html_url stringlengths 51 51 | title stringlengths 9 280 | comments stringlengths 63 24.7k | body stringlengths 51 58.6k | comments length int64 16 1.45k | text stringlengths 186 59.1k |
|---|---|---|---|---|---|
https://github.com/huggingface/datasets/issues/6360 | Add support for `Sequence(Audio/Image)` feature in `push_to_hub` | This issue stems from https://github.com/huggingface/datasets/blob/6d2f2a5e0fea3827eccfd1717d8021c15fc4292a/src/datasets/table.py#L2203-L2205
I'll address it as part of https://github.com/huggingface/datasets/pull/6283.
In the meantime, this should work
```python
import pyarrow as pa
from datasets import Ima... | ### Feature request
Allow for `Sequence` of `Image` (or `Audio`) to be embedded inside the shards.
### Motivation
Currently, thanks to #3685, when `embed_external_files` is set to True (which is the default) in `push_to_hub`, features of type `Image` and `Audio` are embedded inside the arrow/parquet shards, instead ... | 58 | Add support for `Sequence(Audio/Image)` feature in `push_to_hub`
### Feature request
Allow for `Sequence` of `Image` (or `Audio`) to be embedded inside the shards.
### Motivation
Currently, thanks to #3685, when `embed_external_files` is set to True (which is the default) in `push_to_hub`, features of type `Image` ... |
https://github.com/huggingface/datasets/issues/5433 | Support latest Docker image in CI benchmarks | Sorry, it was us:[^1] https://github.com/iterative/cml/pull/1317 & https://github.com/iterative/cml/issues/1319#issuecomment-1385599559; should be fixed with [v0.18.17](https://github.com/iterative/cml/releases/tag/v0.18.17).
[^1]: More or less, see https://github.com/yargs/yargs/issues/873. | Once we find out the root cause of:
- #5431
we should revert the temporary pin on the Docker image version introduced by:
- #5432 | 18 | Support latest Docker image in CI benchmarks
Once we find out the root cause of:
- #5431
we should revert the temporary pin on the Docker image version introduced by:
- #5432
Sorry, it was us:[^1] https://github.com/iterative/cml/pull/1317 & https://github.com/iterative/cml/issues/1319#issuecomment-1385599559; sh... |
https://github.com/huggingface/datasets/issues/5433 | Support latest Docker image in CI benchmarks | Hi @0x2b3bfa0, thanks a lot for the investigation, the context about the the root cause and for fixing it!!
We are reviewing your PR to unpin the container image. | Once we find out the root cause of:
- #5431
we should revert the temporary pin on the Docker image version introduced by:
- #5432 | 29 | Support latest Docker image in CI benchmarks
Once we find out the root cause of:
- #5431
we should revert the temporary pin on the Docker image version introduced by:
- #5432
Hi @0x2b3bfa0, thanks a lot for the investigation, the context about the the root cause and for fixing it!!
We are reviewing your PR to ... |
https://github.com/huggingface/datasets/issues/5910 | Cannot use both set_format and set_transform | Currently, it's not possible to chain `set_format`/`set_transform` calls (plus, this is a breaking change if we decide to implement it), so I see two possible solutions:
* using `set_format`/`set_transform` for the 1st transform and then passing the transformed example/batch to the 2nd transform
* implementing and re... | ### Describe the bug
I need to process some data using the set_transform method but I also need the data to be formatted for pytorch before processing it.
I don't see anywhere in the documentation something that says that both methods cannot be used at the same time.
### Steps to reproduce the bug
```
from... | 69 | Cannot use both set_format and set_transform
### Describe the bug
I need to process some data using the set_transform method but I also need the data to be formatted for pytorch before processing it.
I don't see anywhere in the documentation something that says that both methods cannot be used at the same time.
... |
https://github.com/huggingface/datasets/issues/5910 | Cannot use both set_format and set_transform | Hey Mario,
Thanks, for getting back to me. the toDouble was just an example my real life case requires many more transforms.
What do you mean by:
> using set_format/set_transform for the 1st transform and then passing the transformed example/batch to the 2nd transform
How would that go, I thought you can't chai... | ### Describe the bug
I need to process some data using the set_transform method but I also need the data to be formatted for pytorch before processing it.
I don't see anywhere in the documentation something that says that both methods cannot be used at the same time.
### Steps to reproduce the bug
```
from... | 86 | Cannot use both set_format and set_transform
### Describe the bug
I need to process some data using the set_transform method but I also need the data to be formatted for pytorch before processing it.
I don't see anywhere in the documentation something that says that both methods cannot be used at the same time.
... |
https://github.com/huggingface/datasets/issues/5910 | Cannot use both set_format and set_transform | > How would that go, I thought you can't chain them?
Yes, they cannot be chained. This is what I meant:
```python
ds.set_transform(first_transform)
# calling the 2nd transform on each accessed batch
second_transform(ds[2:3])
```
> As for the custom formatter, is it possible to reference an existing formatter... | ### Describe the bug
I need to process some data using the set_transform method but I also need the data to be formatted for pytorch before processing it.
