? | Streaming IterableDataset not working with translation pipeline | ### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that IterableDataset has no len().
### Steps to reproduce the bug
CODE:
```
... | 29 | Streaming IterableDataset not working with translation pipeline
### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that Iterable... | [
-0.5097402334213257,
-0.1700969636440277,
-0.03699735552072525,
0.1410585343837738,
0.32595449686050415,
-0.04014071822166443,
0.5023482441902161,
0.08091886341571808,
-0.21698808670043945,
0.3707205653190613,
0.11250735074281693,
-0.12919551134109497,
-0.09642905741930008,
-0.056972090154... |
https://github.com/huggingface/datasets/issues/5847 | Hmm
Iterator (`yield`) :
- Not countable
- Super flexible
- Cannot use `num_workers>1` (threading requires indexing at the correct location, iterators require to iterate in order,so each thread would iterate over the full thing being genuinely a bad idea)
- Can batch
- tqdm doesn't show a nice progress bar (it ... | Streaming IterableDataset not working with translation pipeline | ### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that IterableDataset has no len().
### Steps to reproduce the bug
CODE:
```
... | 188 | Streaming IterableDataset not working with translation pipeline
### Describe the bug
I'm trying to use a streaming dataset for translation inference to avoid downloading the training data.
I'm using a pipeline and a dataset, and following the guidance in the tutorial.
Instead I get an exception that Iterable... | [
-0.5097402334213257,
-0.1700969636440277,
-0.03699735552072525,
0.1410585343837738,
0.32595449686050415,
-0.04014071822166443,
0.5023482441902161,
0.08091886341571808,
-0.21698808670043945,
0.3707205653190613,
0.11250735074281693,
-0.12919551134109497,
-0.09642905741930008,
-0.056972090154... |
https://github.com/huggingface/datasets/issues/5846 | This is due to the slow resolution of the data files: https://github.com/huggingface/datasets/issues/5537.
We plan to switch to `huggingface_hub`'s `HfFileSystem` soon to make the resolution faster (will be up to 20x faster once we merge https://github.com/huggingface/huggingface_hub/pull/1443)
| load_dataset('bigcode/the-stack-dedup', streaming=True) very slow! | ### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one or two seconds.
### Environment info
- `datasets` version: 2.1... | 35 | load_dataset('bigcode/the-stack-dedup', streaming=True) very slow!
### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one o... | [
-0.5835384130477905,
-0.4485812187194824,
-0.06136574596166611,
0.2794361114501953,
0.23265546560287476,
0.024867437779903412,
0.07487990707159042,
0.212795227766037,
0.10379171371459961,
-0.05269263684749603,
-0.15228033065795898,
0.2630672752857208,
-0.02050098590552807,
0.43237778544425... |
https://github.com/huggingface/datasets/issues/5846 | You're right, when I try to parse more than 50GB of text data, I also get very slow, usually taking hours or even tens of hours. | load_dataset('bigcode/the-stack-dedup', streaming=True) very slow! | ### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one or two seconds.
### Environment info
- `datasets` version: 2.1... | 26 | load_dataset('bigcode/the-stack-dedup', streaming=True) very slow!
### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one o... | [
-0.6187251806259155,
-0.1993492990732193,
-0.1015913113951683,
0.25945907831192017,
0.1967843770980835,
0.07397590577602386,
0.15669342875480652,
0.23013529181480408,
0.05619114637374878,
-0.0416470468044281,
-0.04566510394215584,
0.29201459884643555,
0.07512497901916504,
0.285486608743667... |
https://github.com/huggingface/datasets/issues/5846 | > You're right, when I try to parse more than 50GB of text data, I also get very slow, usually taking hours or even tens of hours.
That's unrelated to the problem discussed in this issue. | load_dataset('bigcode/the-stack-dedup', streaming=True) very slow! | ### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one or two seconds.
### Environment info
- `datasets` version: 2.1... | 36 | load_dataset('bigcode/the-stack-dedup', streaming=True) very slow!
### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one o... | [
-0.5950614809989929,
-0.1920388787984848,
-0.10719016194343567,
0.20276544988155365,
0.20339643955230713,
0.03242965042591095,
0.18380331993103027,
0.25342410802841187,
0.05174969136714935,
-0.013370223343372345,
-0.057919375598430634,
0.3059339225292206,
0.05831361562013626,
0.30812805891... |
https://github.com/huggingface/datasets/issues/5846 | > > You're right, when I try to parse more than 50GB of text data, I also get very slow, usually taking hours or even tens of hours.
>
> That's unrelated to the problem discussed in this issue.
Sorry, I misunderstood it. | load_dataset('bigcode/the-stack-dedup', streaming=True) very slow! | ### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one or two seconds.
### Environment info
- `datasets` version: 2.1... | 43 | load_dataset('bigcode/the-stack-dedup', streaming=True) very slow!
### Describe the bug
Running
```
import datasets
ds = datasets.load_dataset('bigcode/the-stack-dedup', streaming=True)
```
takes about 2.5 minutes!
I would expect this to be near instantaneous. With other datasets, the runtime is one o... | [
-0.5947154760360718,
-0.24217668175697327,
-0.1078680157661438,
0.22747907042503357,
0.21352587640285492,
0.05493667721748352,
0.16488012671470642,
0.23586004972457886,
0.05870816856622696,
-0.010351285338401794,
-0.08272765576839447,
0.3045477569103241,
0.06317373365163803,
0.317029505968... |
https://github.com/huggingface/datasets/issues/5841 | Hi ! You can try to use the [Image](https://huggingface.co/docs/datasets/v2.12.0/en/package_reference/main_classes#datasets.Image) type which [decodes images on-the-fly](https://huggingface.co/docs/datasets/v2.12.0/en/about_dataset_features#image-feature) into pytorch tensors :)
```python
ds = Dataset.from_dict({"t... | Abusurdly slow on iteration | ### Describe the bug
I am attempting to iterate through an image dataset, but I am encountering a significant slowdown in the iteration speed. In order to investigate this issue, I conducted the following experiment:
```python
a=torch.randn(100,224)
a=torch.stack([a] * 10000)
a.shape
# %%
ds=Dataset.from_d... | 226 | Abusurdly slow on iteration
### Describe the bug
I am attempting to iterate through an image dataset, but I am encountering a significant slowdown in the iteration speed. In order to investigate this issue, I conducted the following experiment:
```python
a=torch.randn(100,224)
a=torch.stack([a] * 10000)
a.sh... | [
-0.1845068335533142,
-0.5200334191322327,
-0.04702053964138031,
0.2566430866718292,
0.12682995200157166,
0.1570071130990982,
0.481078177690506,
0.219479501247406,
-0.2652483582496643,
0.1919068694114685,
-0.06385338306427002,
0.5231744050979614,
-0.08375862240791321,
-0.4237799346446991,
... |
https://github.com/huggingface/datasets/issues/5841 | Anyway, regarding the speed difference between numpy and pytorch, I think the issue is that we first convert numpy sub-arrays to pytorch and then consolidate into one tensor, while we should to the opposite. Indeed converting a numpy array to pytorch has a fix cost that seems to cause a slow down. The current pipeline ... | Abusurdly slow on iteration | ### Describe the bug
I am attempting to iterate through an image dataset, but I am encountering a significant slowdown in the iteration speed. In order to investigate this issue, I conducted the following experiment:
```python
a=torch.randn(100,224)
a=torch.stack([a] * 10000)
a.shape
# %%
ds=Dataset.from_d... | 91 | Abusurdly slow on iteration
### Describe the bug
I am attempting to iterate through an image dataset, but I am encountering a significant slowdown in the iteration speed. In order to investigate this issue, I conducted the following experiment:
```python
a=torch.randn(100,224)
a=torch.stack([a] * 10000)
a.sh... | [
-0.1845068335533142,
-0.5200334191322327,
-0.04702053964138031,
0.2566430866718292,
0.12682995200157166,
0.1570071130990982,
0.481078177690506,
0.219479501247406,
-0.2652483582496643,
0.1919068694114685,
-0.06385338306427002,
0.5231744050979614,
-0.08375862240791321,
-0.4237799346446991,
... |
https://github.com/huggingface/datasets/issues/5841 | I have a similar issue: iterating over a dataset takes 5s without applying any transform, but takes ~30s after applying a transform.
