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/6008 | By default, we write data to disk (so it can be memory-mapped) every 1000 rows/samples. You can control this with the `writer_batch_size` parameter. Also, when working with fixed-size arrays, the `ArrayXD` feature types yield better performance (e.g., in your case, `features=datasets.Features({"i": datasets.Array3D(sha... | Dataset.from_generator consistently freezes at ~1000 rows | ### Describe the bug
Whenever I try to create a dataset which contains images using `Dataset.from_generator`, it freezes around 996 rows. I suppose it has something to do with memory consumption, but there's more memory available. I
Somehow it worked a few times but mostly this makes the datasets library much more ... | 64 | Dataset.from_generator consistently freezes at ~1000 rows
### Describe the bug
Whenever I try to create a dataset which contains images using `Dataset.from_generator`, it freezes around 996 rows. I suppose it has something to do with memory consumption, but there's more memory available. I
Somehow it worked a few... | [
-0.23392441868782043,
0.08639460802078247,
-0.021676111966371536,
0.5308259129524231,
0.25551488995552063,
0.06502610445022583,
0.18270829319953918,
0.12610292434692383,
0.018689557909965515,
0.13800251483917236,
0.3685626983642578,
0.0700070858001709,
-0.11341309547424316,
-0.043522495776... |
https://github.com/huggingface/datasets/issues/6008 | > By default, we write data to disk (so it can be memory-mapped) every 1000 rows/samples. You can control this with the `writer_batch_size` parameter. Also, when working with fixed-size arrays, the `ArrayXD` feature types yield better performance (e.g., in your case, `features=datasets.Features({"i": datasets.Array3D(s... | Dataset.from_generator consistently freezes at ~1000 rows | ### Describe the bug
Whenever I try to create a dataset which contains images using `Dataset.from_generator`, it freezes around 996 rows. I suppose it has something to do with memory consumption, but there's more memory available. I
Somehow it worked a few times but mostly this makes the datasets library much more ... | 101 | Dataset.from_generator consistently freezes at ~1000 rows
### Describe the bug
Whenever I try to create a dataset which contains images using `Dataset.from_generator`, it freezes around 996 rows. I suppose it has something to do with memory consumption, but there's more memory available. I
Somehow it worked a few... | [
-0.2846596837043762,
0.23980244994163513,
-0.03189616650342941,
0.5401877760887146,
0.2745557427406311,
0.07768195867538452,
0.22943264245986938,
0.1456349939107895,
-0.06428185105323792,
0.1479133665561676,
0.47321951389312744,
0.16201141476631165,
-0.09623195230960846,
-0.168942451477050... |
https://github.com/huggingface/datasets/issues/6007 | This error means that one of the int32 (`Value("int32")`) columns in the dataset has a value that is out of the valid (int32) range.
I'll open a PR to print the name of a problematic column to make debugging such errors easier. | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset | ### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encountered the error: "OverflowError: Python int too large to convert to C long"
The error look something like... | 42 | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset
### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encounter... | [
-0.4273005723953247,
-0.3950992226600647,
-0.026656195521354675,
0.3553202748298645,
0.35095882415771484,
-0.00437396764755249,
0.10522252321243286,
0.33112990856170654,
0.3119520843029022,
0.19228410720825195,
-0.027331477031111717,
0.1621960550546646,
-0.11134392023086548,
0.089587256312... |
https://github.com/huggingface/datasets/issues/6007 | I am afraid int32 is not the reason for this error.
I have submitted a commit to use int64 for all ints in the dataset:
https://huggingface.co/datasets/liwu/MNBVC/commit/857ac00d9eab96a6708ad6a82bd9001686042a9e
and I have updated my env to the latest datasets release:
Copy-and-paste the text below in your GitHu... | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset | ### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encountered the error: "OverflowError: Python int too large to convert to C long"
The error look something like... | 763 | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset
### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encounter... | [
-0.4273005723953247,
-0.3950992226600647,
-0.026656195521354675,
0.3553202748298645,
0.35095882415771484,
-0.00437396764755249,
0.10522252321243286,
0.33112990856170654,
0.3119520843029022,
0.19228410720825195,
-0.027331477031111717,
0.1621960550546646,
-0.11134392023086548,
0.089587256312... |
https://github.com/huggingface/datasets/issues/6007 | `simhash` is the problematic column - it has values such as `18329103420363166823` that are out of the int64 range. You can fix this by setting the feature type to `Value("string")` (it's advised to use this type for hash values in general)
> Besides, it works fine when I am using streamed dataset.
Streaming yiel... | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset | ### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encountered the error: "OverflowError: Python int too large to convert to C long"
The error look something like... | 75 | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset
### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encounter... | [
-0.4273005723953247,
-0.3950992226600647,
-0.026656195521354675,
0.3553202748298645,
0.35095882415771484,
-0.00437396764755249,
0.10522252321243286,
0.33112990856170654,
0.3119520843029022,
0.19228410720825195,
-0.027331477031111717,
0.1621960550546646,
-0.11134392023086548,
0.089587256312... |
https://github.com/huggingface/datasets/issues/6007 | i am using uint64 for simhash
uint64 ranges up to about 3.69E19.
18329103420363166823 is less than this value.
moreover, our simhash algorithm use 64 bits. it should fit in uint64.
| Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset | ### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encountered the error: "OverflowError: Python int too large to convert to C long"
The error look something like... | 30 | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset
### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encounter... | [
-0.4273005723953247,
-0.3950992226600647,
-0.026656195521354675,
0.3553202748298645,
0.35095882415771484,
-0.00437396764755249,
0.10522252321243286,
0.33112990856170654,
0.3119520843029022,
0.19228410720825195,
-0.027331477031111717,
0.1621960550546646,
-0.11134392023086548,
0.089587256312... |
https://github.com/huggingface/datasets/issues/6007 | You are right. I overlooked the feature type.
This is a reproducer:
```python
import pyarrow as pa
from datasets.arrow_writer import TypedSequence
pa.array(TypedSequence([18329103420363166823], type=Value("uint64")))
```
`pa.array([18329103420363166823])` also fails with the same error, so it seems PyArrow... | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset | ### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encountered the error: "OverflowError: Python int too large to convert to C long"
The error look something like... | 114 | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset
### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encounter... | [
-0.4273005723953247,
-0.3950992226600647,
-0.026656195521354675,
0.3553202748298645,
0.35095882415771484,
-0.00437396764755249,
0.10522252321243286,
0.33112990856170654,
0.3119520843029022,
0.19228410720825195,
-0.027331477031111717,
0.1621960550546646,
-0.11134392023086548,
0.089587256312... |
https://github.com/huggingface/datasets/issues/6007 | Thank you for checking this issue in detail.
However, it seems that using `np.uint64(simhash)` does not work. The same issue still exists.
https://huggingface.co/datasets/liwu/MNBVC/commit/1e44f1e400b7e61052647d44c99cdae3bae9c830
Anyway, we decide to use string type for these simhash values. Hope pyarrow can f... | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset | ### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encountered the error: "OverflowError: Python int too large to convert to C long"
The error look something like... | 41 | Get an error "OverflowError: Python int too large to convert to C long" when loading a large dataset
### Describe the bug
When load a large dataset with the following code
```python
from datasets import load_dataset
dataset = load_dataset("liwu/MNBVC", 'news_peoples_daily', split='train')
```
We encounter... | [
-0.4273005723953247,
-0.3950992226600647,
-0.026656195521354675,
0.3553202748298645,
0.35095882415771484,
-0.00437396764755249,
0.10522252321243286,
0.33112990856170654,
0.3119520843029022,
0.19228410720825195,
-0.027331477031111717,
0.1621960550546646,
-0.11134392023086548,
0.089587256312... |
https://github.com/huggingface/datasets/issues/5997 | I just noticed the [docs](https://github.com/huggingface/datasets/blob/main/src/datasets/arrow_dataset.py#L2881C11-L2881C200) say:
>If batched is `True` and `batch_size` is `n > 1`, then the function takes a batch of `n` examples as input and can return a batch with `n` examples, or with an arbitrary number of examp... | extend the map function so it can wrap around long text that does not fit in the context window | ### Feature request
I understand `dataset` provides a [`map`](https://github.com/huggingface/datasets/blob/main/src/datasets/arrow_dataset.py#L2849) function. This function in turn takes in a callable that is used to tokenize the text on which a model is trained. Frequently this text will not fit within a models's con... | 49 | extend the map function so it can wrap around long text that does not fit in the context window
### Feature request
I understand `dataset` provides a [`map`](https://github.com/huggingface/datasets/blob/main/src/datasets/arrow_dataset.py#L2849) function. This function in turn takes in a callable that is used to toke... | [
-0.3404634892940521,
0.03542882204055786,
-0.08898577094078064,
0.043310411274433136,
0.17575713992118835,
-0.04931211471557617,
0.30036675930023193,
0.36894655227661133,
-0.09006550908088684,
-0.17397615313529968,
0.09924708306789398,
0.5054342746734619,
-0.03185398131608963,
0.3214940428... |
https://github.com/huggingface/datasets/issues/5997 | All the values in a batch must be of the same length. So one solution is dropping all the input columns:
```python
data = data.map(lambda samples: tokenizer(samples["text"], max_length=tokenizer.model_max_length, truncation=True, stride=4, return_overflowing_tokens=True), batched=True, remove_columns=data.column_name... | extend the map function so it can wrap around long text that does not fit in the context window | ### Feature request
I understand `dataset` provides a [`map`](https://github.com/huggingface/datasets/blob/main/src/datasets/arrow_dataset.py#L2849) function. This function in turn takes in a callable that is used to tokenize the text on which a model is trained. Frequently this text will not fit within a models's con... | 46 | extend the map function so it can wrap around long text that does not fit in the context window
### Feature request
I understand `dataset` provides a [`map`](https://github.com/huggingface/datasets/blob/main/src/datasets/arrow_dataset.py#L2849) function. This function in turn takes in a callable that is used to toke... | [
-0.3404634892940521,
0.03542882204055786,
-0.08898577094078064,
0.043310411274433136,
0.17575713992118835,
-0.04931211471557617,
0.30036675930023193,
0.36894655227661133,
-0.09006550908088684,
-0.17397615313529968,
0.09924708306789398,
0.5054342746734619,
-0.03185398131608963,
0.3214940428... |
https://github.com/huggingface/datasets/issues/5993 | We'll do a new release of `datasets` soon to make the fix available :)
In the meantime you can use `datasets` from source (main) | ValueError: Table schema does not match schema used to create file | ### Describe the bug
Saving a dataset as parquet fails with a `ValueError: Table schema does not match schema used to create file` if the dataset was obtained out of a `.select_columns()` call with columns selected out of order.