I don't see anywhere in the documentation something that says that both methods cannot be used at the same time.
### Steps to reproduce the bug
```
from... | 74 | Cannot use both set_format and set_transform
### Describe the bug
I need to process some data using the set_transform method but I also need the data to be formatted for pytorch before processing it.
I don't see anywhere in the documentation something that says that both methods cannot be used at the same time.
... |
https://github.com/huggingface/datasets/issues/3918 | datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files | Hi @willowdong! These issues were fixed on master. We will have a new release of `datasets` later today. In the meantime, you can avoid these issues by installing `datasets` from master as follows:
```bash
pip install git+https://github.com/huggingface/datasets.git
``` | ## Describe the bug
Can't load the dataset
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
```
from datasets import load_dataset
dataset = load_dataset('multi_news')
dataset_2=load_dataset("reddit_tifu", "long")
## Actual results
raise NonMatchingChecksumError(error_msg + s... | 38 | datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files
## Describe the bug
Can't load the dataset
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
```
from datasets import load_dataset
dataset = load_dataset('multi_news')
dataset_2=load_... |
https://github.com/huggingface/datasets/issues/7720 | Datasets 4.0 map function causing column not found | Hi, I tried to reproduce this issue on the latest `main` branch but it seems to be working correctly now. My test script (which creates a dummy dataset and applies the `.map()` function) successfully creates and accesses the new column without a `KeyError`.
It's possible this was fixed by a recent commit. The maintain... | ### Describe the bug
Column returned after mapping is not found in new instance of the dataset.
### Steps to reproduce the bug
Code for reproduction. After running get_total_audio_length, it is errored out due to `data` not having `duration`
```
def compute_duration(x):
return {"duration": len(x["audio"]["array"... | 61 | Datasets 4.0 map function causing column not found
### Describe the bug
Column returned after mapping is not found in new instance of the dataset.
### Steps to reproduce the bug
Code for reproduction. After running get_total_audio_length, it is errored out due to `data` not having `duration`
```
def compute_duration... |
https://github.com/huggingface/datasets/issues/7720 | Datasets 4.0 map function causing column not found | Hi, have you tried on a large dataset (200GB+) perhaps? I will try my best to do a rerun with main branch when I have the time. | ### Describe the bug
Column returned after mapping is not found in new instance of the dataset.
### Steps to reproduce the bug
Code for reproduction. After running get_total_audio_length, it is errored out due to `data` not having `duration`
```
def compute_duration(x):
return {"duration": len(x["audio"]["array"... | 27 | Datasets 4.0 map function causing column not found
### Describe the bug
Column returned after mapping is not found in new instance of the dataset.
### Steps to reproduce the bug
Code for reproduction. After running get_total_audio_length, it is errored out due to `data` not having `duration`
```
def compute_duration... |
https://github.com/huggingface/datasets/issues/7720 | Datasets 4.0 map function causing column not found | I ran it on a small dataset, maybe that’s why I didn’t hit the issue. If it still shows up on your side with the latest main, let me know. I can try it on a bigger set too. | ### Describe the bug
Column returned after mapping is not found in new instance of the dataset.
### Steps to reproduce the bug
Code for reproduction. After running get_total_audio_length, it is errored out due to `data` not having `duration`
```
def compute_duration(x):
return {"duration": len(x["audio"]["array"... | 39 | Datasets 4.0 map function causing column not found
### Describe the bug
Column returned after mapping is not found in new instance of the dataset.
### Steps to reproduce the bug
Code for reproduction. After running get_total_audio_length, it is errored out due to `data` not having `duration`
```
def compute_duration... |
https://github.com/huggingface/datasets/issues/5797 | load_dataset is case sentitive? | Hi @haonan-li , thank you for the report! It seems to be a bug on the [`huggingface_hub`](https://github.com/huggingface/huggingface_hub) site, there is even no such dataset as `mbzuai/bactrian-x` on the Hub. I opened and [issue](https://github.com/huggingface/huggingface_hub/issues/1453) there. | ### Describe the bug
load_dataset() function is case sensitive?
### Steps to reproduce the bug
The following two code, get totally different behavior.
1. load_dataset('mbzuai/bactrian-x','en')
2. load_dataset('MBZUAI/Bactrian-X','en')
### Expected behavior
Compare 1 and 2.
1 will download all 52 subsets, sh... | 34 | load_dataset is case sentitive?
### Describe the bug
load_dataset() function is case sensitive?
### Steps to reproduce the bug
The following two code, get totally different behavior.
1. load_dataset('mbzuai/bactrian-x','en')
2. load_dataset('MBZUAI/Bactrian-X','en')
### Expected behavior
Compare 1 and 2.
1 ... |
https://github.com/huggingface/datasets/issues/5797 | load_dataset is case sentitive? | I think `load_dataset("mbzuai/bactrian-x")` shouldn't be loaded at all and raise an error but because of [this fallback](https://github.com/huggingface/datasets/blob/main/src/datasets/load.py#L1194) to packaged loaders when no other options are applicable, it loads the dataset with standard `json` loader instead of the... | ### Describe the bug
load_dataset() function is case sensitive?