Here is the minimum code to reproduce the problem
```python
import numpy as np
from datasets import Dataset, DatasetDict, load_dataset, Array3D, Image, Features
from torch.utils.da... | Abusurdly slow on iteration | ### Describe the bug
I am attempting to iterate through an image dataset, but I am encountering a significant slowdown in the iteration speed. In order to investigate this issue, I conducted the following experiment:
```python
a=torch.randn(100,224)
a=torch.stack([a] * 10000)
a.shape
# %%
ds=Dataset.from_d... | 187 | Abusurdly slow on iteration
### Describe the bug
I am attempting to iterate through an image dataset, but I am encountering a significant slowdown in the iteration speed. In order to investigate this issue, I conducted the following experiment:
```python
a=torch.randn(100,224)
a=torch.stack([a] * 10000)
a.sh... | [
-0.1845068335533142,
-0.5200334191322327,
-0.04702053964138031,
0.2566430866718292,
0.12682995200157166,
0.1570071130990982,
0.481078177690506,
0.219479501247406,
-0.2652483582496643,
0.1919068694114685,
-0.06385338306427002,
0.5231744050979614,
-0.08375862240791321,
-0.4237799346446991,
... |
https://github.com/huggingface/datasets/issues/5842 | This method has been recently added to the `IterableDataset`, so you need to update the `datasets`' installation (`pip install -U datasets`) to use it. | Remove columns in interable dataset | ### Feature request
Right now, remove_columns() produces a NotImplementedError for iterable style datasets
### Motivation
It would be great to have the same functionality irrespective of whether one is using an iterable or a map-style dataset
### Your contribution
hope and courage. | 24 | Remove columns in interable dataset
### Feature request
Right now, remove_columns() produces a NotImplementedError for iterable style datasets
### Motivation
It would be great to have the same functionality irrespective of whether one is using an iterable or a map-style dataset
### Your contribution
hope and cou... | [
-0.1692637801170349,
-0.11372427642345428,
-0.22988520562648773,
0.10435251891613007,
0.3636786937713623,
0.12261197715997696,
0.3428803086280823,
0.3954080641269684,
0.1778421252965927,
0.3825918436050415,
-0.41839078068733215,
0.6086258888244629,
-0.04357260465621948,
0.09833016991615295... |
https://github.com/huggingface/datasets/issues/5838 | As the name says, `load_from_disk` load the data from your disk. If the data is hosted on S3, it is first downloaded locally and then loaded from your disk.
There is a discussion on streaming data from S3 here though: #5281 | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 41 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.19136852025985718,
-0.42585277557373047,
-0.1447661966085434,
0.17709669470787048,
-0.05689805746078491,
-0.22552555799484253,
-0.010217159986495972,
-0.014616171829402447,
0.27002018690109253,
0.0845087319612503,
0.007977951318025589,
0.35091349482536316,
-0.015262658707797527,
0.43650... |
https://github.com/huggingface/datasets/issues/5838 | @lhoestq
Thanks for your comment. I have checked out the discussion before and attempted at replicating the mentioned changes in the main branch (#5580). What I found was that if a dataset is saved using `save_to_disk`, it cannot be read by `load_dataset`. The error message asks me to to use `load_from_disk` instead.... | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 65 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.21534283459186554,
-0.13543955981731415,
-0.10849569737911224,
0.0045851487666368484,
0.05231412127614021,
-0.1832382082939148,
-0.017614327371120453,
0.03951312601566315,
0.18168671429157257,
0.08003409206867218,
-0.10446079075336456,
0.31715551018714905,
-0.14613616466522217,
0.398148... |
https://github.com/huggingface/datasets/issues/5838 | Using `push_to_hub` you can save the dataset on the HF Hub as parquet files, and reload it / stream it using `load_dataset` :)
If you want to save your dataset somewhere else you can use `.to_parquet` to get a parquet file. If your dataset is big it's usually recommended to shard it into multi parquet files (around ... | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 59 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.4376581907272339,
-0.3938426971435547,
-0.08103150129318237,
0.10626921057701111,
-0.11427478492259979,
-0.12265785783529282,
-0.11332442611455917,
0.1117803156375885,
0.2856537699699402,
0.0019084811210632324,
-0.309587299823761,
0.2367701232433319,
-0.0507931187748909,
0.5266981124877... |
https://github.com/huggingface/datasets/issues/5838 | @lhoestq
I tried the method you mentioned. This the current scenario I'm facing:
- The parquet file can be read from disk and streaming can be enabled.
- The parquet file can be read from `s3` (local MinIO).
- When `streaming=True` is enabled for `s3`, I get the error mentioned below:
```
File ~/.../lib/pytho... | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 106 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.18137714266777039,
-0.354874849319458,
0.02730192244052887,
0.2140737622976303,
-0.07594193518161774,
-0.16613176465034485,
0.1421043574810028,
0.024777110666036606,
0.29747745394706726,
-0.005054458975791931,
-0.0833112895488739,
0.3988795876502991,
0.07521635293960571,
0.5150984525680... |
https://github.com/huggingface/datasets/issues/5838 | Streaming from S3 is still experimental, there might be a few bugs unfortunately.
Can you share the full stack trace ? | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 21 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.23841492831707,
-0.394839882850647,
-0.1468847393989563,
0.1194128692150116,
-0.00791324581950903,
-0.2820230722427368,
-0.0357992984354496,
0.03679409250617027,
0.18196415901184082,
0.06367497146129608,
-0.009758085012435913,
0.36067014932632446,
-0.07663559913635254,
0.469088912010192... |
https://github.com/huggingface/datasets/issues/5838 | @lhoestq
Sure, here you go:
```python
TypeError Traceback (most recent call last)
Cell In[8], line 1
----> 1 dataset = load_dataset("parquet", data_files=["s3://<bucket name>/<data folder>/data-parquet"], storage_options=fs.storage_options, streaming=True)
File ~/.../datasets/... | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 1,037 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.34546247124671936,
-0.24966630339622498,
-0.13603632152080536,
0.11836671829223633,
0.07666708528995514,
-0.06216322258114815,
0.1502552330493927,
0.13144414126873016,
0.30351272225379944,
0.00011910498142242432,
-0.09259092807769775,
0.3016706705093384,
-0.1856059581041336,
0.496601611... |
https://github.com/huggingface/datasets/issues/5838 | Is `"data-parquet"` a file ? In `data_files` you should pass the paths to the parquet files (not to a directory). Glob patterns are not supported yet for S3 URLs.
The bug seems to happen because your provided data file has no extension. Because of that it tries to infer it from the file content, but fails because `_... | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 63 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.12845727801322937,
-0.3317633867263794,
0.01568622887134552,
0.48633962869644165,
0.08161306381225586,
-0.41716980934143066,
-0.05320350453257561,
0.0006679370999336243,
0.014363624155521393,
0.1316562443971634,
0.08848384022712708,
0.4564896821975708,
-0.10532784461975098,
0.4703978300... |
https://github.com/huggingface/datasets/issues/5838 | @lhoestq
Thank you for your answer. Saving the file with `.parquet` extension solved the issue! This is really great! Really appreciate all the help!
Let me know if I should close the issue or feel free to close it if you want. | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 42 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.20789626240730286,
-0.30446773767471313,
-0.10579246282577515,
0.19229444861412048,
0.19193899631500244,
-0.22231219708919525,
0.05201942101120949,
0.08504946529865265,
0.19627942144870758,
0.0800582766532898,
0.08831930160522461,
0.37090200185775757,
-0.04738371819257736,
0.45688819885... |
https://github.com/huggingface/datasets/issues/5838 | Cool ! I'm glad it worked out :)
Sure feel free to close the issue, since the original question about streaming with load_from_disk has been answered anyway | Streaming support for `load_from_disk` | ### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very large and being able to stream the data ... | 27 | Streaming support for `load_from_disk`
### Feature request
Support for streaming datasets stored in object stores in `load_from_disk`.
### Motivation
The `load_from_disk` function supports fetching datasets stored in object stores such as `s3`. In many cases, the datasets that are stored in object stores are very... | [
-0.16210711002349854,
-0.34814807772636414,
-0.19091832637786865,
0.09209589660167694,
0.08707378804683685,
-0.3386496305465698,
0.03595179691910744,
0.10286790877580643,
0.17164400219917297,
0.07638229429721832,
0.0013486891984939575,
0.3652254045009613,
-0.061528172343969345,
0.455440253... |
https://github.com/huggingface/datasets/issues/5837 | Hi ! Doing `load_dataset("path/to/data.csv")` is not supported yet, but you can do
```python
ds = load_dataset("csv", data_files=["path/to/data.csv"])
``` | Use DeepSpeed load myself " .csv " dataset. | ### Describe the bug
When I use DeepSpeed train a model with my own " XXX.csv" dataset I got the follow question:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/fm001/.conda/envs/hzl/lib/python3.8/site-packages/datasets/load.py", line 1767, in load_dataset
builder_instan... | 18 | Use DeepSpeed load myself " .csv " dataset.
### Describe the bug
When I use DeepSpeed train a model with my own " XXX.csv" dataset I got the follow question:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/fm001/.conda/envs/hzl/lib/python3.8/site-packages/datasets/load.py", ... | [
-0.25815436244010925,
-0.10589193552732468,
-0.02953505702316761,
0.3224860429763794,
0.14691399037837982,
0.2572140395641327,
0.5154327154159546,
0.1292022317647934,
0.23117075860500336,
0.2553311586380005,
0.03211133927106857,
0.17823092639446259,
-0.12606866657733917,
0.3205708861351013... |
https://github.com/huggingface/datasets/issues/5837 | The other question:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/fm001/.conda/envs/hzl/lib/python3.8/site-packages/datasets/load.py", line 1767, in load_dataset
builder_instance = load_dataset_builder(
File "/home/fm001/.conda/envs/hzl/lib/python3.8/site-packages/da... | Use DeepSpeed load myself " .csv " dataset. | ### Describe the bug
When I use DeepSpeed train a model with my own " XXX.csv" dataset I got the follow question:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/fm001/.conda/envs/hzl/lib/python3.8/site-packages/datasets/load.py", line 1767, in load_dataset
builder_instan... | 229 | Use DeepSpeed load myself " .csv " dataset.