### Steps to reproduce the bug
```python
import datasets
dataset = datasets.Dataset... | 24 | ValueError: Table schema does not match schema used to create file
### Describe the bug
Saving a dataset as parquet fails with a `ValueError: Table schema does not match schema used to create file` if the dataset was obtained out of a `.select_columns()` call with columns selected out of order.
### Steps to reprod... | [
-0.2826020419597626,
-0.03604024648666382,
0.04691947251558304,
0.303589791059494,
0.12757602334022522,
0.0786287784576416,
-0.09735140204429626,
0.290692538022995,
0.07794879376888275,
0.2212046980857849,
0.32674533128738403,
0.6393993496894836,
-0.09413140267133713,
0.41772839426994324,
... |
https://github.com/huggingface/datasets/issues/5989 | in this case we need to decide what to do with the existing datasets with white space characters (there shouldn't be a lot of them I think) | Set a rule on the config and split names | > should we actually allow characters like spaces? maybe it's better to add validation for whitespace symbols and directly in datasets and raise
https://github.com/huggingface/datasets-server/issues/853
| 27 | Set a rule on the config and split names
> should we actually allow characters like spaces? maybe it's better to add validation for whitespace symbols and directly in datasets and raise
https://github.com/huggingface/datasets-server/issues/853
in this case we need to decide what to do with the existing dataset... | [
-0.059814464300870895,
-0.04150398075580597,
-0.08315175771713257,
0.05934174358844757,
0.19995608925819397,
-0.1714780628681183,
0.46470189094543457,
0.13817793130874634,
0.13507847487926483,
0.22843386232852936,
-0.12750555574893951,
-0.20322933793067932,
-0.012206869199872017,
0.4376853... |
https://github.com/huggingface/datasets/issues/5988 | Unfortunately, I can't reproduce the error. What does the following code return for you?
```python
import requests
from huggingface_hub import hf_hub_url
r = requests.get(hf_hub_url("codeparrot/codeparrot-clean-train", "dataset_infos.json", repo_type="dataset"))
```
Also, can you provide more info about your ne... | ConnectionError: Couldn't reach dataset_infos.json | ### Describe the bug
I'm trying to load codeparrot/codeparrot-clean-train, but get the following error:
ConnectionError: Couldn't reach https://huggingface.co/datasets/codeparrot/codeparrot-clean-train/resolve/main/dataset_infos.json (ConnectionError(ProtocolError('Connection aborted.', ConnectionResetError(104, 'C... | 39 | ConnectionError: Couldn't reach dataset_infos.json
### Describe the bug
I'm trying to load codeparrot/codeparrot-clean-train, but get the following error:
ConnectionError: Couldn't reach https://huggingface.co/datasets/codeparrot/codeparrot-clean-train/resolve/main/dataset_infos.json (ConnectionError(ProtocolErr... | [
-0.36494722962379456,
-0.2202487587928772,
-0.05098268389701843,
0.30980199575424194,
0.14730001986026764,
-0.14014923572540283,
-0.03622368723154068,
0.34880274534225464,
0.2515786290168762,
0.1794554740190506,
-0.23319512605667114,
0.24075737595558167,
0.3046640157699585,
0.2398941814899... |
https://github.com/huggingface/datasets/issues/5987 | Can you explain your use case for `max_shard_size`?
On some systems, there is a limit to the size of a memory-mapped file, so we could consider exposing this parameter in `load_dataset`. | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare | ### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepare` instead.
### Steps to reproduce the bug
https://github.com/huggingface/datasets/blo... | 31 | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare
### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepa... | [
-0.3037675619125366,
-0.7298216223716736,
-0.06928074359893799,
0.5111607909202576,
0.1843271255493164,
-0.10994713008403778,
0.309076189994812,
0.018675237894058228,
0.10166899114847183,
0.26645544171333313,
0.028241507709026337,
-0.14632511138916016,
-0.0025963541120290756,
0.17893892526... |
https://github.com/huggingface/datasets/issues/5987 | In my use case, users may choose a proper size to balance the cost and benefit of using large shard size. (On azure blob or hdfs which may automatically download the shard from background) | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare | ### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepare` instead.
### Steps to reproduce the bug
https://github.com/huggingface/datasets/blo... | 34 | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare
### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepa... | [
-0.36279332637786865,
-0.6249353885650635,
-0.04186156764626503,
0.4104877710342407,
0.08719819784164429,
-0.13526828587055206,
0.3613433539867401,
0.0005356771871447563,
0.06673719733953476,
0.2791599631309509,
-0.04848305508494377,
-0.13423478603363037,
0.08890283107757568,
0.30845573544... |
https://github.com/huggingface/datasets/issues/5987 | But `load_dataset` doesn't support caching (and reading) Arrow datasets from remote storage.
`load_datset_builder` + `download_and_prepare` is not equal to `load_dataset`. The latter has one more step, `builder.as_dataset`, that memory-maps Arrow files, which only works for local files. | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare | ### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepare` instead.
### Steps to reproduce the bug
https://github.com/huggingface/datasets/blo... | 37 | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare
### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepa... | [
-0.29973113536834717,
-0.53922039270401,
-0.03143972530961037,
0.4921461045742035,
0.06836811453104019,
-0.1485273241996765,
0.3775423765182495,
0.06017141789197922,
0.1547858566045761,
0.19435280561447144,
0.05218140035867691,
-0.0291878841817379,
-0.027666974812746048,
0.1110568344593048... |
https://github.com/huggingface/datasets/issues/5987 | Thanks. So if I want to use `IterableDataset` and control the size of single arrow file, how should I organize the data loader? Maybe `load_dataset_build` + `download_and_prepare` + `builder.as_dataset` + `dataset.to_iterable_dataset`? | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare | ### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepare` instead.
### Steps to reproduce the bug
https://github.com/huggingface/datasets/blo... | 31 | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare
### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepa... | [
-0.39479708671569824,
-0.5203035473823547,
-0.08260675519704819,
0.4154074192047119,
0.12109225243330002,
-0.10688042640686035,
0.4182496666908264,
0.002110140398144722,
0.04207978397607803,
0.20657119154930115,
-0.006453279405832291,
-0.10002326220273972,
-0.12214703857898712,
0.327320665... |
https://github.com/huggingface/datasets/issues/5987 | Yes, this should work.
I think we can expose `max_shard_size` in `load_dataset`, so feel free to open a PR. | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare | ### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepare` instead.
### Steps to reproduce the bug
https://github.com/huggingface/datasets/blo... | 19 | Why max_shard_size is not supported in load_dataset and passed to download_and_prepare
### Describe the bug
https://github.com/huggingface/datasets/blob/a8a797cc92e860c8d0df71e0aa826f4d2690713e/src/datasets/load.py#L1809
What I can to is break the `load_dataset` and use `load_datset_builder` + `download_and_prepa... | [
-0.31491613388061523,
-0.7279372215270996,
-0.04494934901595116,
0.4540124833583832,
0.2202315777540207,
-0.13597308099269867,
0.30974528193473816,
0.026850003749132156,
0.13989853858947754,
0.24466417729854584,
-0.044594064354896545,
-0.1049724593758583,
0.028704285621643066,
0.3151349425... |
https://github.com/huggingface/datasets/issues/5985 | This is a known issue: https://github.com/huggingface/datasets/issues/3847.
Fixing this requires significant work - rewriting the `tokenizers` lib to make them immutable.
The current solution is to pass `cache_file_name` to `map` to use that file for caching or calling a tokenizer before `map` (with the same set ... | Cannot reuse tokenizer object for dataset map | ### Describe the bug
Related to https://github.com/huggingface/transformers/issues/24441. Not sure if this is a tokenizer issue or caching issue, so filing in both.
Passing the tokenizer to the dataset map function causes the tokenizer to be fingerprinted weirdly. After calling the tokenizer with arguments like pad... | 54 | Cannot reuse tokenizer object for dataset map
### Describe the bug
Related to https://github.com/huggingface/transformers/issues/24441. Not sure if this is a tokenizer issue or caching issue, so filing in both.
Passing the tokenizer to the dataset map function causes the tokenizer to be fingerprinted weirdly. Aft... | [
0.056222617626190186,
-0.2532831132411957,
0.16724401712417603,
0.24009664356708527,
0.31940531730651855,
-0.15103591978549957,
0.43399614095687866,
-0.1082683876156807,
-0.0809454694390297,
0.011936616152524948,
-0.17815035581588745,
0.34636059403419495,
-0.10086217522621155,
-0.057456616... |
https://github.com/huggingface/datasets/issues/5984 | For this to be possible, we would have to switch from the "Streaming" Arrow format to the "Random Access" (IPC/Feather) format, which allows reading arbitrary record batches (explained [here](https://arrow.apache.org/docs/python/ipc.html)). We could then use these batches to construct shards.
@lhoestq @albertvillano... | AutoSharding IterableDataset's when num_workers > 1 | ### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
Too many dataloader workers: 2 (max is dataset.n_shard... | 86 | AutoSharding IterableDataset's when num_workers > 1
### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
... | [
-0.41809478402137756,
-0.10188451409339905,
-0.1456300914287567,
0.07439666986465454,
-0.303873747587204,
-0.1833363175392151,
0.6748954653739929,
0.1863306760787964,
-0.15313678979873657,
0.19728778302669525,
0.06172633171081543,
0.15588393807411194,
-0.30952349305152893,
-0.2680183351039... |
https://github.com/huggingface/datasets/issues/5984 | Alternatively we could support multiprocessing map for iterable datasets and let the user do the CPU intensive task there ?
This way it would work on arrow data but also on any iterable dataset | AutoSharding IterableDataset's when num_workers > 1 | ### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
Too many dataloader workers: 2 (max is dataset.n_shard... | 34 | AutoSharding IterableDataset's when num_workers > 1
### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
... | [
-0.40869325399398804,
-0.1982613503932953,
-0.14433178305625916,
0.12288518249988556,
-0.23263713717460632,
-0.10178466141223907,
0.5957220196723938,
0.1270076036453247,
-0.023237287998199463,
0.24934904277324677,
0.12391463667154312,
0.3039184510707855,
-0.30772465467453003,
-0.1494929045... |
https://github.com/huggingface/datasets/issues/5984 | > For this to be possible, we would have to switch from the "Streaming" Arrow format to the "Random Access" (IPC/Feather) format, which allows reading arbitrary record batches (explained [here](https://arrow.apache.org/docs/python/ipc.html)). We could then use these batches to construct shards.