### Steps to reproduce the bug
The following two code, get totally different behavior.
1. load_dataset('mbzuai/bactrian-x','en')
2. load_dataset('MBZUAI/Bactrian-X','en')
### Expected behavior
Compare 1 and 2.
1 will download all 52 subsets, sh... | 40 | load_dataset is case sentitive?
### Describe the bug
load_dataset() function is case sensitive?
### Steps to reproduce the bug
The following two code, get totally different behavior.
1. load_dataset('mbzuai/bactrian-x','en')
2. load_dataset('MBZUAI/Bactrian-X','en')
### Expected behavior
Compare 1 and 2.
1 ... |
https://github.com/huggingface/datasets/issues/6267 | Multi label class encoding | You can use a `Sequence(ClassLabel(...))` feature type to represent a list of labels, and `cast_column`/`cast` to perform the "string to label" conversion (`class_encode_column` does support nested fields), e.g., in your case:
```python
from datasets import Dataset, Sequence, ClassLabel
data = {
'text': ['one'... | ### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like to encode:
```
data = {
... | 66 | Multi label class encoding
### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like t... |
https://github.com/huggingface/datasets/issues/6267 | Multi label class encoding | Great! Can you elaborate on "class_encode_column does support nested fields"? Do you mean that there is a way to `class_encode_column` on a Sequence? | ### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like to encode:
```
data = {
... | 23 | Multi label class encoding
### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like t... |
https://github.com/huggingface/datasets/issues/6267 | Multi label class encoding | Sorry, I'm still not following. Are you saying that there currently exists a way to call `class_encode_column` on a `Sequence(ClassLabel)` type? Or that the underlying data structures support it and a contribution of a method to do that would be welcome? | ### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like to encode:
```
data = {
... | 41 | Multi label class encoding
### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like t... |
https://github.com/huggingface/datasets/issues/6267 | Multi label class encoding | `class_encode_column ` currently does not support `Sequence(ClassLabel)`. Implementing support for this would be a nice contribution.
In the meantime, this limitation can be circumvented by fetching (unique) labels and calling `.cast_column(col, Sequence(ClassLabel(names=labels)))`. | ### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like to encode:
```
data = {
... | 32 | Multi label class encoding
### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like t... |
https://github.com/huggingface/datasets/issues/6267 | Multi label class encoding | Ok makes sense, can you take a look at the POC implementation I did [here](https://github.com/huggingface/datasets/commit/15443098e9ce053943172f7ec6fce3769d7dff6e)? Happy to take another pass / submit as a PR but would be helpful if I got a thumbs up that this was directionally correct with respect to implementation /... | ### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like to encode:
```
data = {
... | 46 | Multi label class encoding
### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like t... |
https://github.com/huggingface/datasets/issues/6267 | Multi label class encoding | There is no need to introduce a new type (`MultiLabel`) for this feature. Also, I think we can keep the logic inside a single method instead of separating the two cases.
Maybe https://github.com/huggingface/datasets/pull/4277 can help with the implementation. We extended `align_labels_with_mapping` to support `Seque... | ### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like to encode:
```
data = {
... | 53 | Multi label class encoding
### Feature request
I have a multi label dataset and I'd like to be able to class encode the column and store the mapping directly in the features just as I can with a single label column. `class_encode_column` currently does not support multi labels.
Here's an example of what I'd like t... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Something that'd be nice to have is "manual update of state". One of the learning from training LLMs is the ability to skip some batches whenever we notice huge spike might be handy. | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 33 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Your outline spec is very sound and clear, @lhoestq - thank you!
@thomasw21, indeed that would be a wonderful extra feature. In Megatron-Deepspeed we manually drained the dataloader for the range we wanted. I wasn't very satisfied with the way we did it, since its behavior would change if you were to do multiple ran... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 97 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Hi there! I think this is a critical issue and have an urgent need for it, in my attempt to train on a super large-scale dataset using `datasets`. It is impossible to resume a time-consuming (like one month) experiment by iterating all seen data again, which could possibly cost several days.
@stas00 @thomasw21 @lhoe... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 63 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | No update so far, I wonder if someone implemented a resumable pytorch Sampler somwhere.
Then regarding resuming a streaming dataset, we'd first like to have an efficient way to skip shards automatically but this is not implemented yet | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 38 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | I opened a draft here for IterableDataset: https://github.com/huggingface/datasets/pull/6658
```python
"""Requires https://github.com/huggingface/datasets/pull/6658 (WIP)"""
from datasets import load_dataset
from torch.utils.data import DataLoader
ds = load_dataset(..., streaming=True)
# ds = ds.map(token... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 72 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Hi @lhoestq - can you provide more information and how to implement on saving and restoring vanilla DataLoader states with map-style datasets?
| It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 22 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | For now the easiest is probably to use the vanilla DataLoader only for batching and multiprocessing, and implement the resuming logic using a `Dataset` (it has `.select()` to skip examples) and a `dataset_state_dict`:
```python
from datasets import load_dataset
from torch.utils.data import DataLoader
ds = loa... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 100 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Hello, I found a similar implementation online that seems to solve your problem. https://github.com/facebookresearch/vissl/blob/main/vissl/data/data_helper.py#L93
it looks like we can set_start_iter in StatefulDistributedSampler to implement the stateful resume requirement we want.