### Describe the bug
When I use DeepSpeed train a model with my own " XXX.csv" dataset I got the follow question:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/fm001/.conda/envs/hzl/lib/python3.8/site-packages/datasets/load.py", ... | [
-0.2316916137933731,
-0.08913345634937286,
-0.03008565865457058,
0.3897944390773773,
0.13162685930728912,
0.2130463421344757,
0.546649158000946,
0.09271926432847977,
0.23029524087905884,
0.25812938809394836,
0.02489587664604187,
0.17865730822086334,
-0.06958972662687302,
0.3221999406814575... |
https://github.com/huggingface/datasets/issues/5834 | Hi ! The numpy formatting detaults to int64 and float32 - but you can use uint8 using
```python
ds = ds.with_format("numpy", dtype=np.uint8)
``` | Is uint8 supported? | ### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as well.
Is there a way ... | 23 | Is uint8 supported?
### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as ... | [
-0.0773063451051712,
-0.37703418731689453,
0.007305428385734558,
0.2908572554588318,
0.5897737145423889,
-0.02267012745141983,
0.5053195953369141,
0.2177438735961914,
0.016887344419956207,
-0.12525634467601776,
-0.2633092701435089,
0.10247842967510223,
0.05486701801419258,
0.00420443713665... |
https://github.com/huggingface/datasets/issues/5834 | Thank you!
By setting `ds.with_format("numpy", dtype=np.uint8)`, the dataset returns the data in `uint8`.
However, `with_format` and `set_format` seem to cast the data on-the-fly.
I want to reduce the dataset size by using `uint8` instead of `int64` and I observe no difference between using `int64` and `uint8` for... | Is uint8 supported? | ### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as well.
Is there a way ... | 74 | Is uint8 supported?
### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as ... | [
-0.07757597416639328,
-0.31044113636016846,
0.009639076888561249,
0.3099156618118286,
0.6234179735183716,
-0.06039707362651825,
0.44952890276908875,
0.24180036783218384,
-0.0026682615280151367,
-0.12752535939216614,
-0.2918073534965515,
0.16088151931762695,
-0.007652686908841133,
0.0309169... |
https://github.com/huggingface/datasets/issues/5834 | If the feature type is `Value("uint8")` then it's written an uint8 on disk using the uint8 Arrow dtype.
e.g.
```python
ds = Dataset.from_dict({"a": range(10)}, features=Features({"a": Value("uint8")}))
ds.data.nbytes
# 10
``` | Is uint8 supported? | ### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as well.
Is there a way ... | 30 | Is uint8 supported?
### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as ... | [
-0.04128550738096237,
-0.408857136964798,
0.0004546605050563812,
0.26819339394569397,
0.5924240946769714,
-0.038157060742378235,
0.49679771065711975,
0.20345473289489746,
0.04306764155626297,
-0.13159647583961487,
-0.25536489486694336,
0.14052356779575348,
0.033094555139541626,
0.036020174... |
https://github.com/huggingface/datasets/issues/5834 | Oh, I understand now.
The data was stored in `uint8` from the beginning (when the dataset returns `int64`).
Thank you for your time!
My question is fully resolved. | Is uint8 supported? | ### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as well.
Is there a way ... | 28 | Is uint8 supported?
### Describe the bug
I expect the dataset to store the data in the `uint8` data type, but it's returning `int64` instead.
While I've found that `datasets` doesn't yet support float16 (https://github.com/huggingface/datasets/issues/4981), I'm wondering if this is the case for other data types as ... | [
-0.014799565076828003,
-0.42494139075279236,
0.0009363330900669098,
0.2994321882724762,
0.5796529054641724,
-0.0617339164018631,
0.4964897632598877,
0.17840290069580078,
0.015032373368740082,
-0.11957670748233795,
-0.2177356630563736,
0.09623894095420837,
0.0689069926738739,
-0.00031278282... |
https://github.com/huggingface/datasets/issues/5833 | Thanks for reporting, @agombert.
In this case, I think the root issue is authentication: before pushing to Hub, you should authenticate. See our docs: https://huggingface.co/docs/datasets/upload_dataset#upload-with-python
> 2. To upload a dataset on the Hub in Python, you need to log in to your Hugging Face account... | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 50 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | Hey @albertvillanova well I actually did :D
<img width="1079" alt="Capture d’écran 2023-05-09 à 18 02 58" src="https://github.com/huggingface/datasets/assets/17645711/e091aa20-06b1-4dd3-bfdb-35e832c66f8d">
| Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 17 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | That is weird that you get a Forbidden error if you are properly authenticated...
Today we had a big outage issue affecting the Hugging Face Hub. Could you please retry to push_to_hub your dataset? Maybe that was the cause... | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 39 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | Login successful but also got this error "Forbidden: pass `create_pr=1` as a query parameter to create a Pull Request" | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 19 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | Make sure your API token has a `write` role. I had the same issue as you with the `read` token. Creating a `write` token and using that solved the issue. | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 30 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | > Make sure your API token has a `write` role. I had the same issue as you with the `read` token. Creating a `write` token and using that solved the issue.
I generate a token with write role. It works! thank you so much. | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 44 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | @dmitrijsk amazing thanks so much !
The error should be clearer when the token is read-only – I wasted a lot of time there.. | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 24 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | Based on the number of reactions (https://github.com/huggingface/datasets/issues/5833#issuecomment-1586521001), many users have issues debugging this. @Wauplin Maybe a more informative error can be thrown in `hfh` if a token's role is insufficient for an op. WDYT? | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 34 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | Yes indeed. I created an issue some time ago about it: https://github.com/huggingface/huggingface_hub/issues/1653. I'll prioritize it more and let you know. Thanks for the ping. | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 24 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | Hey everyone :wave: The error message has been fixed to be more informative. As mentioned in https://github.com/huggingface/datasets/issues/5833#issuecomment-1586521001, the n°1 reason why this is happening is that a `read` token has been used instead of `write`. The fix has being shipped on the server meaning that you... | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 100 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | @Wauplin Reopening it. Indeed, the above error message is thrown if pushing to an **existing** repo with a `read` token. However, if the repo does not exist and needs to be created (by calling `create_repo` in `push_to_hub`), then passing a `read` token will raise the following:
```python
HTTPError: 403 Client Error:... | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 140 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5833 | > However, if the repo does not exist and needs to be created (by calling create_repo in push_to_hub), then passing a read token will raise the following:
This seems to be a different issue than the one reported above right? Still agree that a more informative message would be nice. Can you open an issue on moon-lan... | Unable to push dataset - `create_pr` problem | ### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
```python
dataset.push_to_hub("agomberto/Fr... | 73 | Unable to push dataset - `create_pr` problem
### Describe the bug
I can't upload to the hub the dataset I manually created locally (Image dataset). I have a problem when using the method `.push_to_hub` which asks for a `create_pr` attribute which is not compatible.
### Steps to reproduce the bug
here what I have:
... | [
-0.3313612937927246,
-0.5114108920097351,
0.0488574355840683,
0.2377043217420578,
0.16082608699798584,
-0.022989168763160706,
0.147964209318161,
0.22991344332695007,
-0.06724781543016434,
0.05386371165513992,
0.11776133626699448,
0.38434380292892456,
-0.08146622776985168,
0.264676392078399... |
https://github.com/huggingface/datasets/issues/5831 | I am experiencing the same problem with the following environment:
* `datasets` version: 2.11.0
* Platform: `Linux 5.19.0-41-generic x86_64 GNU/Linux`
* Python version: `3.8.5`
* Huggingface_hub version: 0.13.3
* PyArrow version: `11.0.0`
* Pandas version: `1.5.3`
Trying to get some diagnostics, I got the fo... | [Bug]504 Server Error when loading dataset which was already cached | ### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it again using the same machine, I got this error:
```
Traceback (most rece... | 120 | [Bug]504 Server Error when loading dataset which was already cached
### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it agai... | [
-0.1444699466228485,
0.002489805221557617,
0.03585709258913994,
0.3607310950756073,
0.345577210187912,
0.07848291844129562,
0.4264727532863617,
0.2986140847206116,
-0.215998575091362,
-0.09981349855661392,
-0.14252692461013794,
0.34242287278175354,
0.07249689847230911,
0.011891968548297882... |
https://github.com/huggingface/datasets/issues/5831 | same problem with a private dataset repo, seems the huggingface hub server got some connection problem? | [Bug]504 Server Error when loading dataset which was already cached | ### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it again using the same machine, I got this error:
```
Traceback (most rece... | 16 | [Bug]504 Server Error when loading dataset which was already cached
### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it agai... | [
-0.1444699466228485,
0.002489805221557617,
0.03585709258913994,
0.3607310950756073,
0.345577210187912,
0.07848291844129562,
0.4264727532863617,
0.2986140847206116,
-0.215998575091362,
-0.09981349855661392,
-0.14252692461013794,
0.34242287278175354,
0.07249689847230911,
0.011891968548297882... |
https://github.com/huggingface/datasets/issues/5831 | @SingL3 You can avoid this error by setting the [`HF_DATASETS_OFFLINE`](https://huggingface.co/docs/datasets/v2.12.0/en/loading#offline) env variable to 1. By default, if an internet connection is available, we check whether the cache of a cached dataset is up-to-date.