>
> @lhoestq @albertv... | AutoSharding IterableDataset's when num_workers > 1 | ### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
Too many dataloader workers: 2 (max is dataset.n_shard... | 166 | AutoSharding IterableDataset's when num_workers > 1
### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
... | [
-0.41506555676460266,
-0.11562855541706085,
-0.14575722813606262,
0.08082820475101471,
-0.29021868109703064,
-0.17778001725673676,
0.6815129518508911,
0.202985480427742,
-0.1653580367565155,
0.19864985346794128,
0.050410643219947815,
0.16098478436470032,
-0.2835122346878052,
-0.28057730197... |
https://github.com/huggingface/datasets/issues/5984 | > Could you explain why you'd need to change the arrow format?
This way workers have random access to the location of the file where its dataset subset starts. Currently we're using the Arrow streaming format which doesn't include the metadata of the record batches offsets. This is needed here to efficiently split a... | AutoSharding IterableDataset's when num_workers > 1 | ### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
Too many dataloader workers: 2 (max is dataset.n_shard... | 60 | AutoSharding IterableDataset's when num_workers > 1
### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
... | [
-0.44266700744628906,
-0.20271021127700806,
-0.12075897306203842,
0.07688800990581512,
-0.2469574511051178,
-0.19033944606781006,
0.677727222442627,
0.15579548478126526,
-0.08213454484939575,
0.2221689075231552,
0.1517815887928009,
0.2422218918800354,
-0.303825706243515,
-0.155423939228057... |
https://github.com/huggingface/datasets/issues/5984 | > > Could you explain why you'd need to change the arrow format?
>
> This way workers have random access to the location of the file where its dataset subset starts. Currently we're using the Arrow streaming format which doesn't include the metadata of the record batches offsets. This is needed here to efficiently s... | AutoSharding IterableDataset's when num_workers > 1 | ### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
Too many dataloader workers: 2 (max is dataset.n_shard... | 158 | AutoSharding IterableDataset's when num_workers > 1
### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
... | [
-0.46709948778152466,
-0.1791263073682785,
-0.11632896214723587,
0.049152933061122894,
-0.26057004928588867,
-0.17280226945877075,
0.6774322986602783,
0.16799107193946838,
-0.08234615623950958,
0.23683150112628937,
0.11716029047966003,
0.24954739212989807,
-0.23069238662719727,
-0.20272251... |
https://github.com/huggingface/datasets/issues/5984 | > For instance, using pyArrow, you could use RecordBatchStreamReader to zero-copy iterate over records with read_next_batch and then only initiate the next step for records modulo worker shard.
That works indeed ! And what we meant is that you can make it even faster to instantiate. Indeed using RecordBatchStreamRea... | AutoSharding IterableDataset's when num_workers > 1 | ### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
Too many dataloader workers: 2 (max is dataset.n_shard... | 110 | AutoSharding IterableDataset's when num_workers > 1
### Feature request
Minimal Example
```
import torch
from datasets import IterableDataset
d = IterableDataset.from_file(<file_name>)
dl = torch.utils.data.dataloader.DataLoader(d,num_workers=3)
for sample in dl:
print(sample)
```
Warning:
... | [
-0.5463466048240662,
-0.06826047599315643,
-0.12397506833076477,
0.021309219300746918,
-0.2813234329223633,
-0.164022758603096,
0.6526508331298828,
0.17368414998054504,
0.03373450040817261,
0.2698051333427429,
0.10144931823015213,
0.215835303068161,
-0.3545735478401184,
-0.2576067447662353... |
https://github.com/huggingface/datasets/issues/5982 | This wasn’t working for me a bit earlier, but it looks to be back up now | 404 on Datasets Documentation Page | ### Describe the bug
Getting a 404 from the Hugging Face Datasets docs page:
https://huggingface.co/docs/datasets/index
### Steps to reproduce the bug
1. Go to URL https://huggingface.co/docs/datasets/index
2. Notice 404 not found
### Expected behavior
URL should either show docs or redirect to new location
#... | 16 | 404 on Datasets Documentation Page
### Describe the bug
Getting a 404 from the Hugging Face Datasets docs page:
https://huggingface.co/docs/datasets/index
### Steps to reproduce the bug
1. Go to URL https://huggingface.co/docs/datasets/index
2. Notice 404 not found
### Expected behavior
URL should either sho... | [
0.2358924001455307,
-0.0897052064538002,
-0.038213007152080536,
0.19043193757534027,
0.17236018180847168,
0.2296682745218277,
0.030324263498187065,
0.17348481714725494,
-0.0862479954957962,
0.1142839714884758,
-0.04872129485011101,
-0.027137374505400658,
-0.05816924571990967,
0.07129813730... |
https://github.com/huggingface/datasets/issues/5982 | We had a minor issue updating the docs after the latest release. It should work now :). | 404 on Datasets Documentation Page | ### Describe the bug
Getting a 404 from the Hugging Face Datasets docs page:
https://huggingface.co/docs/datasets/index
### Steps to reproduce the bug
1. Go to URL https://huggingface.co/docs/datasets/index
2. Notice 404 not found
### Expected behavior
URL should either show docs or redirect to new location
#... | 17 | 404 on Datasets Documentation Page
### Describe the bug
Getting a 404 from the Hugging Face Datasets docs page:
https://huggingface.co/docs/datasets/index
### Steps to reproduce the bug
1. Go to URL https://huggingface.co/docs/datasets/index
2. Notice 404 not found
### Expected behavior
URL should either sho... | [
0.17083513736724854,
-0.21462519466876984,
-0.02505388669669628,
0.2500726580619812,
0.15249451994895935,
0.2086954116821289,
0.030402902513742447,
0.18846707046031952,
-0.06417596340179443,
0.19951173663139343,
-0.07982753217220306,
0.026617862284183502,
-0.07252615690231323,
0.1539390981... |
https://github.com/huggingface/datasets/issues/5981 | I think it's more likely that this issue is related to PyTorch than Datasets, as PyTorch (on import) registers functions to execute when forking a process. Maybe this is the culprit: https://github.com/pytorch/pytorch/issues/99625 | Only two cores are getting used in sagemaker with pytorch 3.10 kernel | ### Describe the bug
When using the newer pytorch 3.10 kernel, only 2 cores are being used by huggingface filter and map functions. The Pytorch 3.9 kernel would use as many cores as specified in the num_proc field.
We have solved this in our own code by placing the following snippet in the code that is called insi... | 32 | Only two cores are getting used in sagemaker with pytorch 3.10 kernel
### Describe the bug
When using the newer pytorch 3.10 kernel, only 2 cores are being used by huggingface filter and map functions. The Pytorch 3.9 kernel would use as many cores as specified in the num_proc field.
We have solved this in our o... | [
-0.16755031049251556,
-0.698649525642395,
-0.006346769630908966,
0.3419397175312042,
-0.08470866829156876,
-0.290692538022995,
0.17301566898822784,
0.061169158667325974,
0.36152422428131104,
0.2239425778388977,
-0.09223724901676178,
0.6258037090301514,
0.17458635568618774,
0.04712916910648... |
https://github.com/huggingface/datasets/issues/5981 | From reading that ticket, it may be down in mkl? Is it worth hotfixing in the meantime, with the express intention of turning it off? I know that's a horribly crufty solution, but it's also deeply frustrating to be limited to 2 cores for operations as simple as filtration. | Only two cores are getting used in sagemaker with pytorch 3.10 kernel | ### Describe the bug
When using the newer pytorch 3.10 kernel, only 2 cores are being used by huggingface filter and map functions. The Pytorch 3.9 kernel would use as many cores as specified in the num_proc field.
We have solved this in our own code by placing the following snippet in the code that is called insi... | 49 | Only two cores are getting used in sagemaker with pytorch 3.10 kernel
### Describe the bug
When using the newer pytorch 3.10 kernel, only 2 cores are being used by huggingface filter and map functions. The Pytorch 3.9 kernel would use as many cores as specified in the num_proc field.
We have solved this in our o... | [
-0.16755031049251556,
-0.698649525642395,
-0.006346769630908966,
0.3419397175312042,
-0.08470866829156876,
-0.290692538022995,
0.17301566898822784,
0.061169158667325974,
0.36152422428131104,
0.2239425778388977,
-0.09223724901676178,
0.6258037090301514,
0.17458635568618774,
0.04712916910648... |
https://github.com/huggingface/datasets/issues/5981 | @mariosasko @mmr-crexi I had the exact same problem on my kubernetes cluster. the datasets subprocess only user 1 and 17 core | Only two cores are getting used in sagemaker with pytorch 3.10 kernel | ### Describe the bug
When using the newer pytorch 3.10 kernel, only 2 cores are being used by huggingface filter and map functions. The Pytorch 3.9 kernel would use as many cores as specified in the num_proc field.
We have solved this in our own code by placing the following snippet in the code that is called insi... | 21 | Only two cores are getting used in sagemaker with pytorch 3.10 kernel
### Describe the bug
When using the newer pytorch 3.10 kernel, only 2 cores are being used by huggingface filter and map functions. The Pytorch 3.9 kernel would use as many cores as specified in the num_proc field.
We have solved this in our o... | [
-0.16755031049251556,
-0.698649525642395,
-0.006346769630908966,
0.3419397175312042,
-0.08470866829156876,
-0.290692538022995,
0.17301566898822784,
0.061169158667325974,
0.36152422428131104,
0.2239425778388977,
-0.09223724901676178,
0.6258037090301514,
0.17458635568618774,
0.04712916910648... |
https://github.com/huggingface/datasets/issues/5980 | Yes, it seems to be working now. In case it's helpful, the outage lasted several days. It was failing as late as yesterday morning. | Viewing dataset card returns “502 Bad Gateway” | The url is: https://huggingface.co/datasets/Confirm-Labs/pile_ngrams_trigrams
I am able to successfully view the “Files and versions” tab: [Confirm-Labs/pile_ngrams_trigrams at main](https://huggingface.co/datasets/Confirm-Labs/pile_ngrams_trigrams/tree/main)
Any help would be appreciated! Thanks! I hope this is ... | 24 | Viewing dataset card returns “502 Bad Gateway”
The url is: https://huggingface.co/datasets/Confirm-Labs/pile_ngrams_trigrams
I am able to successfully view the “Files and versions” tab: [Confirm-Labs/pile_ngrams_trigrams at main](https://huggingface.co/datasets/Confirm-Labs/pile_ngrams_trigrams/tree/main)
Any h... | [
-0.3045494854450226,
-0.08470316231250763,
0.019787978380918503,
0.2753653824329376,
0.37162843346595764,
-0.18100424110889435,
-0.0117882639169693,
0.2689245343208313,
-0.15327826142311096,
-0.15398728847503662,
-0.10853318870067596,
0.07101313769817352,
0.2886524498462677,
0.616469919681... |
https://github.com/huggingface/datasets/issues/5975 | Hi ! can you try to set the upper case environment variables `HTTP_PROXY` and `HTTPS_PROXY` ?