| It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 30 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Hi y'all, @lhoestq I wanted to flag that we currently have a StatefulDataLoader in `pytorch/data/torchdata` that has state_dict/load_state_dict methods, which will call a dataset's state_dict/load_state_dict methods but also handle multiprocessing under the hood. Any chance we can collaborate on this and try to get the... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 59 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Fantastic ! This will help pushing our IterableDataset state_dict implementation at https://github.com/huggingface/datasets/pull/6658 :) I'll check if there is anything missing to maker them work together, and add tests and some docs referring to the StatefulDataLoader :) | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 36 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Ah I just saw this disclaimer in the torchdata README and it feels like people should not rely on it. Should the StatefulDataLoader live elsewhere @andrewkho ?
> ⚠️ As of July 2023, we have paused active development on TorchData and have paused new releases. We have learnt a lot from building it and hearing from use... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 108 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | @lhoestq Good find, we are in the midst of updating this disclaimer as we're re-starting development and regular releases, though our approach will be to iterate on DL V1 (ie StatefulDataLoader) instead of continuing development on datapipes+DLV2. Let's discuss on a call at some point to figure out the best path forwar... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 52 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | As a heads up, `IterableDataset` state_dict has been added in https://github.com/huggingface/datasets/pull/6658
...and it works out of the box with the `torchdata` `StatefulDataLoader` :)
See the docs at https://huggingface.co/docs/datasets/main/en/use_with_pytorch#checkpoint-and-resume | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 28 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | amazing! Thank you, @lhoestq
does it work with non-iterable dataset as well? the docs only mention iterable dataset | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 18 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | It's for iterable dataset only. For regular dataset I believe the sampler should implement state_dict, but maybe @andrewkho might know best how to resume a regular dataset with torchdata | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 29 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | @stas00 stateful dataloader will save and resume samplers for map style datasets. If no state_dict/load_state_dict is provided by the sampler, it will naively skip samples to fast forward. See here for more details https://github.com/pytorch/data/blob/main/torchdata/stateful_dataloader/README.md
Hope this helps! | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 37 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | 👋 I am trying to use `HF Streaming Dataset + TorchDDP + Stateful Dataloader`, to train using multiple nodes and large datasets.
So far, I have been able to use HF Streaming Dataset + TorchDDP with Vanilla Datasets. To do so, I implemented a custom iterable to make sure that shards are distributed across the mult... | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 282 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | Hi ! Have you tried using `split_dataset_by_node()` and pass the result to the StatefulDataLoader ?
```python
dataloader = StatefulDataLoader(split_dataset_by_node(dataset, rank=process_rank, world_size=world_size))
``` | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 22 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/5454 | Save and resume the state of a DataLoader | > Hi ! Have you tried using split_dataset_by_node() and pass the result to the StatefulDataLoader ?
@lhoestq it took me some time to test, but it works like a charm. Thanks for the pointer. Totally missed this 🤦. | It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to have a PyTorch Sampler state that can be sav... | 38 | Save and resume the state of a DataLoader
It would be nice when using `datasets` with a PyTorch DataLoader to be able to resume a training from a DataLoader state (e.g. to resume a training that crashed)
What I have in mind (but lmk if you have other ideas or comments):
For map-style datasets, this requires to ha... |
https://github.com/huggingface/datasets/issues/7811 | SIGSEGV when Python exits due to near null deref | The issue seems to come from `dill` which is a `datasets` dependency, e.g. this segfaults:
```python
import dill
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
`tqdm` seems to segfault when `dill` is imported. I only found this about segfault but it's may... | ### Describe the bug
When I run the following python script using datasets I get a segfault.
```python
from datasets import load_dataset
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
```
% lldb -- python3 crashmin.py
(lldb) target create "python3"
Cur... | 51 | SIGSEGV when Python exits due to near null deref
### Describe the bug
When I run the following python script using datasets I get a segfault.
```python
from datasets import load_dataset
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
```
% lldb -- python... |
https://github.com/huggingface/datasets/issues/7811 | SIGSEGV when Python exits due to near null deref | After more investigation it seems to be because of it imports `__main__`. This segfaults:
```python
import __main__
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
I opened an issue at https://github.com/tqdm/tqdm/issues/1687 | ### Describe the bug
When I run the following python script using datasets I get a segfault.
```python
from datasets import load_dataset
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
```
% lldb -- python3 crashmin.py
(lldb) target create "python3"
Cur... | 35 | SIGSEGV when Python exits due to near null deref
### Describe the bug
When I run the following python script using datasets I get a segfault.
```python
from datasets import load_dataset
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
```
% lldb -- python... |
https://github.com/huggingface/datasets/issues/7811 | SIGSEGV when Python exits due to near null deref | Here is a workaround. You can run your code as long as the progress bar is closed before exiting.
```python
from datasets import load_dataset
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
progress_bar.close() # avoids the segfault
``` | ### Describe the bug
When I run the following python script using datasets I get a segfault.