@lucidBrot `datasets`' cache is still not alig... | [Bug]504 Server Error when loading dataset which was already cached | ### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it again using the same machine, I got this error:
```
Traceback (most rece... | 48 | [Bug]504 Server Error when loading dataset which was already cached
### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it agai... | [
-0.1444699466228485,
0.002489805221557617,
0.03585709258913994,
0.3607310950756073,
0.345577210187912,
0.07848291844129562,
0.4264727532863617,
0.2986140847206116,
-0.215998575091362,
-0.09981349855661392,
-0.14252692461013794,
0.34242287278175354,
0.07249689847230911,
0.011891968548297882... |
https://github.com/huggingface/datasets/issues/5831 | Today we had a big issue affecting the Hugging Face Hub, thus all the `504 Server Error: Gateway Time-out` errors.
It is fixed now and loading your datasets should work as expected. | [Bug]504 Server Error when loading dataset which was already cached | ### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it again using the same machine, I got this error:
```
Traceback (most rece... | 32 | [Bug]504 Server Error when loading dataset which was already cached
### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it agai... | [
-0.1444699466228485,
0.002489805221557617,
0.03585709258913994,
0.3607310950756073,
0.345577210187912,
0.07848291844129562,
0.4264727532863617,
0.2986140847206116,
-0.215998575091362,
-0.09981349855661392,
-0.14252692461013794,
0.34242287278175354,
0.07249689847230911,
0.011891968548297882... |
https://github.com/huggingface/datasets/issues/5831 | Hi, @albertvillanova.
If there is a locally cached version of datasets or something cache using huggingface_hub, when a network problem(either client or server) occurs, is it a better way to fallback to use the current cached version rather than raise a exception and exit? | [Bug]504 Server Error when loading dataset which was already cached | ### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it again using the same machine, I got this error:
```
Traceback (most rece... | 44 | [Bug]504 Server Error when loading dataset which was already cached
### Describe the bug
I have already cached the dataset using:
```
dataset = load_dataset("databricks/databricks-dolly-15k",
cache_dir="/mnt/data/llm/datasets/databricks-dolly-15k")
```
After that, I tried to load it agai... | [
-0.1444699466228485,
0.002489805221557617,
0.03585709258913994,
0.3607310950756073,
0.345577210187912,
0.07848291844129562,
0.4264727532863617,
0.2986140847206116,
-0.215998575091362,
-0.09981349855661392,
-0.14252692461013794,
0.34242287278175354,
0.07249689847230911,
0.011891968548297882... |
https://github.com/huggingface/datasets/issues/5829 | That is weird. I can't reproduce it again after reboot.
```python
In [2]: import platform
In [3]: platform.platform()
Out[3]: 'macOS-13.2-arm64-arm-64bit'
In [4]: from datasets import load_dataset
...:
...: jazzy = load_dataset("nomic-ai/gpt4all-j-prompt-generations", revision='v1.2-jazzy')
Found cach... | (mach-o file, but is an incompatible architecture (have 'arm64', need 'x86_64')) | ### Describe the bug
M2 MBP can't run
```python
from datasets import load_dataset
jazzy = load_dataset("nomic-ai/gpt4all-j-prompt-generations", revision='v1.2-jazzy')
```
### Steps to reproduce the bug
1. Use M2 MBP
2. Python 3.10.10 from pyenv
3. Run
```
from datasets import load_dataset
jazzy = load_... | 42 | (mach-o file, but is an incompatible architecture (have 'arm64', need 'x86_64'))
### Describe the bug
M2 MBP can't run
```python
from datasets import load_dataset
jazzy = load_dataset("nomic-ai/gpt4all-j-prompt-generations", revision='v1.2-jazzy')
```
### Steps to reproduce the bug
1. Use M2 MBP
2. Python 3... | [
-0.40310120582580566,
-0.2521466910839081,
0.05834469944238663,
0.12893787026405334,
0.3469509780406952,
0.09621185064315796,
0.12205573916435242,
0.3088306784629822,
-0.28251463174819946,
0.028164349496364594,
-0.1043701022863388,
0.6357762813568115,
-0.2538766860961914,
-0.06267487257719... |
https://github.com/huggingface/datasets/issues/5828 | Hi!
You can call `map` as follows to avoid the error:
```python
augmented_dataset_cln = dataset_cln['train'].map(augment_dataset, features=dataset_cln['train'].features)
``` | Stream data concatenation issue | ### Describe the bug
I am not able to concatenate the augmentation of the stream data. I am using the latest version of dataset.
ValueError: The features can't be aligned because the key audio of features {'audio_id': Value(dtype='string',
id=None), 'audio': {'array': Sequence(feature=Value(dtype='float32', id=... | 17 | Stream data concatenation issue
### Describe the bug
I am not able to concatenate the augmentation of the stream data. I am using the latest version of dataset.
ValueError: The features can't be aligned because the key audio of features {'audio_id': Value(dtype='string',
id=None), 'audio': {'array': Sequence(... | [
-0.2828328609466553,
-0.0790247917175293,
-0.05130556970834732,
0.2551898956298828,
0.10679186135530472,
-0.029187217354774475,
0.19599652290344238,
0.084890216588974,
-0.3168542683124542,
0.2629067003726959,
-0.14963625371456146,
0.04651915282011032,
0.05549899861216545,
0.402705550193786... |
https://github.com/huggingface/datasets/issues/5828 | Hi!
I have run into the same problem with you. Could you please let me know how you solve it? Thanks! | Stream data concatenation issue | ### Describe the bug
I am not able to concatenate the augmentation of the stream data. I am using the latest version of dataset.
ValueError: The features can't be aligned because the key audio of features {'audio_id': Value(dtype='string',
id=None), 'audio': {'array': Sequence(feature=Value(dtype='float32', id=... | 21 | Stream data concatenation issue
### Describe the bug
I am not able to concatenate the augmentation of the stream data. I am using the latest version of dataset.
ValueError: The features can't be aligned because the key audio of features {'audio_id': Value(dtype='string',
id=None), 'audio': {'array': Sequence(... | [
-0.2828328609466553,
-0.0790247917175293,
-0.05130556970834732,
0.2551898956298828,
0.10679186135530472,
-0.029187217354774475,
0.19599652290344238,
0.084890216588974,
-0.3168542683124542,
0.2629067003726959,
-0.14963625371456146,
0.04651915282011032,
0.05549899861216545,
0.402705550193786... |
https://github.com/huggingface/datasets/issues/5827 | Indeed the JSON dataset builder raises an error when it encounters an unexpected type.
There's an old PR open to add away to ignore such elements though, if it can help: https://github.com/huggingface/datasets/pull/2838 | load json dataset interrupt when dtype cast problem occured | ### Describe the bug
i have a json like this:
[
{"id": 1, "name": 1},
{"id": 2, "name": "Nan"},
{"id": 3, "name": 3},
....
]
，which have several problematic rows data like row 2, then i load it with datasets.load_dataset('json', data_files=['xx.json'], split='train'), it will report like this:
... | 32 | load json dataset interrupt when dtype cast problem occured
### Describe the bug
i have a json like this:
[
{"id": 1, "name": 1},
{"id": 2, "name": "Nan"},
{"id": 3, "name": 3},
....
]
，which have several problematic rows data like row 2, then i load it with datasets.load_dataset('json', data... | [
-0.29087209701538086,
0.005442589521408081,
-0.023509161546826363,
0.6575196981430054,
0.6633947491645813,
0.0485328733921051,
0.3084115982055664,
0.336452841758728,
-0.14784428477287292,
-0.007055878639221191,
0.06579495221376419,
0.43404433131217957,
-0.1574418544769287,
0.01036161929368... |
https://github.com/huggingface/datasets/issues/5825 | Hi!
This would only work if `bigscience/xP3` was a no-code dataset, but it isn't (it has a Python builder script).
But this should work:
```python
load_dataset("json", data_files="https://huggingface.co/datasets/bigscience/xP3/resolve/main/ur/xp3_facebook_flores_spa_Latn-urd_Arab_devtest_ab-spa_Latn-urd_Arab.... | FileNotFound even though exists | ### Describe the bug
I'm trying to download https://huggingface.co/datasets/bigscience/xP3/resolve/main/ur/xp3_facebook_flores_spa_Latn-urd_Arab_devtest_ab-spa_Latn-urd_Arab.jsonl which works fine in my webbrowser, but somehow not with datasets. Am I doing sth wrong?