We use `aiohttp` for streaming and it uses case sensitive environment variables | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 28 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.20426958799362183,
-0.12143173813819885,
0.09346703439950943,
-0.11213923990726471,
0.3825293183326721,
0.020822487771511078,
0.13100317120552063,
0.1265920102596283,
-0.06272534281015396,
0.1365622580051422,
0.07379918545484543,
-0.17025922238826752,
0.20304013788700104,
0.098815046250... |
https://github.com/huggingface/datasets/issues/5975 | Hi, thanks for the quick reply.
I set the uppercase env variables with
`
os.environ['HTTP_PROXY'] = "http://example.com:xxxx"
os.environ['HTTPS_PROXY'] = "http://example.com:xxxx"
`
However, I still get the same error.
One thing that could be helpfull: When downloading a dataset without streaming i get... | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 62 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.15595516562461853,
-0.09131626784801483,
0.09897124022245407,
-0.10627853870391846,
0.39636728167533875,
-0.00484660267829895,
0.11199093610048294,
0.07572492212057114,
0.03164759278297424,
0.15219303965568542,
0.08398015052080154,
-0.21240690350532532,
0.17867884039878845,
0.1644148826... |
https://github.com/huggingface/datasets/issues/5975 | Are you able to use `aiohttp` to get the file at `https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json` using your proxy ? | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 16 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.25485262274742126,
-0.0679943710565567,
0.09301558881998062,
-0.12202920019626617,
0.3035862445831299,
-0.011668860912322998,
0.1310204267501831,
0.15154731273651123,
0.009722955524921417,
0.1170535758137703,
0.04780591279268265,
-0.19018292427062988,
0.2405724972486496,
0.1584572046995... |
https://github.com/huggingface/datasets/issues/5975 | It only works when passing trust_env=True when creating the ClientSession, as well as setting ssl=False.
Working Example:
```
import os
os.environ['HTTP_PROXY'] = "xyz"
os.environ['HTTPS_PROXY'] = "xyz"
import asyncio
import aiohttp
async def download_pep(url):
async with aiohttp.ClientSession(tr... | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 98 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.2875823378562927,
0.027243584394454956,
0.11844801157712936,
-0.1200261116027832,
0.31531858444213867,
-0.05895368754863739,
0.21909725666046143,
0.13560554385185242,
0.06980419903993607,
0.1013476699590683,
0.017686940729618073,
-0.1512751579284668,
0.3091881275177002,
0.10330618172883... |
https://github.com/huggingface/datasets/issues/5975 | We do pass `trust_env` as well. Could you share the full stack trace you get when streaming using `datasets` ? That could help locate where we might have forgotten to pass `trust_env` | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 32 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.22055009007453918,
-0.04736737906932831,
0.09124628454446793,
-0.10017752647399902,
0.36972758173942566,
-0.011570446193218231,
0.10142427682876587,
0.1315036416053772,
0.010005481541156769,
0.11178212612867355,
0.09104365110397339,
-0.15345272421836853,
0.20461568236351013,
0.125229984... |
https://github.com/huggingface/datasets/issues/5975 | Is there a way to disable ssl verification when streaming a dataset. I suspect this might be the isssue with my proxy.
Here you go:
```
FileNotFoundError Traceback (most recent call last)
Cell In[8], line 3
1 from datasets import load_dataset
----> 3 ds = load_dataset("facebook... | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 282 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.24882769584655762,
-0.05979403853416443,
0.1007160171866417,
-0.10208457708358765,
0.24106337130069733,
-0.006721898913383484,
0.17118151485919952,
0.06702045351266861,
-0.020424023270606995,
0.1228136420249939,
0.11307883262634277,
-0.21383479237556458,
0.2762974798679352,
0.0067338198... |
https://github.com/huggingface/datasets/issues/5975 | > Is there a way to disable ssl verification when streaming a dataset.
I don't think so.
We use `fsspec` HTTPFileSystem implementation that is based on `aiohttp`. If you register a subclass of HTTPFileSystem that has SSL disabled by default it could work, but I wouldn't recommended it because it can raise securit... | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 54 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.25454550981521606,
-0.01584668457508087,
0.1071445494890213,
-0.10625217854976654,
0.23831488192081451,
-0.018652863800525665,
0.13900716602802277,
0.07588482648134232,
0.0818885862827301,
0.13150793313980103,
0.12248197197914124,
-0.2498541623353958,
0.30348244309425354,
0.170065134763... |
https://github.com/huggingface/datasets/issues/5975 | Okay thanks for your help! I guess I have to figure out how to improve the proxy environment / see if I can make it work with ssl connections. | Streaming Dataset behind Proxy - FileNotFoundError | ### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure to log in with `huggingface-cli login`.
```
I hav... | 29 | Streaming Dataset behind Proxy - FileNotFoundError
### Describe the bug
When trying to stream a dataset i get the following error after a few minutes of waiting.
```
FileNotFoundError: https://huggingface.co/datasets/facebook/voxpopuli/resolve/main/data/n_files.json
If the repo is private or gated, make sure ... | [
-0.22630605101585388,
-0.08385404199361801,
0.08787970244884491,
-0.11146005988121033,
0.31258198618888855,
-0.019106529653072357,
0.1175183653831482,
0.1108732521533966,
-0.001358620822429657,
0.05213000997900963,
0.09835232049226761,
-0.17861908674240112,
0.21179527044296265,
0.142906606... |
https://github.com/huggingface/datasets/issues/5971 | Loading a local dataset also works the same way when `data_files` are not specified, so I agree we should make this info easier to discover
cc @stevhliu | Docs: make "repository structure" easier to find | The page https://huggingface.co/docs/datasets/repository_structure explains how to create a simple repository structure without a dataset script.
It's the simplest way to create a dataset and should be easier to find, particularly on the docs' first pages. | 27 | Docs: make "repository structure" easier to find
The page https://huggingface.co/docs/datasets/repository_structure explains how to create a simple repository structure without a dataset script.
It's the simplest way to create a dataset and should be easier to find, particularly on the docs' first pages.
Loading a... | [
-0.23497682809829712,
0.09987065196037292,
-0.1013665497303009,
0.31530508399009705,
0.13123159110546112,
-0.038293980062007904,
-0.043238017708063126,
0.45117151737213135,
0.034013889729976654,
0.5910804271697998,
0.01932922750711441,
0.1766023337841034,
0.1995181143283844,
0.228219941258... |
https://github.com/huggingface/datasets/issues/5971 | @benjaminbrown038 Yes, it is. Maybe @stevhliu can give some pointers on improving this doc page's discoverability. | Docs: make "repository structure" easier to find | The page https://huggingface.co/docs/datasets/repository_structure explains how to create a simple repository structure without a dataset script.
It's the simplest way to create a dataset and should be easier to find, particularly on the docs' first pages. | 16 | Docs: make "repository structure" easier to find
The page https://huggingface.co/docs/datasets/repository_structure explains how to create a simple repository structure without a dataset script.
It's the simplest way to create a dataset and should be easier to find, particularly on the docs' first pages.
@benjamin... | [
-0.05660982429981232,
0.03940857946872711,
-0.1552848368883133,
0.22599880397319794,
0.10666590929031372,
0.0058924853801727295,
-0.18545351922512054,
0.4604344069957733,
-0.10095638781785965,
0.3205561339855194,
0.030891314148902893,
0.16898559033870697,
0.09326285123825073,
0.14402456581... |
https://github.com/huggingface/datasets/issues/5971 | I think we can add a version of the [Main use-case](https://huggingface.co/docs/datasets/repository_structure#main-usecase) section to the [Share a dataset to the Hub](https://huggingface.co/docs/datasets/upload_dataset) tutorial.
Currently, it doesn't tell you *how* to structure the repository; it only tells you h... | Docs: make "repository structure" easier to find | The page https://huggingface.co/docs/datasets/repository_structure explains how to create a simple repository structure without a dataset script.
It's the simplest way to create a dataset and should be easier to find, particularly on the docs' first pages. | 77 | Docs: make "repository structure" easier to find
The page https://huggingface.co/docs/datasets/repository_structure explains how to create a simple repository structure without a dataset script.
It's the simplest way to create a dataset and should be easier to find, particularly on the docs' first pages.
I think w... | [
-0.022013843059539795,
-0.17372795939445496,
-0.025757355615496635,
0.12676502764225006,
0.11845462024211884,
0.15033085644245148,
-0.09882107377052307,
0.09547268599271774,
-0.17971496284008026,
0.3140157461166382,
0.13280057907104492,
0.15143832564353943,
0.1374928504228592,
0.2845121026... |
https://github.com/huggingface/datasets/issues/5970 | Here's a minimal way to reproduce the bug, for the sake of convenience.
````
from datasets import Dataset, DatasetInfo, load_dataset
episodes_dict = {"test":[1,2,3],"test2": [1,2,4]}
hugging_face_dataset = Dataset.from_dict(
episodes_dict, info=DatasetInfo(description="test_str")
)
print(hugging_face_d... | description disappearing from Info when Uploading a Dataset Created with `from_dict` | ### Describe the bug
When uploading a dataset created locally using `from_dict` with a specified `description` field. It appears before upload, but is missing after upload and re-download.
### Steps to reproduce the bug
I think the most relevant pattern in the code might be the following lines:
```
descr... | 37 | description disappearing from Info when Uploading a Dataset Created with `from_dict`
### Describe the bug
When uploading a dataset created locally using `from_dict` with a specified `description` field. It appears before upload, but is missing after upload and re-download.
### Steps to reproduce the bug
I ... | [
0.2038726657629013,
0.06310242414474487,
0.08928681910037994,
0.23943296074867249,
0.43252235651016235,
-0.08675259351730347,
0.1411246955394745,
0.22590745985507965,
-0.469280481338501,
0.09071217477321625,
0.3411741256713867,
0.14465267956256866,
0.3179815709590912,
0.3623771071434021,
... |
https://github.com/huggingface/datasets/issues/5970 | Thanks for reporting !
For now I would recommend uploading a separate JSON file for your metadata.
Alternatively you can upload a second configuration of the dataset containing your metadata but this feature is not released yet (though you can already use it from [here](https://github.com/huggingface/datasets/pul... | description disappearing from Info when Uploading a Dataset Created with `from_dict` | ### Describe the bug
When uploading a dataset created locally using `from_dict` with a specified `description` field. It appears before upload, but is missing after upload and re-download.
### Steps to reproduce the bug
I think the most relevant pattern in the code might be the following lines:
```
descr... | 50 | description disappearing from Info when Uploading a Dataset Created with `from_dict`
### Describe the bug
When uploading a dataset created locally using `from_dict` with a specified `description` field. It appears before upload, but is missing after upload and re-download.