```python
from datasets import load_dataset
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
```
% lldb -- python3 crashmin.py
(lldb) target create "python3"
Cur... | 41 | SIGSEGV when Python exits due to near null deref
### Describe the bug
When I run the following python script using datasets I get a segfault.
```python
from datasets import load_dataset
from tqdm import tqdm
progress_bar = tqdm(total=(1000), unit='cols', desc='cols ')
progress_bar.update(1)
```
```
% lldb -- python... |
https://github.com/huggingface/datasets/issues/3598 | Readme info not being parsed to show on Dataset card page | i suspect a markdown parsing error, @severo do you want to take a quick look at it when you have some time? | ## Describe the bug
The info contained in the README.md file is not being shown in the dataset main page. Basic info and table of contents are properly formatted in the README.
## Steps to reproduce the bug
# Sample code to reproduce the bug
The README file is this one: https://huggingface.co/datasets/softcatal... | 22 | Readme info not being parsed to show on Dataset card page
## Describe the bug
The info contained in the README.md file is not being shown in the dataset main page. Basic info and table of contents are properly formatted in the README.
## Steps to reproduce the bug
# Sample code to reproduce the bug
The README f... |
https://github.com/huggingface/datasets/issues/3598 | Readme info not being parsed to show on Dataset card page | # Problem
The issue seems to coming from the front matter of the README
```---
annotations_creators:
- no-annotation
language_creators:
- machine-generated
languages:
- 'ca'
- 'de'
licenses:
- cc-by-4.0
multilinguality:
- translation
pretty_name: Catalan-German aligned corpora to train NMT systems.
size_... | ## Describe the bug
The info contained in the README.md file is not being shown in the dataset main page. Basic info and table of contents are properly formatted in the README.
## Steps to reproduce the bug
# Sample code to reproduce the bug
The README file is this one: https://huggingface.co/datasets/softcatal... | 121 | Readme info not being parsed to show on Dataset card page
## Describe the bug
The info contained in the README.md file is not being shown in the dataset main page. Basic info and table of contents are properly formatted in the README.
## Steps to reproduce the bug
# Sample code to reproduce the bug
The README f... |
https://github.com/huggingface/datasets/issues/3598 | Readme info not being parsed to show on Dataset card page | Thank you. It finally worked implementing your changes and leaving a white line between title and text in the description. | ## Describe the bug
The info contained in the README.md file is not being shown in the dataset main page. Basic info and table of contents are properly formatted in the README.
## Steps to reproduce the bug
# Sample code to reproduce the bug
The README file is this one: https://huggingface.co/datasets/softcatal... | 20 | Readme info not being parsed to show on Dataset card page
## Describe the bug
The info contained in the README.md file is not being shown in the dataset main page. Basic info and table of contents are properly formatted in the README.
## Steps to reproduce the bug
# Sample code to reproduce the bug
The README f... |
https://github.com/huggingface/datasets/issues/6389 | Index 339 out of range for dataset of size 339 <-- save_to_file() | I managed a workaround eventually but I don't know what it was (I made a lot of changes to seq2seq). I'll try to include generating code in the future. (If I close, I don't know if you see it. Feel free to close; I'll re-open if I encounter it again (if I can)). | ### Describe the bug
When saving out some Audio() data.
The data is audio recordings with associated 'sentences'.
(They use the audio 'bytes' approach because they're clips within audio files).
Code is below the traceback (I can't upload the voice audio/text (it's not even me)).
```
Traceback (most recent call ... | 53 | Index 339 out of range for dataset of size 339 <-- save_to_file()
### Describe the bug
When saving out some Audio() data.
The data is audio recordings with associated 'sentences'.
(They use the audio 'bytes' approach because they're clips within audio files).
Code is below the traceback (I can't upload the voice au... |
https://github.com/huggingface/datasets/issues/3809 | Checksums didn't match for datasets on Google Drive | Hi @muelletm, thanks for reporting.
This issue was already reported and its root cause is a change in the Google Drive service. See:
- #3786
We have already fixed it. See:
- #3787
Until our next `datasets` library release, you can get this fix by installing our library from the GitHub master branch:
```sh... | ## Describe the bug
Datasets hosted on Google Drive do not seem to work right now.
Loading them fails with a checksum error.
## Steps to reproduce the bug
```python
from datasets import load_dataset
for dataset in ["head_qa", "yelp_review_full"]:
try:
load_dataset(dataset)
except Exception as excep... | 103 | Checksums didn't match for datasets on Google Drive
## Describe the bug
Datasets hosted on Google Drive do not seem to work right now.
Loading them fails with a checksum error.
## Steps to reproduce the bug
```python
from datasets import load_dataset
for dataset in ["head_qa", "yelp_review_full"]:
try:
... |
https://github.com/huggingface/datasets/issues/8052 | [Security] Zip/Rar/7z Archive Extraction Path Traversal (Incomplete CVE-2007-4559 Mitigation) | Hi! I'm currently exploring the codebase and came across this issue. I'd be interested in investigating it further and trying to reproduce the behavior locally.