```
Downloading builder script: 100%
2.82k/2.8... | 28 | FileNotFound even though exists
### Describe the bug
I'm trying to download https://huggingface.co/datasets/bigscience/xP3/resolve/main/ur/xp3_facebook_flores_spa_Latn-urd_Arab_devtest_ab-spa_Latn-urd_Arab.jsonl which works fine in my webbrowser, but somehow not with datasets. Am I doing sth wrong?
```
Downloadi... | [
-0.2510882019996643,
0.14677296578884125,
0.09450720995664597,
0.31087517738342285,
0.3156176805496216,
-0.10854393988847733,
0.19184064865112305,
0.47219040989875793,
-0.021109148859977722,
-0.014219947159290314,
0.039334096014499664,
-0.04836209863424301,
-0.006035400554537773,
0.0782377... |
https://github.com/huggingface/datasets/issues/5825 | > I see, it's not compatible w/ regex right? e.g. `load_dataset("json", data_files="https://huggingface.co/datasets/bigscience/xP3/resolve/main/ur/*")`
It should work for patterns that "reference" the local filesystem, but to make this work with the Hub, we must implement https://github.com/huggingface/datasets/issu... | FileNotFound even though exists | ### Describe the bug
I'm trying to download https://huggingface.co/datasets/bigscience/xP3/resolve/main/ur/xp3_facebook_flores_spa_Latn-urd_Arab_devtest_ab-spa_Latn-urd_Arab.jsonl which works fine in my webbrowser, but somehow not with datasets. Am I doing sth wrong?
```
Downloading builder script: 100%
2.82k/2.8... | 106 | FileNotFound even though exists
### Describe the bug
I'm trying to download https://huggingface.co/datasets/bigscience/xP3/resolve/main/ur/xp3_facebook_flores_spa_Latn-urd_Arab_devtest_ab-spa_Latn-urd_Arab.jsonl which works fine in my webbrowser, but somehow not with datasets. Am I doing sth wrong?
```
Downloadi... | [
-0.2510882019996643,
0.14677296578884125,
0.09450720995664597,
0.31087517738342285,
0.3156176805496216,
-0.10854393988847733,
0.19184064865112305,
0.47219040989875793,
-0.021109148859977722,
-0.014219947159290314,
0.039334096014499664,
-0.04836209863424301,
-0.006035400554537773,
0.0782377... |
https://github.com/huggingface/datasets/issues/5820 | Thanks for reporting! You are more than welcome to improve `BuilderConfig`'s docstring.
This class serves an identical purpose as `tensorflow_datasets`'s `BuilderConfig`, and its docstring is [here](https://github.com/tensorflow/datasets/blob/a95e38b5bb018312c3d3720619c2a8ef83ebf57f/tensorflow_datasets/core/dataset_... | Incomplete docstring for `BuilderConfig` | Hi guys !
I stumbled upon this docstring while working on a project.
Some of the attributes have missing descriptions.
https://github.com/huggingface/datasets/blob/bc5fef5b6d91f009e4101684adcb374df2c170f6/src/datasets/builder.py#L104-L117 | 34 | Incomplete docstring for `BuilderConfig`
Hi guys !
I stumbled upon this docstring while working on a project.
Some of the attributes have missing descriptions.
https://github.com/huggingface/datasets/blob/bc5fef5b6d91f009e4101684adcb374df2c170f6/src/datasets/builder.py#L104-L117
Thanks for reporting! You are m... | [
-0.3216456174850464,
0.14040017127990723,
-0.1410626471042633,
0.21151545643806458,
0.30552026629447937,
0.19221021234989166,
0.10483421385288239,
0.2315959334373474,
-0.21633338928222656,
0.1964913010597229,
0.35195815563201904,
-0.13295713067054749,
0.06480507552623749,
0.044080652296543... |
https://github.com/huggingface/datasets/issues/5819 | Hi! It should work if you put `model = torch.compile(model)` inside the `generate_data` function. If a referenced object is outside, it needs to be pickable, and that's not the case for the compiled models (or functions). | Cannot pickle error in Dataset.from_generator() | ### Describe the bug
I'm trying to use Dataset.from_generator() to generate a large dataset.
### Steps to reproduce the bug
Code to reproduce:
```
from transformers import T5Tokenizer, T5ForConditionalGeneration, GenerationConfig
import torch
from tqdm import tqdm
from datasets import load_dataset
tokenizer... | 36 | Cannot pickle error in Dataset.from_generator()
### Describe the bug
I'm trying to use Dataset.from_generator() to generate a large dataset.
### Steps to reproduce the bug
Code to reproduce:
```
from transformers import T5Tokenizer, T5ForConditionalGeneration, GenerationConfig
import torch
from tqdm import tq... | [
-0.4289095103740692,
-0.2743408679962158,
0.027513988316059113,
0.5192881226539612,
0.38269156217575073,
-0.06168058514595032,
0.21634811162948608,
0.3536655604839325,
-0.17799019813537598,
0.40683653950691223,
0.15663272142410278,
0.33345890045166016,
-0.24602733552455902,
0.1091277748346... |
https://github.com/huggingface/datasets/issues/5819 | > Hi! It should work if you put `model = torch.compile(model)` inside the `generate_data` function. If a referenced object is outside, it needs to be pickable, and that's not the case for the compiled models (or functions).
Hi! Thank you for your reply! Everything works perfectly with your suggestion!
Closing the... | Cannot pickle error in Dataset.from_generator() | ### Describe the bug
I'm trying to use Dataset.from_generator() to generate a large dataset.
### Steps to reproduce the bug
Code to reproduce:
```
from transformers import T5Tokenizer, T5ForConditionalGeneration, GenerationConfig
import torch
from tqdm import tqdm
from datasets import load_dataset
tokenizer... | 52 | Cannot pickle error in Dataset.from_generator()
### Describe the bug
I'm trying to use Dataset.from_generator() to generate a large dataset.
### Steps to reproduce the bug
Code to reproduce:
```
from transformers import T5Tokenizer, T5ForConditionalGeneration, GenerationConfig
import torch
from tqdm import tq... | [
-0.4289095103740692,
-0.2743408679962158,
0.027513988316059113,
0.5192881226539612,
0.38269156217575073,
-0.06168058514595032,
0.21634811162948608,
0.3536655604839325,
-0.17799019813537598,
0.40683653950691223,
0.15663272142410278,
0.33345890045166016,
-0.24602733552455902,
0.1091277748346... |
https://github.com/huggingface/datasets/issues/5818 | In this case, I think we can avoid the `PermissionError` by unpacking the underlying `ConcatenationTable` and saving only the newly added data blocks (in new files). | Ability to update a dataset | ### Feature request
The ability to load a dataset, add or change something, and save it back to disk.
Maybe it's possible, but I can't work out how to do it, e.g. this fails:
```py
import datasets
dataset = datasets.load_from_disk("data/test1")
dataset = dataset.add_item({"text": "A new item"})
dataset.sav... | 26 | Ability to update a dataset
### Feature request
The ability to load a dataset, add or change something, and save it back to disk.
Maybe it's possible, but I can't work out how to do it, e.g. this fails:
```py
import datasets
dataset = datasets.load_from_disk("data/test1")
dataset = dataset.add_item({"text... | [
-0.19011037051677704,
0.4618924856185913,
-0.11184918880462646,
-0.0979379266500473,
0.02292514219880104,
0.2477240264415741,
0.10128538310527802,
0.007762257009744644,
-0.31282854080200195,
0.20791491866111755,
-0.032510872930288315,
0.15257567167282104,
-0.3104686737060547,
0.35682746767... |
https://github.com/huggingface/datasets/issues/5818 | Thanks @stevhliu and @mariosasko , so saving to individual files then loading them later, concatenating again and saving again is the recommended way. Good to know.
Question that I hope doesn't sound rude: is this sort of thing (processing a dataset that doesn't fit in memory) outside of `datasets`'s core area of fo... | Ability to update a dataset | ### Feature request
The ability to load a dataset, add or change something, and save it back to disk.
Maybe it's possible, but I can't work out how to do it, e.g. this fails:
```py
import datasets
dataset = datasets.load_from_disk("data/test1")
dataset = dataset.add_item({"text": "A new item"})
dataset.sav... | 89 | Ability to update a dataset
### Feature request
The ability to load a dataset, add or change something, and save it back to disk.
Maybe it's possible, but I can't work out how to do it, e.g. this fails:
```py
import datasets
dataset = datasets.load_from_disk("data/test1")
dataset = dataset.add_item({"text... | [
-0.19202479720115662,
0.30247825384140015,
-0.14454354345798492,
-0.03796469420194626,
0.06152549013495445,
0.2540952265262604,
-0.019155899062752724,
-0.047730665653944016,
-0.06347855180501938,
0.19598427414894104,
0.08385588228702545,
0.029836494475603104,
-0.24973738193511963,
0.421500... |
https://github.com/huggingface/datasets/issues/5817 | Hi ! Unfortunately I couldn't reproduce on my side locally and with datasets 2.11 and python 3.10.11 on colab.