### Steps to reproduce the bug
I ... | [
0.2038726657629013,
0.06310242414474487,
0.08928681910037994,
0.23943296074867249,
0.43252235651016235,
-0.08675259351730347,
0.1411246955394745,
0.22590745985507965,
-0.469280481338501,
0.09071217477321625,
0.3411741256713867,
0.14465267956256866,
0.3179815709590912,
0.3623771071434021,
... |
https://github.com/huggingface/datasets/issues/5968 | The issue commes from the dataset itself and is not related to the `datasets` lib
see https://huggingface.co/datasets/mozilla-foundation/common_voice_6_1/blob/2c475b3b88e0f2e5828f830a4b91618a25ff20b7/common_voice_6_1.py#L148-L152 | Common Voice datasets still need `use_auth_token=True` | ### Describe the bug
We don't need to pass `use_auth_token=True` anymore to download gated datasets or models, so the following should work if correctly logged in.
```py
from datasets import load_dataset
load_dataset("mozilla-foundation/common_voice_6_1", "tr", split="train+validation")
```
However it throw... | 17 | Common Voice datasets still need `use_auth_token=True`
### Describe the bug
We don't need to pass `use_auth_token=True` anymore to download gated datasets or models, so the following should work if correctly logged in.
```py
from datasets import load_dataset
load_dataset("mozilla-foundation/common_voice_6_1",... | [
-0.19967004656791687,
0.01185956597328186,
0.1525600701570511,
-0.005796443670988083,
0.44708141684532166,
0.05265425145626068,
0.5136598348617554,
0.12041813135147095,
0.031704775989055634,
-0.047319553792476654,
-0.5297954678535461,
0.042947519570589066,
0.03492205962538719,
-0.029982902... |
https://github.com/huggingface/datasets/issues/5968 | Addressed in:
* `mozilla-foundation/common_voice_1_0` [PR](https://huggingface.co/datasets/mozilla-foundation/common_voice_1_0/discussions/4)
* `mozilla-foundation/common_voice_2_0` [PR](https://huggingface.co/datasets/mozilla-foundation/common_voice_2_0/discussions/3)
* `mozilla-foundation/common_voice_3_0` [PR](... | Common Voice datasets still need `use_auth_token=True` | ### Describe the bug
We don't need to pass `use_auth_token=True` anymore to download gated datasets or models, so the following should work if correctly logged in.
```py
from datasets import load_dataset
load_dataset("mozilla-foundation/common_voice_6_1", "tr", split="train+validation")
```
However it throw... | 38 | Common Voice datasets still need `use_auth_token=True`
### Describe the bug
We don't need to pass `use_auth_token=True` anymore to download gated datasets or models, so the following should work if correctly logged in.
```py
from datasets import load_dataset
load_dataset("mozilla-foundation/common_voice_6_1",... | [
-0.19967004656791687,
0.01185956597328186,
0.1525600701570511,
-0.005796443670988083,
0.44708141684532166,
0.05265425145626068,
0.5136598348617554,
0.12041813135147095,
0.031704775989055634,
-0.047319553792476654,
-0.5297954678535461,
0.042947519570589066,
0.03492205962538719,
-0.029982902... |
https://github.com/huggingface/datasets/issues/5967 | This must be due to DatasetInfo.from_merge which drops them and is used in `concatenate_datasets`.
And you're experiencing this issue because multiprocessing does concatenate the resulting datasets from each process.
Maybe they should be kept if all the subdatasets share the same values for config_name and split | Config name / split name lost after map with multiproc | ### Describe the bug
Performing a `.map` method on a dataset loses it's config name / split name only if run with multiproc
### Steps to reproduce the bug
```python
from datasets import Audio, load_dataset
from transformers import AutoFeatureExtractor
import numpy as np
# load dummy dataset
libri = load_datas... | 46 | Config name / split name lost after map with multiproc
### Describe the bug
Performing a `.map` method on a dataset loses it's config name / split name only if run with multiproc
### Steps to reproduce the bug
```python
from datasets import Audio, load_dataset
from transformers import AutoFeatureExtractor
impor... | [
-0.3114316463470459,
-0.3305574655532837,
0.0021502934396266937,
0.09281058609485626,
0.29288220405578613,
-0.11196547001600266,
0.59432053565979,
0.13082531094551086,
-0.08384830504655838,
0.6229397058486938,
0.04151308536529541,
0.5787537097930908,
-0.13195489346981049,
0.307406395673751... |
https://github.com/huggingface/datasets/issues/5965 | Thanks for reporting!
Specifying the target features explicitly should avoid this error:
```python
dataset = dataset.map(
batch_process,
batched=True,
batch_size=1,
num_proc=1,
remove_columns=dataset.column_names,
features=datasets.Features({"texts": datasets.Sequence(datasets.Value(... | "Couldn't cast array of type" in complex datasets | ### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competing for the same field value.
This is prone to hap... | 61 | "Couldn't cast array of type" in complex datasets
### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competi... | [
-0.3982987403869629,
-0.056902915239334106,
-0.04033725708723068,
0.17765331268310547,
0.4318235218524933,
0.226649209856987,
0.33259859681129456,
0.3314976692199707,
0.3767266571521759,
-0.019378453493118286,
0.023040346801280975,
0.3431892991065979,
-0.15088258683681488,
0.02486550062894... |
https://github.com/huggingface/datasets/issues/5965 | Hi @mariosasko thanks for the context, this is helpful to know. Would it be worth having some logic to generate this explicit feature specification automatically if a type annotation for a .map returns a dataclass that can be inferred?
Feels like something that would be easy to implement and could save memory / deal... | "Couldn't cast array of type" in complex datasets | ### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competing for the same field value.
This is prone to hap... | 61 | "Couldn't cast array of type" in complex datasets
### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competi... | [
-0.3982987403869629,
-0.056902915239334106,
-0.04033725708723068,
0.17765331268310547,
0.4318235218524933,
0.226649209856987,
0.33259859681129456,
0.3314976692199707,
0.3767266571521759,
-0.019378453493118286,
0.023040346801280975,
0.3431892991065979,
-0.15088258683681488,
0.02486550062894... |
https://github.com/huggingface/datasets/issues/5965 | > . Would it be worth having some logic to generate this explicit feature specification automatically if a type annotation for a .map returns a dataclass that can be inferred?
Interesting proposal! Yes, we could consider doing this if the (return) type hint is `TypedDict`, and raise an error that type hints are inco... | "Couldn't cast array of type" in complex datasets | ### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competing for the same field value.
This is prone to hap... | 62 | "Couldn't cast array of type" in complex datasets
### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competi... | [
-0.3982987403869629,
-0.056902915239334106,
-0.04033725708723068,
0.17765331268310547,
0.4318235218524933,
0.226649209856987,
0.33259859681129456,
0.3314976692199707,
0.3767266571521759,
-0.019378453493118286,
0.023040346801280975,
0.3431892991065979,
-0.15088258683681488,
0.02486550062894... |
https://github.com/huggingface/datasets/issues/5965 | @mariosasko Put up an initial PR to implement this proposal. Let me know your thoughts on direction and what else should be in-scope here. | "Couldn't cast array of type" in complex datasets | ### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competing for the same field value.
This is prone to hap... | 24 | "Couldn't cast array of type" in complex datasets
### Describe the bug
When doing a map of a dataset with complex types, sometimes `datasets` is unable to interpret the valid schema of a returned datasets.map() function. This often comes from conflicting types, like when both empty lists and filled lists are competi... | [
-0.3982987403869629,
-0.056902915239334106,
-0.04033725708723068,
0.17765331268310547,
0.4318235218524933,
0.226649209856987,
0.33259859681129456,
0.3314976692199707,
0.3767266571521759,
-0.019378453493118286,
0.023040346801280975,
0.3431892991065979,
-0.15088258683681488,
0.02486550062894... |
https://github.com/huggingface/datasets/issues/5963 | i got error using method from_spark when using multi-node Spark cluster. seems could only use "from_spark" in local? | Got an error _pickle.PicklingError use Dataset.from_spark. | python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark(df, keep_in_memory=True,
cache_dir="... | 18 | Got an error _pickle.PicklingError use Dataset.from_spark.
python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark... | [
-0.48843914270401,
-0.16567064821720123,
0.023851711302995682,
0.342393159866333,
0.29546353220939636,
-0.0016700103878974915,
0.5267864465713501,
0.13773950934410095,
0.08715443313121796,
0.09700290113687515,
0.08231262862682343,
0.6951181888580322,
-0.18221795558929443,
0.534705400466919... |
https://github.com/huggingface/datasets/issues/5963 | cc @maddiedawson it looks like there an issue with `_validate_cache_dir` ?
It looks like the function passed to mapPartitions has a reference to the Spark dataset builder, and therefore contains the SparkContext itself.
I think it can be fixed by defining `create_cache_and_write_probe` outside the Spark dataset b... | Got an error _pickle.PicklingError use Dataset.from_spark. | python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark(df, keep_in_memory=True,
cache_dir="... | 54 | Got an error _pickle.PicklingError use Dataset.from_spark.
python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark... | [
-0.48843914270401,
-0.16567064821720123,
0.023851711302995682,
0.342393159866333,
0.29546353220939636,
-0.0016700103878974915,
0.5267864465713501,
0.13773950934410095,
0.08715443313121796,
0.09700290113687515,
0.08231262862682343,
0.6951181888580322,
-0.18221795558929443,
0.534705400466919... |
https://github.com/huggingface/datasets/issues/5963 | Just saw this; thanks for flagging! Your proposed solution sounds good. I can prepare a PR | Got an error _pickle.PicklingError use Dataset.from_spark. | python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark(df, keep_in_memory=True,
cache_dir="... | 16 | Got an error _pickle.PicklingError use Dataset.from_spark.
python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark... | [
-0.48843914270401,
-0.16567064821720123,
0.023851711302995682,
0.342393159866333,
0.29546353220939636,
-0.0016700103878974915,
0.5267864465713501,
0.13773950934410095,
0.08715443313121796,
0.09700290113687515,
0.08231262862682343,
0.6951181888580322,
-0.18221795558929443,
0.534705400466919... |
https://github.com/huggingface/datasets/issues/5963 | @maddiedawson can you show me the demo ,so i can test in local .before your PR | Got an error _pickle.PicklingError use Dataset.from_spark. | python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark(df, keep_in_memory=True,
cache_dir="... | 16 | Got an error _pickle.PicklingError use Dataset.from_spark.
python 3.9.2
Got an error _pickle.PicklingError use Dataset.from_spark.