If no one is actively working on it, I'd be happy to take a look and attempt a fix.
| ## Security Vulnerability Report
**Reporter:** Conner Webber (conner.webber000@gmail.com)
**Severity:** HIGH (CVSS 3.1: 8.2)
**CWE:** CWE-22 (Improper Limitation of a Pathname to a Restricted Directory)
## Summary
The `ZipExtractor`, `RarExtractor`, and `SevenZipExtractor` classes in `datasets/utils/extract.py` call... | 44 | [Security] Zip/Rar/7z Archive Extraction Path Traversal (Incomplete CVE-2007-4559 Mitigation)
## Security Vulnerability Report
**Reporter:** Conner Webber (conner.webber000@gmail.com)
**Severity:** HIGH (CVSS 3.1: 8.2)
**CWE:** CWE-22 (Improper Limitation of a Pathname to a Restricted Directory)
## Summary
The `ZipE... |
https://github.com/huggingface/datasets/issues/8052 | [Security] Zip/Rar/7z Archive Extraction Path Traversal (Incomplete CVE-2007-4559 Mitigation) | The steps to reproduce don't cause any file extraction outside of output_path
see https://docs.python.org/2/library/zipfile.html#zipfile.ZipFile.extract
> Note If a member filename is an absolute path, a drive/UNC sharepoint and leading (back)slashes will be stripped, e.g.: ///foo/bar becomes foo/bar on Unix, and C:\... | ## Security Vulnerability Report
**Reporter:** Conner Webber (conner.webber000@gmail.com)
**Severity:** HIGH (CVSS 3.1: 8.2)
**CWE:** CWE-22 (Improper Limitation of a Pathname to a Restricted Directory)
## Summary
The `ZipExtractor`, `RarExtractor`, and `SevenZipExtractor` classes in `datasets/utils/extract.py` call... | 83 | [Security] Zip/Rar/7z Archive Extraction Path Traversal (Incomplete CVE-2007-4559 Mitigation)
## Security Vulnerability Report
**Reporter:** Conner Webber (conner.webber000@gmail.com)
**Severity:** HIGH (CVSS 3.1: 8.2)
**CWE:** CWE-22 (Improper Limitation of a Pathname to a Restricted Directory)
## Summary
The `ZipE... |
https://github.com/huggingface/datasets/issues/8052 | [Security] Zip/Rar/7z Archive Extraction Path Traversal (Incomplete CVE-2007-4559 Mitigation) | Thanks @lhoestq — you're right, I missed that \ already strips \ components and absolute paths in modern Python. Appreciate you adding the extra safeguards anyway as defense-in-depth. Good catch on the stdlib behavior. | ## Security Vulnerability Report
**Reporter:** Conner Webber (conner.webber000@gmail.com)
**Severity:** HIGH (CVSS 3.1: 8.2)
**CWE:** CWE-22 (Improper Limitation of a Pathname to a Restricted Directory)
## Summary
The `ZipExtractor`, `RarExtractor`, and `SevenZipExtractor` classes in `datasets/utils/extract.py` call... | 34 | [Security] Zip/Rar/7z Archive Extraction Path Traversal (Incomplete CVE-2007-4559 Mitigation)
## Security Vulnerability Report
**Reporter:** Conner Webber (conner.webber000@gmail.com)
**Severity:** HIGH (CVSS 3.1: 8.2)
**CWE:** CWE-22 (Improper Limitation of a Pathname to a Restricted Directory)
## Summary
The `ZipE... |
https://github.com/huggingface/datasets/issues/7691 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming | It seems to me that if we have something that is so large that it cannot fit in pa.table, the fallback method should be to just set it as "binary" type, perhaps? | ### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, even with streaming.
I made a config for the dataset that specifically inclu... | 32 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming
### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, e... |
https://github.com/huggingface/datasets/issues/7691 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming | I also tried creating a dataset_info.json but the webdataset builder didn't seem to look for it and load it | ### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, even with streaming.
I made a config for the dataset that specifically inclu... | 19 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming
### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, e... |
https://github.com/huggingface/datasets/issues/7691 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming | Workaround on my end, removed all videos larger than 2GB for now. The dataset no longer crashes. | ### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, even with streaming.
I made a config for the dataset that specifically inclu... | 17 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming
### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, e... |
https://github.com/huggingface/datasets/issues/7691 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming | Potential patch to webdataset.py could be like so:
```python
LARGE_THRESHOLD = 2 * 1024 * 1024 * 1024 # 2 GB
large_fields = set()
# Replace large binary fields with None for schema inference
processed_examples = []
for example in first_examples:
new_example = {}
for k, v in example.items():
if isinst... | ### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, even with streaming.
I made a config for the dataset that specifically inclu... | 109 | Large WebDataset: pyarrow.lib.ArrowCapacityError on load() even with streaming
### Describe the bug
I am creating a large WebDataset-format dataset for sign language processing research, and a number of the videos are over 2GB. The instant I hit one of the shards with one of those videos, I get a ArrowCapacityError, e... |
https://github.com/huggingface/datasets/issues/6568 | keep_in_memory=True does not seem to work | Seems like I just used the old code which did not have `keep_in_memory=True` argument, sorry.