What version of `multiprocess` are you using ? | Setting `num_proc` errors when `.map` returns additional items. | ### Describe the bug
I'm using a map function that returns more rows than are passed in.
If I try to use `num_proc` I get:
```
File "/home/davidg/.virtualenvs/learning/lib/python3.10/site-packages/datasets/arrow_dataset.py", line 563, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kw... | 27 | Setting `num_proc` errors when `.map` returns additional items.
### Describe the bug
I'm using a map function that returns more rows than are passed in.
If I try to use `num_proc` I get:
```
File "/home/davidg/.virtualenvs/learning/lib/python3.10/site-packages/datasets/arrow_dataset.py", line 563, in wrapper
... | [
0.24224074184894562,
-0.41235876083374023,
-0.07637587189674377,
0.37606674432754517,
0.0636359453201294,
-0.3487008810043335,
-0.0032279640436172485,
0.3338972330093384,
0.023244023323059082,
0.30512771010398865,
0.4175057113170624,
0.4686207175254822,
-0.2382592111825943,
0.1526001095771... |
https://github.com/huggingface/datasets/issues/5817 | I've got `multiprocess` version `0.70.14`.
I've done some more testing and the error only occurs in PyCharm's Python Console. It seems to be [this PyCharm bug](https://youtrack.jetbrains.com/issue/PY-51922/Multiprocessing-bug.-Can-only-run-in-debugger.), I'll close this. | Setting `num_proc` errors when `.map` returns additional items. | ### Describe the bug
I'm using a map function that returns more rows than are passed in.
If I try to use `num_proc` I get:
```
File "/home/davidg/.virtualenvs/learning/lib/python3.10/site-packages/datasets/arrow_dataset.py", line 563, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kw... | 29 | Setting `num_proc` errors when `.map` returns additional items.
### Describe the bug
I'm using a map function that returns more rows than are passed in.
If I try to use `num_proc` I get:
```
File "/home/davidg/.virtualenvs/learning/lib/python3.10/site-packages/datasets/arrow_dataset.py", line 563, in wrapper
... | [
0.24224074184894562,
-0.41235876083374023,
-0.07637587189674377,
0.37606674432754517,
0.0636359453201294,
-0.3487008810043335,
-0.0032279640436172485,
0.3338972330093384,
0.023244023323059082,
0.30512771010398865,
0.4175057113170624,
0.4686207175254822,
-0.2382592111825943,
0.1526001095771... |
https://github.com/huggingface/datasets/issues/5817 | For other users facing this, my workaround is to conditionally set `num_proc` so I can work interactively in the PyCharm Python Console while developing, then when I'm ready to run on the whole dataset, run it as a script and use multiprocessing.
```py
mapped_ds = ds.map(
my_map_function,
batched=True,
... | Setting `num_proc` errors when `.map` returns additional items. | ### Describe the bug
I'm using a map function that returns more rows than are passed in.
If I try to use `num_proc` I get:
```
File "/home/davidg/.virtualenvs/learning/lib/python3.10/site-packages/datasets/arrow_dataset.py", line 563, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kw... | 58 | Setting `num_proc` errors when `.map` returns additional items.
### Describe the bug
I'm using a map function that returns more rows than are passed in.
If I try to use `num_proc` I get:
```
File "/home/davidg/.virtualenvs/learning/lib/python3.10/site-packages/datasets/arrow_dataset.py", line 563, in wrapper
... | [
0.24224074184894562,
-0.41235876083374023,
-0.07637587189674377,
0.37606674432754517,
0.0636359453201294,
-0.3487008810043335,
-0.0032279640436172485,
0.3338972330093384,
0.023244023323059082,
0.30512771010398865,
0.4175057113170624,
0.4686207175254822,
-0.2382592111825943,
0.1526001095771... |
https://github.com/huggingface/datasets/issues/5815 | Hi @hrbigelow , I'm no expert for such a question so I'll ping @lhoestq from the `datasets` library (also this issue could be moved there if someone with permission can do it :) ) | Easy way to create a Kaggle dataset from a Huggingface dataset? | I'm not sure whether this is more appropriately addressed with HuggingFace or Kaggle. I would like to somehow directly create a Kaggle dataset from a HuggingFace Dataset.
While Kaggle does provide the option to create a dataset from a URI, that URI must point to a single file. For example:
![image](https://user... | 34 | Easy way to create a Kaggle dataset from a Huggingface dataset?
I'm not sure whether this is more appropriately addressed with HuggingFace or Kaggle. I would like to somehow directly create a Kaggle dataset from a HuggingFace Dataset.
While Kaggle does provide the option to create a dataset from a URI, that URI m... | [
-0.06938397884368896,
-0.46260684728622437,
0.07999630272388458,
0.14770963788032532,
0.29280203580856323,
0.1902616173028946,
-0.006398454308509827,
0.14528387784957886,
0.09063258767127991,
0.10242429375648499,
-0.20503051578998566,
0.23473498225212097,
-0.21115469932556152,
0.7492237091... |
https://github.com/huggingface/datasets/issues/5815 | Hi ! Many datasets are made of several files, and how they are parsed often requires a python script. Because of that, datasets like wmt14 are not available as a single file on HF. Though you can create this file using `datasets`:
```python
from datasets import load_dataset
ds = load_dataset("wmt14", "de-en", sp... | Easy way to create a Kaggle dataset from a Huggingface dataset? | I'm not sure whether this is more appropriately addressed with HuggingFace or Kaggle. I would like to somehow directly create a Kaggle dataset from a HuggingFace Dataset.
While Kaggle does provide the option to create a dataset from a URI, that URI must point to a single file. For example:
![image](https://user... | 102 | Easy way to create a Kaggle dataset from a Huggingface dataset?
I'm not sure whether this is more appropriately addressed with HuggingFace or Kaggle. I would like to somehow directly create a Kaggle dataset from a HuggingFace Dataset.
While Kaggle does provide the option to create a dataset from a URI, that URI m... | [
-0.13704515993595123,
-0.5377284288406372,
0.11752719432115555,
0.11216665804386139,
0.24724948406219482,
0.2054954171180725,
-0.025303073227405548,
0.17020028829574585,
0.1390443593263626,
0.11979781091213226,
-0.26239433884620667,
0.1523076444864273,
-0.13135060667991638,
0.7215529680252... |
https://github.com/huggingface/datasets/issues/5815 | I've requested support for creating a Kaggle dataset from an imported HF dataset repo on their "forum" here: https://www.kaggle.com/discussions/product-feedback/427142 (upvotes appreciated 🙂) | Easy way to create a Kaggle dataset from a Huggingface dataset? | I'm not sure whether this is more appropriately addressed with HuggingFace or Kaggle. I would like to somehow directly create a Kaggle dataset from a HuggingFace Dataset.
While Kaggle does provide the option to create a dataset from a URI, that URI must point to a single file. For example:
Also, we should improve the error to make it more obvious what the pr... | load_dataset: TypeError: 'NoneType' object is not callable, on local dataset filename changes | ### Describe the bug
I've adapted Databrick's [train_dolly.py](/databrickslabs/dolly/blob/master/train_dolly.py) to train using a local dataset, which has been working. Upon changing the filenames of the `.json` & `.py` files in my local dataset directory, `dataset = load_dataset(path_or_dataset)["train"]` throws th... | 45 | load_dataset: TypeError: 'NoneType' object is not callable, on local dataset filename changes
### Describe the bug
I've adapted Databrick's [train_dolly.py](/databrickslabs/dolly/blob/master/train_dolly.py) to train using a local dataset, which has been working. Upon changing the filenames of the `.json` & `.py` f... | [
-0.03892234340310097,
0.3020404577255249,
0.15307991206645966,
0.35685789585113525,
0.4620305299758911,
0.17396527528762817,
0.6785809397697449,
0.33185064792633057,
0.2880825102329254,
0.2160206288099289,
0.2014547884464264,
0.34833091497421265,
-0.26490601897239685,
-0.11683293431997299,... |
https://github.com/huggingface/datasets/issues/5809 | Hi ! I don't remember exactly how it was done, but maybe you have to embed `f"{title}<sep>{text}"` ?
Using a HF tokenizer it corresponds to doing
```python
tokenized = tokenizer(titles, texts)
``` | wiki_dpr details for Open Domain Question Answering tasks | Hey guys!
Thanks for creating the wiki_dpr dataset!
I am currently trying to combine wiki_dpr and my own datasets. but I don't know how to make the embedding value the same way as wiki_dpr.
As an experiment, I embeds the text of id="7" of wiki_dpr, but this result was very different from wiki_dpr. | 32 | wiki_dpr details for Open Domain Question Answering tasks
Hey guys!
Thanks for creating the wiki_dpr dataset!
I am currently trying to combine wiki_dpr and my own datasets. but I don't know how to make the embedding value the same way as wiki_dpr.