Did the dataset import load data from spark dataframe using multi-node Spark cluster
df = spark.read.parquet(args.input_data).repartition(50)
ds = Dataset.from_spark... | [
-0.48843914270401,
-0.16567064821720123,
0.023851711302995682,
0.342393159866333,
0.29546353220939636,
-0.0016700103878974915,
0.5267864465713501,
0.13773950934410095,
0.08715443313121796,
0.09700290113687515,
0.08231262862682343,
0.6951181888580322,
-0.18221795558929443,
0.534705400466919... |
https://github.com/huggingface/datasets/issues/5961 | Does "number of shards" refer to the total number of data?
my config:
nproc_per_node=2
ds=ds['train'] = load_dataset(streaming=True).take(50000)
I'm test again: in prepare_data(), data have the same for each GPU
| IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 29 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.44223552942276,
-0.3692326247692108,
-0.06835275143384933,
0.2758820652961731,
0.11674202233552933,
-0.10412335395812988,
0.35168322920799255,
0.17539653182029724,
-0.2317139059305191,
0.17232121527194977,
0.05940701812505722,
0.4387744963169098,
0.06595718860626221,
0.37266895174980164... |
https://github.com/huggingface/datasets/issues/5961 | The number of shards is `ds.n_shards`. It corresponds generally to the number of files the dataset is made of, to be able to distribute to several nodes.
**You don't end up with the same data per GPU**. But all the samples are going through your preprocessing function you pass to map. They are just skipped afterward... | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 64 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.4653586149215698,
-0.46394428610801697,
-0.09561261534690857,
0.24369829893112183,
0.0934542790055275,
-0.053702130913734436,
0.32686930894851685,
0.2737382650375366,
0.046611398458480835,
0.1674361675977707,
-0.0384560264647007,
0.46263110637664795,
0.19708502292633057,
0.2577080726623... |
https://github.com/huggingface/datasets/issues/5961 | For each GPU, although see the same data in prepare_data(), the actual training data will not be the same in the end.
Is my understanding correct?
Where can I print the actual training data for each GPU? | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 37 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.33584466576576233,
-0.42952442169189453,
-0.08595259487628937,
0.35857486724853516,
0.07497124373912811,
0.0813596099615097,
0.3470378816127777,
0.24788913130760193,
-0.16065239906311035,
0.04934052377939224,
-0.035910945385694504,
0.3949976861476898,
0.20944832265377045,
0.338742285966... |
https://github.com/huggingface/datasets/issues/5961 | > For each GPU, although see the same data in prepare_data(), the actual training data will not be the same in the end.
Is my understanding correct?
Yes exactly :)
> Where can I print the actual training data for each GPU?
You should call print in the data_collator | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 49 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.347914457321167,
-0.4232020378112793,
-0.09536373615264893,
0.37167900800704956,
0.12340044975280762,
0.06726889312267303,
0.3374471664428711,
0.24970708787441254,
-0.17176666855812073,
0.045485466718673706,
-0.06359327584505081,
0.4129130244255066,
0.1829591542482376,
0.352101325988769... |
https://github.com/huggingface/datasets/issues/5961 | I print out n_shards, and under multiple GPUs, this value is always 1.
Is this value correct? | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 17 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.4411472678184509,
-0.5177356004714966,
-0.08604525774717331,
0.26262933015823364,
0.015819620341062546,
-0.052527137100696564,
0.4174102246761322,
0.22069090604782104,
-0.012561824172735214,
0.0985802710056305,
0.07519344985485077,
0.29676541686058044,
0.17903944849967957,
0.30831497907... |
https://github.com/huggingface/datasets/issues/5961 | Yes it's correct, and it explains why you always have the same data passed to your map function (the data can't be split).
But after being passed to `map`, each GPU keeps one example out of n(GPUs) so that you don't end up with duplicate data across GPUs | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 48 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.3965552747249603,
-0.5092836618423462,
-0.10583838820457458,
0.2740623354911804,
0.09118995070457458,
-0.07790995389223099,
0.3442014157772064,
0.22046878933906555,
0.0705917626619339,
0.10969898849725723,
-0.05882649123668671,
0.474252849817276,
0.2462736815214157,
0.3017978072166443,
... |
https://github.com/huggingface/datasets/issues/5961 | > > For each GPU, although see the same data in prepare_data(), the actual training data will not be the same in the end.
> > Is my understanding correct?
>
> Yes exactly :)
>
> > Where can I print the actual training data for each GPU?
>
> You should call print in the data_collator
OK, when printing the t... | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 77 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.286543071269989,
-0.44449901580810547,
-0.09795355796813965,
0.3963581323623657,
0.11219131201505661,
0.07839414477348328,
0.3804986774921417,
0.24164199829101562,
-0.16741898655891418,
0.06556814908981323,
-0.09847146272659302,
0.401059091091156,
0.1971670687198639,
0.37721991539001465... |
https://github.com/huggingface/datasets/issues/5961 | Do we have a solution for this one? Or it's required to get "number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU" | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 33 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.4867711663246155,
-0.4351811110973358,
-0.07999777048826218,
0.20005813241004944,
0.09749133884906769,
-0.0396166667342186,
0.3830694258213043,
0.24505838751792908,
-0.03761538490653038,
0.13198687136173248,
0.10013092309236526,
0.3976881206035614,
0.12254050374031067,
0.302592545747756... |
https://github.com/huggingface/datasets/issues/5961 | For now it's required to have a number of shards that is a factor of the number of GPUs to not have all the workers process the same data (and then skip the right ones to not end up training on duplicate data).
It would be quite complex to implement a strategy that would utilize all the GPUs with an arbitrary number... | IterableDataset: split by node and map may preprocess samples that will be skipped anyway | There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the data and at the end keeps 1 sample out of n(GPUs) - skipping the others.
In case 2. it's ... | 70 | IterableDataset: split by node and map may preprocess samples that will be skipped anyway
There are two ways an iterable dataset can be split by node:
1. if the number of shards is a factor of number of GPUs: in that case the shards are evenly distributed per GPU
2. otherwise, each GPU iterate on the ... | [
-0.4646584987640381,
-0.3735218346118927,
-0.0920911431312561,
0.15261659026145935,
0.11911855638027191,
-0.04775995761156082,
0.4881906807422638,
0.25509804487228394,
-0.039605945348739624,
0.1438959836959839,
0.09474242478609085,
0.3893817067146301,
0.10234491527080536,
0.329986870288848... |
https://github.com/huggingface/datasets/issues/5955 | This is the actual error:
```
Failed to read file '/home/lakala/hjc/code/pycode/glm/temp.json' with error <class 'pyarrow.lib.ArrowInvalid'>: cannot mix list and non-list, non-null values
```
Which means some samples are incorrectly formatted.
PyArrow, a storage backend that we use under the hood, requires that ... | Strange bug in loading local JSON files, using load_dataset | ### Describe the bug
I am using 'load_dataset 'loads a JSON file, but I found a strange bug: an error will be reported when the length of the JSON file exceeds 160000 (uncertain exact number). I have checked the data through the following code and there are no issues. So I cannot determine the true reason for this err... | 84 | Strange bug in loading local JSON files, using load_dataset
### Describe the bug
I am using 'load_dataset 'loads a JSON file, but I found a strange bug: an error will be reported when the length of the JSON file exceeds 160000 (uncertain exact number). I have checked the data through the following code and there are... | [
-0.1079191341996193,
-0.09877710044384003,
-0.04527151584625244,
0.54625403881073,
0.23800578713417053,
0.10583063960075378,
0.32995471358299255,
0.41471725702285767,
0.433463454246521,
0.18027374148368835,
0.30697494745254517,
0.3544989824295044,
0.08857817947864532,
-0.13640546798706055,... |
https://github.com/huggingface/datasets/issues/5955 | @mariosasko
I used the same operation to check the original data before and after slicing.
This is reflected in my code.
160000 is not a specific number.
I can also get output using 150000.
This doesn't seem to align very well with what you said.
Because if only some sample formats are incorrect.
So there shoul... | Strange bug in loading local JSON files, using load_dataset | ### Describe the bug
I am using 'load_dataset 'loads a JSON file, but I found a strange bug: an error will be reported when the length of the JSON file exceeds 160000 (uncertain exact number). I have checked the data through the following code and there are no issues. So I cannot determine the true reason for this err... | 72 | Strange bug in loading local JSON files, using load_dataset
### Describe the bug
I am using 'load_dataset 'loads a JSON file, but I found a strange bug: an error will be reported when the length of the JSON file exceeds 160000 (uncertain exact number). I have checked the data through the following code and there are... | [
-0.1079191341996193,
-0.09877710044384003,
-0.04527151584625244,
0.54625403881073,
0.23800578713417053,
0.10583063960075378,
0.32995471358299255,
0.41471725702285767,
0.433463454246521,
0.18027374148368835,
0.30697494745254517,
0.3544989824295044,
0.08857817947864532,
-0.13640546798706055,... |
https://github.com/huggingface/datasets/issues/5955 | Our JSON loader does the following in your case:
```python
import json
import pyarrow as pa
with open(file, encoding="utf-8") as f:
dataset = json.load(f)
keys = set().union(*[row.keys() for row in dataset])
mapping = {col: [row.get(col) for row in dataset] for col in keys}
pa_table = pa.Table.from_pydi... | Strange bug in loading local JSON files, using load_dataset | ### Describe the bug
I am using 'load_dataset 'loads a JSON file, but I found a strange bug: an error will be reported when the length of the JSON file exceeds 160000 (uncertain exact number). I have checked the data through the following code and there are no issues. So I cannot determine the true reason for this err... | 156 | Strange bug in loading local JSON files, using load_dataset
### Describe the bug
I am using 'load_dataset 'loads a JSON file, but I found a strange bug: an error will be reported when the length of the JSON file exceeds 160000 (uncertain exact number). I have checked the data through the following code and there are... | [
-0.1079191341996193,
-0.09877710044384003,
-0.04527151584625244,
0.54625403881073,
0.23800578713417053,
0.10583063960075378,
0.32995471358299255,
0.41471725702285767,
0.433463454246521,
0.18027374148368835,
0.30697494745254517,
0.3544989824295044,
0.08857817947864532,
-0.13640546798706055,... |
https://github.com/huggingface/datasets/issues/5953 | cc @sanchit-gandhi @Vaibhavs10 @lhoestq - this is mainly for demos that use Common Voice datasets as done here: https://github.com/facebookresearch/fairseq/tree/main/examples/mms#-transformers
| Bad error message when trying to download gated dataset | ### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make sure you specified the correct `repo_id` and `repo_type`.
I... | 19 | Bad error message when trying to download gated dataset
### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make su... | [
-0.2910834550857544,
-0.24233008921146393,
0.11376997828483582,
0.28113362193107605,
0.4399706721305847,
0.1367226243019104,
0.28315269947052,
0.18023063242435455,
0.02009575068950653,
0.15855589509010315,
-0.41839614510536194,
-0.16160747408866882,
0.03821331635117531,
0.06896946579217911... |
https://github.com/huggingface/datasets/issues/5953 | Hi ! the error for me is
```
FileNotFoundError: Couldn't find a dataset script at /content/mozilla-foundation/common_voice_13_0/common_voice_13_0.py or any data file in the same directory. Couldn't find 'mozilla-foundation/common_voice_13_0' on the Hugging Face Hub either: FileNotFoundError: Dataset 'mozilla-founda... | Bad error message when trying to download gated dataset | ### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make sure you specified the correct `repo_id` and `repo_type`.