Although i encountered a different problem – at 97% my python process just hung for around 11 minutes with no logs (when running dataset.map without `keep_in_memory=True` over around 3 million of dataset samples)... | UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :( | 48 | keep_in_memory=True does not seem to work
UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :(
Seems like I just used the old code which did not have `keep_in_memory=True` argument, sorry.
Although i encountered a different problem – at 97% my python ... |
https://github.com/huggingface/datasets/issues/6568 | keep_in_memory=True does not seem to work | Can you open a new issue and provide a bit more details ? What kind of map operations did you run ? | UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :( | 22 | keep_in_memory=True does not seem to work
UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :(
Can you open a new issue and provide a bit more details ? What kind of map operations did you run ? |
https://github.com/huggingface/datasets/issues/6568 | keep_in_memory=True does not seem to work | Hey. I will try to find some free time to describe it.
(can't do it now, cause i need to reproduce it myself to be sure about everything, which requires spinning a new Azuree VM, copying a huge dataset to drive from network disk for a long time etc...) | UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :( | 49 | keep_in_memory=True does not seem to work
UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :(
Hey. I will try to find some free time to describe it.
(can't do it now, cause i need to reproduce it myself to be sure about everything, which requires spi... |
https://github.com/huggingface/datasets/issues/6568 | keep_in_memory=True does not seem to work | @lhoestq loading dataset like this does not spawn 50 python processes:
```
datasets.load_dataset("/preprocessed_2256k/train", num_proc=50)
```
I have 64 vCPU so i hoped it could speed up the dataset loading...
My dataset onlly has images and metadata.csv with text column alongside image file path column | UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :( | 44 | keep_in_memory=True does not seem to work
UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :(
@lhoestq loading dataset like this does not spawn 50 python processes:
```
datasets.load_dataset("/preprocessed_2256k/train", num_proc=50)
```
I have 6... |
https://github.com/huggingface/datasets/issues/6568 | keep_in_memory=True does not seem to work | now noticed
```
'Setting num_proc from 50 back to 1 for the train split to disable multiprocessing as it only contains one shard
```
Any way to work around this? | UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :( | 30 | keep_in_memory=True does not seem to work
UPD: [Fixed](https://github.com/huggingface/datasets/issues/6568#issuecomment-1880817794) . But a new issue came up :(
now noticed
```
'Setting num_proc from 50 back to 1 for the train split to disable multiprocessing as it only contains one shard
```
Any way to work arou... |
https://github.com/huggingface/datasets/issues/3464 | struct.error: 'i' format requires -2147483648 <= number <= 2147483647 | Hi ! Can you try setting `datasets.config.MAX_TABLE_NBYTES_FOR_PICKLING` to a smaller value than `4 << 30` (4GiB), for example `500 << 20` (500MiB) ? It should reduce the maximum size of the arrow table being pickled during multiprocessing.
If it fixes the issue, we can consider lowering the default value for everyo... | ## Describe the bug
A clear and concise description of what the bug is.
using latest datasets=datasets-1.16.1-py3-none-any.whl
process my own multilingual dataset by following codes, and the number of rows in all dataset is 306000, the max_length of each sentence is 256:
` | ## Describe the bug
After discussion with @lhoestq, just want to mention here that `glob.glob(...)` should always be used in combination with `sorted(...)` to make sure the list of files returned by `glob.glob(...)` doesn't change depending on the OS system.
There are currently multiple datasets that use `glob.g... | 48 | Order of dataset changes due to glob.glob.
## Describe the bug
After discussion with @lhoestq, just want to mention here that `glob.glob(...)` should always be used in combination with `sorted(...)` to make sure the list of files returned by `glob.glob(...)` doesn't change depending on the OS system.
There are c... |
https://github.com/huggingface/datasets/issues/3298 | Agnews dataset viewer is not working | Hi ! Thanks for reporting
We've already fixed the code that generates the preview for this dataset, we'll release the fix soon :) | ## Dataset viewer issue for '*name of the dataset*'
**Link:** https://huggingface.co/datasets/ag_news
Hi there, the `ag_news` dataset viewer is not working.
Am I the one who added this dataset? No
| 23 | Agnews dataset viewer is not working
## Dataset viewer issue for '*name of the dataset*'
**Link:** https://huggingface.co/datasets/ag_news
Hi there, the `ag_news` dataset viewer is not working.
Am I the one who added this dataset? No
Hi ! Thanks for reporting
We've already fixed the code that generates the ... |
https://github.com/huggingface/datasets/issues/7894 | embed_table_storage crashes (SIGKILL) on sharded datasets with Sequence() nested types | I wasn't able to reproduce the crash on my side (macos arm 54, pyarrow 22 and a nifti file I found [online](https://s3.amazonaws.com/openneuro.org/ds004884/sub-M2001/ses-1076/anat/sub-M2001_ses-1076_acq-tfl3_run-4_T1w.nii.gz?versionId=9aVGb3C.VcoBgxrhNzFnL6O0MvxQsXX7&AWSAccessKeyId=AKIARTA7OOV5WQ3DGSOB&Signature=LQMLzj... | ## Summary
`embed_table_storage` crashes with SIGKILL (exit code 137) when processing sharded datasets containing `Sequence()` nested types like `Sequence(Nifti())`. Likely affects `Sequence(Image())` and `Sequence(Audio())` as well.