As an experiment, I embeds the text of id="7" of wiki_dpr, bu... | [
0.058335259556770325,
-0.6291362643241882,
0.005163419991731644,
0.25820544362068176,
-0.027563676238059998,
0.018641412258148193,
0.29902830719947815,
0.04544338211417198,
0.17212235927581787,
0.5337421894073486,
-0.17683109641075134,
0.17993849515914917,
0.30286556482315063,
0.0636853650... |
https://github.com/huggingface/datasets/issues/5806 | Implementing this makes sense (e.g., `tensorflow_datasets`' imagefolder returns image filenames). Also, in Datasets 3.0, we plan only to store the bytes of an image/audio, not its path, so this feature would be useful when the path info is still needed. | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 40 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.4825645089149475,
-0.09418713301420212,
-0.04889893904328346,
0.03705478087067604,
0.03574010729789734,
-0.14645807445049286,
0.4693204164505005,
0.45564034581184387,
0.10660460591316223,
0.17320871353149414,
0.07409103214740753,
0.5581418871879578,
-0.41195744276046753,
0.3097176253795... |
https://github.com/huggingface/datasets/issues/5806 | Hey @mariosasko, Can I work on this issue, this one seems interesting to implement. I have contributed to jupyterlab recently, and would love to contribute here as well. | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 28 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.2670653462409973,
-0.030552253127098083,
-0.017161186784505844,
-0.03512344881892204,
0.018165092915296555,
-0.20466110110282898,
0.4512977600097656,
0.540688157081604,
0.20236021280288696,
0.15843483805656433,
0.05343162268400192,
0.5594682097434998,
-0.34587737917900085,
0.49699687957... |
https://github.com/huggingface/datasets/issues/5806 | @tsabbir96 if you are planning to start working on this, you can take on this issue by writing a comment with only the keyword: #self-assign | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 25 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.4191776216030121,
-0.1814727783203125,
-0.030569300055503845,
-0.01647360995411873,
0.07291547954082489,
-0.13672377169132233,
0.4060405492782593,
0.4840847849845886,
0.2491959035396576,
0.09856682270765305,
0.07769029587507248,
0.5259687304496765,
-0.29867953062057495,
0.47859397530555... |
https://github.com/huggingface/datasets/issues/5806 | @albertvillanova thank you for letting me contribute here.
@albertvillanova @mariosasko As I am totally new to this repo, could you tell me something more about this issue or perhaps give me some idea on how I can proceed with it? Thanks! | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 41 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.3265039920806885,
-0.2647354006767273,
-0.01758318766951561,
0.07004538178443909,
0.0992976576089859,
-0.13961559534072876,
0.3466265797615051,
0.4475677013397217,
0.22081124782562256,
0.09441199153661728,
0.06968678534030914,
0.4536360204219818,
-0.23294107615947723,
0.467743843793869,... |
https://github.com/huggingface/datasets/issues/5806 | Hello there, is this issue resolved? @tsabbir96 are you still working on it? Otherwise I would love to give it a try | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 22 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.39790356159210205,
-0.22820815443992615,
-0.02644972875714302,
0.023000480607151985,
0.09245386719703674,
-0.1665101945400238,
0.3990684747695923,
0.49507105350494385,
0.22137081623077393,
0.0845126062631607,
0.08824421465396881,
0.504729151725769,
-0.2669532895088196,
0.484145551919937... |
https://github.com/huggingface/datasets/issues/5806 | Hey @mariosasko, I've taken the time to take a look at how we load the datasets usually. My main question now is about the final solution.
So the idea is that whenever we load the datasets we also add a new column in the _generate_tables() method from the builders called filename (or file_name) that should be relate... | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 74 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.36966389417648315,
-0.047463566064834595,
-0.025749318301677704,
0.20428672432899475,
0.03717111051082611,
-0.0648699551820755,
0.35760340094566345,
0.4892924427986145,
0.23243914544582367,
0.0483609102666378,
0.0418221578001976,
0.4462888538837433,
-0.24884460866451263,
0.4740105271339... |
https://github.com/huggingface/datasets/issues/5806 | > Is this issue still open? If yes, I'd like to work upon on it. Thanks
Definitely still open. I gave it a try, but then didn't get any feedback on my last question so I stopped . Feel free to work on it. | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 44 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.34719961881637573,
-0.22424393892288208,
-0.027185918763279915,
0.005224757827818394,
0.11257457733154297,
-0.18082734942436218,
0.39576083421707153,
0.47865381836891174,
0.22163629531860352,
0.09539312869310379,
0.11878401041030884,
0.5177738070487976,
-0.2579129934310913,
0.4836080968... |
https://github.com/huggingface/datasets/issues/5806 | It's still open, so feel free to work on it. This can be implemented by adding a param to the packaged builders' configs that inserts a column with file names (in `_generate_tables`) when `True`. Naming this column `file_name` sounds good to me. | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 42 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.5077106952667236,
-0.14263205230236053,
-0.03315097838640213,
0.036459363996982574,
0.05918165296316147,
-0.1509508490562439,
0.38432031869888306,
0.5087962746620178,
0.21042674779891968,
0.11009090393781662,
0.16244655847549438,
0.5358562469482422,
-0.33372265100479126,
0.4870533943176... |
https://github.com/huggingface/datasets/issues/5806 | Hi is the issues still open, is see no activity since September but it shows that it is still assigned to tsabbir96. If
tsabbir96 is not planning to continue, can i get it assigned to me. | Return the name of the currently loaded file in the load_dataset function. | ### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Motivation
When training large language models, machine problems may interrupt... | 36 | Return the name of the currently loaded file in the load_dataset function.
### Feature request
Add an optional parameter return_file_name in the load_dataset function. When it is set to True, the function will include the name of the file corresponding to the current line as a feature in the returned output.
### Mo... | [
-0.3106291592121124,
-0.2386893630027771,
-0.05924660339951515,
0.1655980348587036,
0.0021553095430135727,
-0.18083323538303375,
0.4046478271484375,
0.4016396105289459,
0.20308026671409607,
0.03709171339869499,
0.07435110956430435,
0.4502989649772644,
-0.344589501619339,
0.4916836023330688... |
https://github.com/huggingface/datasets/issues/5805 | I can work on this. The link to the tutorial seems to be broken though @polinaeterna. | Improve `Create a dataset` tutorial | Our [tutorial on how to create a dataset](https://huggingface.co/docs/datasets/create_dataset) is a bit misleading.
1. In **Folder-based builders** section it says that we have two folder-based builders as standard builders, but we also have similar builders (that can be created from directory with data of required f... | 16 | Improve `Create a dataset` tutorial
Our [tutorial on how to create a dataset](https://huggingface.co/docs/datasets/create_dataset) is a bit misleading.
1. In **Folder-based builders** section it says that we have two folder-based builders as standard builders, but we also have similar builders (that can be created... | [
-0.19121858477592468,
-0.2695015072822571,
-0.04800918698310852,
0.5280870199203491,
0.23932234942913055,
-0.03763539344072342,
0.004639771301299334,
0.024110592901706696,
-0.0192304328083992,
0.25799429416656494,
-0.025773076340556145,
0.08891582489013672,
-0.12057146430015564,
0.32013255... |
https://github.com/huggingface/datasets/issues/5805 | @isunitha98selvan would be great, thank you! which link are you talking about? I think it should work: https://huggingface.co/docs/datasets/create_dataset | Improve `Create a dataset` tutorial | Our [tutorial on how to create a dataset](https://huggingface.co/docs/datasets/create_dataset) is a bit misleading.
1. In **Folder-based builders** section it says that we have two folder-based builders as standard builders, but we also have similar builders (that can be created from directory with data of required f... | 18 | Improve `Create a dataset` tutorial
Our [tutorial on how to create a dataset](https://huggingface.co/docs/datasets/create_dataset) is a bit misleading.
1. In **Folder-based builders** section it says that we have two folder-based builders as standard builders, but we also have similar builders (that can be created... | [
-0.18374519050121307,
-0.2780703604221344,
-0.04612888768315315,
0.5162710547447205,
0.23995791375637054,
-0.013876698911190033,
0.030477095395326614,
0.017136752605438232,
-0.026692785322666168,
0.2696177363395691,
0.022135674953460693,
0.08110903203487396,
-0.13184525072574615,
0.3071809... |
https://github.com/huggingface/datasets/issues/5798 | Hi ! We're using process pools for parallelism right now. I was wondering if there's a package that implements the same API as a process pool but runs with Spark under the hood ? That or something similar would be cool because users could use whatever distributed framework they want this way.
Feel free to ping us wh... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 81 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.5311525464057922,
-0.4519221782684326,
-0.28954362869262695,
0.08191453665494919,
-0.04725201427936554,
-0.007369197905063629,
0.2613649368286133,
0.19786617159843445,
0.2873057425022125,
0.1246139332652092,
0.019359782338142395,
0.4726947844028473,
-0.083057701587677,
0.526435196399688... |
https://github.com/huggingface/datasets/issues/5798 | Hi, thanks for taking a look and providing your input! I don't know of such packages, and even it exists, I don't think with the process pool API it's possible to run Spark as backend properly; otherwise I understand a unified API would be preferable.