I... | 75 | Bad error message when trying to download gated dataset
### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make su... | [
-0.2910834550857544,
-0.24233008921146393,
0.11376997828483582,
0.28113362193107605,
0.4399706721305847,
0.1367226243019104,
0.28315269947052,
0.18023063242435455,
0.02009575068950653,
0.15855589509010315,
-0.41839614510536194,
-0.16160747408866882,
0.03821331635117531,
0.06896946579217911... |
https://github.com/huggingface/datasets/issues/5953 | Okay, I am able to reproduce @patrickvonplaten's original error: https://github.com/Vaibhavs10/scratchpad/blob/main/cv13_datasets_test.ipynb
Also not sure why it looks for `n_shards.json` | Bad error message when trying to download gated dataset | ### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make sure you specified the correct `repo_id` and `repo_type`.
I... | 18 | Bad error message when trying to download gated dataset
### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make su... | [
-0.2910834550857544,
-0.24233008921146393,
0.11376997828483582,
0.28113362193107605,
0.4399706721305847,
0.1367226243019104,
0.28315269947052,
0.18023063242435455,
0.02009575068950653,
0.15855589509010315,
-0.41839614510536194,
-0.16160747408866882,
0.03821331635117531,
0.06896946579217911... |
https://github.com/huggingface/datasets/issues/5953 | Ok I see, this file is downloaded from the CV dataset script - let me investigate | Bad error message when trying to download gated dataset | ### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make sure you specified the correct `repo_id` and `repo_type`.
I... | 16 | Bad error message when trying to download gated dataset
### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make su... | [
-0.2910834550857544,
-0.24233008921146393,
0.11376997828483582,
0.28113362193107605,
0.4399706721305847,
0.1367226243019104,
0.28315269947052,
0.18023063242435455,
0.02009575068950653,
0.15855589509010315,
-0.41839614510536194,
-0.16160747408866882,
0.03821331635117531,
0.06896946579217911... |
https://github.com/huggingface/datasets/issues/5953 | Ok I see: when you log out you no longer have access to the repository.
Therefore the dataset script is loaded from cache:
```
WARNING:datasets.load:Using the latest cached version of the module from /root/.cache/huggingface/modules/datasets_modules/datasets/mozilla-foundation--common_voice_13_0/22809012aac1fc9803... | Bad error message when trying to download gated dataset | ### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make sure you specified the correct `repo_id` and `repo_type`.
I... | 68 | Bad error message when trying to download gated dataset
### Describe the bug
When I attempt to download a model from the Hub that is gated without being logged in, I get a nice error message. E.g.:
E.g.
```sh
Repository Not Found for url: https://huggingface.co/api/models/DeepFloyd/IF-I-XL-v1.0.
Please make su... | [
-0.2910834550857544,
-0.24233008921146393,
0.11376997828483582,
0.28113362193107605,
0.4399706721305847,
0.1367226243019104,
0.28315269947052,
0.18023063242435455,
0.02009575068950653,
0.15855589509010315,
-0.41839614510536194,
-0.16160747408866882,
0.03821331635117531,
0.06896946579217911... |
https://github.com/huggingface/datasets/issues/5950 | Hi ! We use the Arrow columnar format under the hood, which doesn't support such dictionaries: each field must have a fixed type and exist in each sample.
Instead you can restructure your data like
```
{
"index": 0,
"keys": ["2 * x + y >= 3"],
"values": [["2 * x + y >= 3", "4 * x + 2 * y >= 6"]],
... | Support for data with instance-wise dictionary as features | ### Feature request
I notice that when loading data instances with feature type of python dictionary, the dictionary keys would be broadcast so that every instance has the same set of keys. Please see an example in the Motivation section.
It is possible to avoid this behavior, i.e., load dictionary features as it i... | 87 | Support for data with instance-wise dictionary as features
### Feature request
I notice that when loading data instances with feature type of python dictionary, the dictionary keys would be broadcast so that every instance has the same set of keys. Please see an example in the Motivation section.
It is possible t... | [
0.19504207372665405,
-0.34095388650894165,
0.002453260123729706,
0.29262271523475647,
0.2256864458322525,
-0.06849919259548187,
0.2951243817806244,
0.02916017547249794,
0.14267361164093018,
0.2032242715358734,
0.21794208884239197,
0.4819650948047638,
-0.3849047124385834,
0.7559517025947571... |
https://github.com/huggingface/datasets/issues/5947 | Hi ! The audio data don't always exist as files on disk - the blobs are often stored in the Arrow files. For now I'd suggest disabling decoding with `.cast_column("audio", Audio(decode=False))` and apply your own decoding that handles corrupted files (maybe to filter them out ?)
cc @sanchit-gandhi since it's related... | Return the audio filename when decoding fails due to corrupt files | ### Feature request
Return the audio filename when the audio decoding fails. Although currently there are some checks for mp3 and opus formats with the library version there are still cases when the audio decoding could fail, eg. Corrupt file.
### Motivation
When you try to load an object file dataset and the... | 71 | Return the audio filename when decoding fails due to corrupt files
### Feature request
Return the audio filename when the audio decoding fails. Although currently there are some checks for mp3 and opus formats with the library version there are still cases when the audio decoding could fail, eg. Corrupt file.
... | [
-0.32508015632629395,
0.2062225639820099,
-0.10531511902809143,
0.2657145857810974,
0.12691596150398254,
-0.3168007433414459,
-0.04729864373803139,
0.3888251781463623,
-0.3635145425796509,
0.6399989724159241,
-0.183310866355896,
0.47412925958633423,
-0.3171817362308502,
-0.223241925239563,... |
https://github.com/huggingface/datasets/issues/5947 | Thanks @lhoestq, I wasn't aware of the decode flag. It makes more sense as you say to show a warning when there are corrupted files together with some metadata of the file that allows to filter them from the dataset.
My workaround was to catch the LibsndfileError and generate a dummy audio with an unsual sample rate... | Return the audio filename when decoding fails due to corrupt files | ### Feature request
Return the audio filename when the audio decoding fails. Although currently there are some checks for mp3 and opus formats with the library version there are still cases when the audio decoding could fail, eg. Corrupt file.
### Motivation
When you try to load an object file dataset and the... | 81 | Return the audio filename when decoding fails due to corrupt files
### Feature request
Return the audio filename when the audio decoding fails. Although currently there are some checks for mp3 and opus formats with the library version there are still cases when the audio decoding could fail, eg. Corrupt file.
... | [
-0.2349940538406372,
0.25069093704223633,
-0.08755676448345184,
0.023980338126420975,
0.0556185282766819,
-0.2654074430465698,
-0.08733473718166351,
0.4549439549446106,
-0.26085078716278076,
0.6982671022415161,
-0.07927210628986359,
0.5527421236038208,
-0.3463272452354431,
-0.0602911263704... |
https://github.com/huggingface/datasets/issues/5946 | > Looks related to https://discuss.huggingface.co/t/indexerror-invalid-key-16-is-out-of-bounds-for-size-0/14298/4?u=lhoestq
The problem has not been solved, I have tried this before, but the problem is the same | IndexError Not Solving -> IndexError: Invalid key: ?? is out of bounds for size 0 or ?? | ### Describe the bug
in <cell line: 1>:1 │
│ │
│ /usr/local/lib/python3.10/dist-packages/transformers/trainer.py:1537 in train ... | 22 | IndexError Not Solving -> IndexError: Invalid key: ?? is out of bounds for size 0 or ??
### Describe the bug
in <cell line: 1>:1 │
│ │
│ /usr... | [
-0.18714331090450287,
-0.29146265983581543,
-0.20474481582641602,
0.2455594688653946,
0.14909164607524872,
-0.0931643545627594,
0.32553473114967346,
0.5325318574905396,
-0.18145307898521423,
0.26250627636909485,
-0.03803275153040886,
0.2623401880264282,
-0.2751728594303131,
0.0846320539712... |
https://github.com/huggingface/datasets/issues/5946 | data = data["train"].shuffle().map(generate_and_tokenize_prompt, batched = False) # change this line to -
data["train"] = data["train"].shuffle().map(generate_and_tokenize_prompt, batched = False)
After doing this change you code should run fine. | IndexError Not Solving -> IndexError: Invalid key: ?? is out of bounds for size 0 or ?? | ### Describe the bug
in <cell line: 1>:1 │
│ │
│ /usr/local/lib/python3.10/dist-packages/transformers/trainer.py:1537 in train ... | 27 | IndexError Not Solving -> IndexError: Invalid key: ?? is out of bounds for size 0 or ??
### Describe the bug
in <cell line: 1>:1 │
│ │
│ /usr... | [
-0.18714331090450287,
-0.29146265983581543,
-0.20474481582641602,
0.2455594688653946,
0.14909164607524872,
-0.0931643545627594,
0.32553473114967346,
0.5325318574905396,
-0.18145307898521423,
0.26250627636909485,
-0.03803275153040886,
0.2623401880264282,
-0.2751728594303131,
0.0846320539712... |
https://github.com/huggingface/datasets/issues/5946 | > >
>
> @syngokhan did u solve it? I am desperate
refer to my earlier comment. you will find the solution. | IndexError Not Solving -> IndexError: Invalid key: ?? is out of bounds for size 0 or ?? | ### Describe the bug
in <cell line: 1>:1 │
│ │
│ /usr/local/lib/python3.10/dist-packages/transformers/trainer.py:1537 in train ... | 22 | IndexError Not Solving -> IndexError: Invalid key: ?? is out of bounds for size 0 or ??
### Describe the bug
in <cell line: 1>:1 │
│ │
│ /usr... | [
-0.18714331090450287,
-0.29146265983581543,
-0.20474481582641602,
0.2455594688653946,
0.14909164607524872,
-0.0931643545627594,
0.32553473114967346,
0.5325318574905396,
-0.18145307898521423,
0.26250627636909485,
-0.03803275153040886,
0.2623401880264282,
-0.2751728594303131,
0.0846320539712... |
https://github.com/huggingface/datasets/issues/5945 | Hi ! Feel free to re-run your code later, it will resume automatically where you left | Failing to upload dataset to the hub | ### Describe the bug
Trying to upload a dataset of hundreds of thousands of audio samples (the total volume is not very large, 60 gb) to the hub with push_to_hub, it doesn't work.