The process pool API requires splitting the workload to a fixed n... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 132 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.5754362344741821,
-0.26495084166526794,
-0.27761897444725037,
0.03848211094737053,
-0.06436130404472351,
-0.011157408356666565,
0.3694995641708374,
0.21103748679161072,
0.25697705149650574,
0.05093444511294365,
-0.021644312888383865,
0.5398142337799072,
-0.04608116298913956,
0.553321719... |
https://github.com/huggingface/datasets/issues/5798 | I see ! I think we only need an equivalent of `pool.map`. We use it to run download and conversion of data files on disk. That would require less changes in the internal code - and therefore less tests to write ;)
We also use `pool.apply_async` in some places with a `Queue` to get progress updates of the running job... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 84 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.5281373858451843,
-0.18202292919158936,
-0.2799943685531616,
0.029995812103152275,
0.0023557282984256744,
-0.012214362621307373,
0.3677160441875458,
0.36550411581993103,
0.26282089948654175,
0.1473686397075653,
-0.027858640998601913,
0.5398534536361694,
-0.02363302744925022,
0.355804681... |
https://github.com/huggingface/datasets/issues/5798 | For Spark, `rdd.map` (where `rdd` can be created by `sparkContext.parallelize`) is the most similar as `pool.map`, but it requires creating a Spark RDD first that is used for distributing the `iterable` and the actual parallelization is managed by the Spark framework; `pool.map` takes the splits of `iterable` that are ... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 127 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.4450608193874359,
-0.29352670907974243,
-0.2786247730255127,
-0.005575243383646011,
-0.029473690316081047,
0.013661235570907593,
0.3585393726825714,
0.24527771770954132,
0.2660626173019409,
0.11462710797786713,
-0.17100560665130615,
0.5936993360519409,
-0.10378982871770859,
0.3936449587... |
https://github.com/huggingface/datasets/issues/5798 | Indeed I think the current use of multiprocessing.Pool in `map_nested` can be rewritten to work like `sparkContext.parallelize` - without splitting the iterable.
Maybe from the user's perspective it's ok to let multiprocessing.Pool or spark distribute the load on their own, as long as it takes a list and runs jobs i... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 56 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.5608075261116028,
-0.2904577851295471,
-0.28479528427124023,
0.12091349065303802,
-0.09748193621635437,
0.04721705615520477,
0.4310897886753082,
0.21069492399692535,
0.3261694312095642,
0.09282231330871582,
-0.1244276836514473,
0.477003812789917,
-0.03488113731145859,
0.4295397102832794... |
https://github.com/huggingface/datasets/issues/5798 | From your feedback, seems to me there are two paths to consider now for supporting spark's `map` function in `map_nested` now:
1. Keep the current `pool.map` implementation, and add an if statement for the spark's `map` code (which is what I did in my current PR) -- the code change is just a few lines in the `map_nest... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 185 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.473716676235199,
-0.3848658800125122,
-0.25377580523490906,
0.05113663896918297,
-0.21205277740955353,
-0.031120195984840393,
0.3600761890411377,
0.3059087097644806,
0.3719865083694458,
-0.014005929231643677,
-0.03372737765312195,
0.5151513814926147,
-0.07134740054607391,
0.200395733118... |
https://github.com/huggingface/datasets/issues/5798 | Yup correct ! I think it's a nice path because it would be possible for users to define whatever parallel processing backend they want. I think we still need to discuss how that would look like in the `datasets` API : how to specify it has to use the "spark" parallel backend ? And how to specify the spark session param... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 94 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.36371368169784546,
-0.18536823987960815,
-0.2672674059867859,
0.06818731129169464,
-0.1409541815519333,
-0.04652605950832367,
0.3026992678642273,
0.2803654372692108,
0.24448606371879578,
0.01529906690120697,
0.05880527198314667,
0.49277788400650024,
-0.07215771824121475,
0.5174586772918... |
https://github.com/huggingface/datasets/issues/5798 | Sure, please let me know if you have more updates regarding the API and implementation from the team.
For parameters we don't need to worry about setting them for Spark, because Spark will figure out the environment / number of worker nodes by itself, so it's preferable to just provide some parameter such as `use_sp... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 61 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.4812101125717163,
-0.3032202124595642,
-0.290436714887619,
0.07637719064950943,
-0.020943492650985718,
-0.10099832713603973,
0.34414246678352356,
0.2184220850467682,
0.2572854161262512,
0.07534430176019669,
-0.00004504993557929993,
0.5947644114494324,
-0.14878487586975098,
0.42328602075... |
https://github.com/huggingface/datasets/issues/5798 | Hi! I wanted to check in to see if there is any update from the team.
A potential change of API I can think of is change the argument to `distributed_backend=...`, which accepts `str`, such as `load_dataset(..., distributed_backend="spark")`.
Implementation wise, we can add a class / function to abstract away the... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 66 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.4750991463661194,
-0.3348017930984497,
-0.2552705705165863,
0.09217201173305511,
-0.1191941648721695,
-0.07862095534801483,
0.3384152352809906,
0.20579347014427185,
0.3066742718219757,
0.09049621224403381,
0.0406356006860733,
0.5110883712768555,
-0.0026250481605529785,
0.479317814111709... |
https://github.com/huggingface/datasets/issues/5798 | I found this quite interesting: https://github.com/joblib/joblib-spark with this syntax:
```python
with parallel_backend('spark', n_jobs=3):
...
```
cc @lu-wang-dl who might know better | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 21 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.46477270126342773,
-0.2248917818069458,
-0.2864758372306824,
0.15008693933486938,
-0.0498322919011116,
-0.02512441575527191,
0.43911606073379517,
0.19149240851402283,
0.27764448523521423,
0.10016019642353058,
0.010865975171327591,
0.5445321798324585,
-0.07295788824558258,
0.486570060253... |
https://github.com/huggingface/datasets/issues/5798 | Joblib spark is providing Spark backend for joblib. We can implement a general parallel backend like
```
with parallel_backend("<parallel-backedn>", n_jobs=..):
```
It can support multiprocessing , spark, ray, and etc. https://joblib.readthedocs.io/en/latest/parallel.html#joblib.parallel_backend | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 31 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.5077944993972778,
-0.21592821180820465,
-0.296323299407959,
0.10189136862754822,
-0.049831487238407135,
-0.051647864282131195,
0.43653202056884766,
0.22513099014759064,
0.26978829503059387,
0.1350201815366745,
-0.1023634523153305,
0.5332247614860535,
-0.03646765649318695,
0.537842035293... |
https://github.com/huggingface/datasets/issues/5798 | Thank you @lhoestq for finding this repo. I validated that it can distribute downloading jobs with Spark to arbitrary cluster worker nodes evenly with `n_jobs=-1`.
For the API, I think it makes sense to define it as
```python
load_dataset(..., parallel_backend=<str>)
```
where `parallel_backend` can be `spark`, ... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 82 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.4051167368888855,
-0.1918499618768692,
-0.26870977878570557,
0.046584147959947586,
-0.263689249753952,
-0.15612906217575073,
0.4492082893848419,
0.159150168299675,
0.36480194330215454,
-0.006678774952888489,
0.06409858912229538,
0.5088892579078674,
-0.07883021235466003,
0.50977134704589... |
https://github.com/huggingface/datasets/issues/5798 | Hello @lhoestq, I wanted to follow up on my previous comment with some prototyping code that demonstrates how `map_nested` would be like if we unify `multiprocessing` and `spark` with `joblib`. The snippet hasn't hashed out the details such as dealing with `tqdm` yet.
In terms of API, the way of using multiprocessin... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 291 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.5167416930198669,
-0.3007856607437134,
-0.2685208022594452,
0.08404047787189484,
-0.08992109447717667,
-0.07849778234958649,
0.448391854763031,
0.2081310898065567,
0.29894667863845825,
0.08510483056306839,
-0.07281912863254547,
0.5960126519203186,
-0.07160244882106781,
0.490204751491546... |
https://github.com/huggingface/datasets/issues/5798 | Nice ! It should be quite easy to make the change then :)
I think adding spark support can actually be less than 20 lines of code and would roughly require one line of code to change in map_nested:
Maybe we can define a new `datasets.parallel` submodule that has the `parallel_backend()` context manager and a `par... | Support parallelized downloading and processing in load_dataset with Spark | ### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```python
load_dataset(..., use_spark=True)
```
### Motivation
Further speed up ... | 108 | Support parallelized downloading and processing in load_dataset with Spark
### Feature request
When calling `load_dataset` for datasets that have multiple files, support using Spark to distribute the downloading and processing job to worker nodes when `cache_dir` is a cloud file system shared among nodes.
```pyth... | [
-0.4788985252380371,
-0.27889353036880493,
-0.29145094752311707,
0.0654510036110878,
-0.12983141839504242,
0.10031931102275848,
0.34198427200317383,
0.3615080416202545,
0.36372554302215576,
0.05006475746631622,
-0.01716097816824913,
0.5124855041503906,
-0.11140985786914825,
0.4177769422531... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.