From time to time one piece of the data (parquet) gets pushed and then I get RemoteDisconnected even though my internet is stable.
Please... | 16 | Failing to upload dataset to the hub
### Describe the bug
Trying to upload a dataset of hundreds of thousands of audio samples (the total volume is not very large, 60 gb) to the hub with push_to_hub, it doesn't work.
From time to time one piece of the data (parquet) gets pushed and then I get RemoteDisconnected eve... | [
-0.3492951989173889,
-0.43613675236701965,
0.011495642364025116,
0.13727526366710663,
0.14236441254615784,
-0.1314420849084854,
0.1631830632686615,
0.24041369557380676,
-0.09034931659698486,
0.07815927267074585,
0.15260639786720276,
0.2544565498828888,
-0.19957593083381653,
0.1263261735439... |
https://github.com/huggingface/datasets/issues/5945 | Alternatively you can save your dataset in parquet files locally and upload them to the hub manually
```python
from tqdm import tqdm
num_shards = 60
for index in tqdm(range(num_shards)):
ds.shard(num_shards=num_shards, index=index, contiguous=True).to_parquet(f"{index:05d}.parquet")
```` | Failing to upload dataset to the hub | ### Describe the bug
Trying to upload a dataset of hundreds of thousands of audio samples (the total volume is not very large, 60 gb) to the hub with push_to_hub, it doesn't work.
From time to time one piece of the data (parquet) gets pushed and then I get RemoteDisconnected even though my internet is stable.
Please... | 33 | Failing to upload dataset to the hub
### Describe the bug
Trying to upload a dataset of hundreds of thousands of audio samples (the total volume is not very large, 60 gb) to the hub with push_to_hub, it doesn't work.
From time to time one piece of the data (parquet) gets pushed and then I get RemoteDisconnected eve... | [
-0.3584897518157959,
-0.4968307316303253,
0.041109416633844376,
0.15697543323040009,
0.1653338372707367,
-0.17984625697135925,
0.18572813272476196,
0.24879494309425354,
-0.26125916838645935,
0.13924777507781982,
0.08525009453296661,
0.33221179246902466,
-0.21793578565120697,
0.122852161526... |
https://github.com/huggingface/datasets/issues/5941 | already did，but not useful for step Generating train split，it works in step "Resolving data files" & "Downloading data files" | Load Data Sets Too Slow In Train Seq2seq Model | ### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ run_speech_recognition_seq2seq.py](https://github... | 19 | Load Data Sets Too Slow In Train Seq2seq Model
### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ ... | [
-0.4304403066635132,
-0.21508798003196716,
-0.0036190710961818695,
0.40801307559013367,
0.20815572142601013,
0.016328558325767517,
0.27846330404281616,
0.553817093372345,
0.13296890258789062,
-0.04803692549467087,
-0.3303144574165344,
0.2580045461654663,
-0.25866442918777466,
0.17247849702... |
https://github.com/huggingface/datasets/issues/5941 | We need more info about the issue to provide help.
Can you interrupt the process (with `num_proc=None`) after the `load_dataset` call when the slowdown occurs? So we can know what part of the code is causing it.
The `audiofolder` \ `imagefolder` with metadata is not performant for large datasets. Luckily, we can... | Load Data Sets Too Slow In Train Seq2seq Model | ### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ run_speech_recognition_seq2seq.py](https://github... | 101 | Load Data Sets Too Slow In Train Seq2seq Model
### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ ... | [
-0.4304403066635132,
-0.21508798003196716,
-0.0036190710961818695,
0.40801307559013367,
0.20815572142601013,
0.016328558325767517,
0.27846330404281616,
0.553817093372345,
0.13296890258789062,
-0.04803692549467087,
-0.3303144574165344,
0.2580045461654663,
-0.25866442918777466,
0.17247849702... |
https://github.com/huggingface/datasets/issues/5941 | Can you interrupt the process (with num_proc=None) after the load_dataset call when the slowdown occurs? So we can know what part of the code is causing it.
（I'll try this operation）
The audiofolder \ imagefolder with metadata is not performant for large datasets. Luckily, we can make them much faster if drop the nes... | Load Data Sets Too Slow In Train Seq2seq Model | ### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ run_speech_recognition_seq2seq.py](https://github... | 140 | Load Data Sets Too Slow In Train Seq2seq Model
### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ ... | [
-0.4304403066635132,
-0.21508798003196716,
-0.0036190710961818695,
0.40801307559013367,
0.20815572142601013,
0.016328558325767517,
0.27846330404281616,
0.553817093372345,
0.13296890258789062,
-0.04803692549467087,
-0.3303144574165344,
0.2580045461654663,
-0.25866442918777466,
0.17247849702... |
https://github.com/huggingface/datasets/issues/5941 | Can you interrupt the process (with num_proc=None) after the load_dataset call when the slowdown occurs? So we can know what part of the code is causing it.
================================================================================
Here is the log：
[load_dataset.log](https://github.com/huggingface/datasets/fil... | Load Data Sets Too Slow In Train Seq2seq Model | ### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ run_speech_recognition_seq2seq.py](https://github... | 43 | Load Data Sets Too Slow In Train Seq2seq Model
### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ ... | [
-0.4304403066635132,
-0.21508798003196716,
-0.0036190710961818695,
0.40801307559013367,
0.20815572142601013,
0.016328558325767517,
0.27846330404281616,
0.553817093372345,
0.13296890258789062,
-0.04803692549467087,
-0.3303144574165344,
0.2580045461654663,
-0.25866442918777466,
0.17247849702... |
https://github.com/huggingface/datasets/issues/5941 | In the meantime, it's better to use Dataset.from_generator (requires replacing the load_dataset calls in the transformers script with Dataset.from_generator) or write a dataset loading script for large datasets.
================================================================================
I tried ‘Dataset. from_ge... | Load Data Sets Too Slow In Train Seq2seq Model | ### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ run_speech_recognition_seq2seq.py](https://github... | 43 | Load Data Sets Too Slow In Train Seq2seq Model
### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ ... | [
-0.4304403066635132,
-0.21508798003196716,
-0.0036190710961818695,
0.40801307559013367,
0.20815572142601013,
0.016328558325767517,
0.27846330404281616,
0.553817093372345,
0.13296890258789062,
-0.04803692549467087,
-0.3303144574165344,
0.2580045461654663,
-0.25866442918777466,
0.17247849702... |
https://github.com/huggingface/datasets/issues/5941 | I have already solved this problem, referring to #5990 : read audio frist, then use data_generator to change format . | Load Data Sets Too Slow In Train Seq2seq Model | ### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ run_speech_recognition_seq2seq.py](https://github... | 20 | Load Data Sets Too Slow In Train Seq2seq Model
### Describe the bug
step 'Generating train split' in load_dataset is too slow：
### Steps to reproduce the bug
Data： own data，16K16B Mono wav
Oficial Script:[ ... | [
-0.4304403066635132,
-0.21508798003196716,
-0.0036190710961818695,
0.40801307559013367,
0.20815572142601013,
0.016328558325767517,
0.27846330404281616,
0.553817093372345,
0.13296890258789062,
-0.04803692549467087,
-0.3303144574165344,
0.2580045461654663,
-0.25866442918777466,
0.17247849702... |
https://github.com/huggingface/datasets/issues/5990 | Hi @AntreasAntoniou , sorry to know you are facing this issue. To help debugging it, could you tell me:
- What is the total dataset size?
- Is it always failing on the same shard or is the hanging problem happening randomly?
- Were you able to save the dataset as parquet locally? This would help us determine if the ... | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 81 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | One trick that can also help is to check the traceback when you kill your python process: it will show where in the code it was hanging | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 27 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | Right. So I did the trick @lhoestq suggested. Here is where things seem to hang
```
Error while uploading 'data/train-00120-of-00195-466c2dbab2eb9989.parquet' to the Hub.
Pushing split train to the Hub. ... | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 556 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | @Wauplin
>What is the total dataset size?
There are three variants, and the random hanging happens on all three. The sizes are 2TB, 1TB, and 200GB.
>Is it always failing on the same shard or is the hanging problem happening randomly?
It seems to be very much random, as restarting can help move past the pr... | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 101 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | Hmm it looks like an issue with TQDM lock. Maybe you can try updating TQDM ? | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 16 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | I am using the latest version of tqdm
```
⬢ [Docker] ❯ pip install tqdm --upgrade
Requirement already satisfied: tqdm in /opt/conda/envs/main/lib/python3.10/site-packages (4.65.0)
WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It... | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 54 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | I tried trying to catch the hanging issue in action again
```
Pushing dataset shards to the dataset hub: 65%|█████████████████████████████████████████████████████████████████▊ | 127/195 [2:28:02<1:19:15, 69.94s/it]
Error while upload... | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 867 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | I'm on my phone so can't help that much. What I'd advice to do is to [save_to_disk](https://huggingface.co/docs/datasets/package_reference/main_classes#save_to_disk) if it's not already done and then upload the files/folder to the Hub separately. You can find what you need in the [upload guide](https://huggingface.co/d... | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 58 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | In your last stacktrace it interrupted while embedding external content - in case your dataset in made of images or audio files that live on your disk. Is it the case ? | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 32 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | It's maybe related to https://github.com/apache/arrow/issues/34455: are you using ArrayND features ?
Also what's your `pyarrow` version ? Could you try updating to >= 12.0.1 ? | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 25 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | I was using pyarrow == 12.0.0
I am not explicitly using ArrayND features, unless the hub API automatically converts my files to such. | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 23 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | You can also try to reduce the `max_shard_size` - Sometimes parquet has a hard time working with data bigger than 2GB | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 21 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | So, updating the pyarrow seems to help. It can still throw errors here and there but I can retry when that happens. It's better than hanging.
However, I am a bit confused about something. I have uploaded my datasets, but while earlier I could see all three sets, now I can only see 1. What's going on?
https://hugg... | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 83 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
https://github.com/huggingface/datasets/issues/5990 | It's a bug on our side, I'll update the dataset viewer ;)
Thanks for reporting ! | Pushing a large dataset on the hub consistently hangs | ### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over the past week, and despite my best efforts to try and catc... | 16 | Pushing a large dataset on the hub consistently hangs
### Describe the bug
Once I have locally built a large dataset that I want to push to hub, I use the recommended approach of .push_to_hub to get the dataset on the hub, and after pushing a few shards, it consistently hangs. This has happened over 40 times over th... | [
-0.3251087963581085,
-0.42933112382888794,
-0.06096188724040985,
0.08869528025388718,
0.21317243576049805,
-0.06696006655693054,
0.1970674693584442,
0.5384528636932373,
0.05225852131843567,
0.3061686158180237,
0.08448541909456253,
0.10903309285640717,
0.03724048659205437,
0.220244064927101... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.