The crash occurs at the C++ level with no Python traceback.
### Related Issues
- #... | 41 | embed_table_storage crashes (SIGKILL) on sharded datasets with Sequence() nested types
## Summary
`embed_table_storage` crashes with SIGKILL (exit code 137) when processing sharded datasets containing `Sequence()` nested types like `Sequence(Nifti())`. Likely affects `Sequence(Image())` and `Sequence(Audio())` as well... |
https://github.com/huggingface/datasets/issues/7894 | embed_table_storage crashes (SIGKILL) on sharded datasets with Sequence() nested types | Hi @lhoestq,
Thank you so much for taking the time to investigate this. Your comment about not being able to reproduce it with a single NIfTI file actually helped me understand the bug better.
**Key finding:** This bug is scale-dependent. It only manifests with real, full-scale data, and not with synthetic test files... | ## Summary
`embed_table_storage` crashes with SIGKILL (exit code 137) when processing sharded datasets containing `Sequence()` nested types like `Sequence(Nifti())`. Likely affects `Sequence(Image())` and `Sequence(Audio())` as well.
The crash occurs at the C++ level with no Python traceback.
### Related Issues
- #... | 251 | embed_table_storage crashes (SIGKILL) on sharded datasets with Sequence() nested types
## Summary
`embed_table_storage` crashes with SIGKILL (exit code 137) when processing sharded datasets containing `Sequence()` nested types like `Sequence(Nifti())`. Likely affects `Sequence(Image())` and `Sequence(Audio())` as well... |
https://github.com/huggingface/datasets/issues/7894 | embed_table_storage crashes (SIGKILL) on sharded datasets with Sequence() nested types | @lhoestq Brief update - I've added a reproduction that uses standard `ds.push_to_hub()` (no custom code).
**Reproduction branch:** https://github.com/The-Obstacle-Is-The-Way/arc-aphasia-bids/tree/sandbox/reproduce-bug-7894
**To reproduce with standard library:**
```bash
git clone -b sandbox/reproduce-bug-7894 https:/... | ## Summary
`embed_table_storage` crashes with SIGKILL (exit code 137) when processing sharded datasets containing `Sequence()` nested types like `Sequence(Nifti())`. Likely affects `Sequence(Image())` and `Sequence(Audio())` as well.
The crash occurs at the C++ level with no Python traceback.
### Related Issues
- #... | 83 | embed_table_storage crashes (SIGKILL) on sharded datasets with Sequence() nested types
## Summary
`embed_table_storage` crashes with SIGKILL (exit code 137) when processing sharded datasets containing `Sequence()` nested types like `Sequence(Nifti())`. Likely affects `Sequence(Image())` and `Sequence(Audio())` as well... |
https://github.com/huggingface/datasets/issues/7901 | ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint | Hi ! As you can read in the logs, the shuffle buffer content is lost when resuming a shuffled dataset. The default size is 1000 examples, but you can tweak it
e.g. if you run your code with this
```diff
- ds = Dataset.from_dict({"a": range(12)}).to_iterable_dataset(num_shards=1)
- ds = ds.shuffle(seed=42)
+ ds = Data... | ### Describe the bug
ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Steps to reproduce the bug
1. The reproducible code is as follows：
```
from datasets import Dataset, concatenate_datasets, interleave_datasets
ds = Dataset.from_dict({"a": range(12)}).to_iterable_dataset(num_sha... | 130 | ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Describe the bug
ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Steps to reproduce the bug
1. The reproducible code is as follows：
```
from datasets import Dataset, concatenate_datasets, interlea... |
https://github.com/huggingface/datasets/issues/7901 | ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint | > Hi ! As you can read in the logs, the shuffle buffer content is lost when resuming a shuffled dataset. The default size is 1000 examples, but you can tweak it
>
> e.g. if you run your code with this
>
> - ds = Dataset.from_dict({"a": range(12)}).to_iterable_dataset(num_shards=1)
> - ds = ds.shuffle(seed=42)
> + ds ... | ### Describe the bug
ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Steps to reproduce the bug
1. The reproducible code is as follows：
```
from datasets import Dataset, concatenate_datasets, interleave_datasets
ds = Dataset.from_dict({"a": range(12)}).to_iterable_dataset(num_sha... | 173 | ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Describe the bug
ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Steps to reproduce the bug
1. The reproducible code is as follows：
```
from datasets import Dataset, concatenate_datasets, interlea... |
https://github.com/huggingface/datasets/issues/7901 | ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint | Yes correct. This is because the state_dict doesn't save the content of the buffer, so when resuming the buffer starts empty and the examples that were in the buffer are lost. | ### Describe the bug
ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Steps to reproduce the bug
1. The reproducible code is as follows：
```
from datasets import Dataset, concatenate_datasets, interleave_datasets
ds = Dataset.from_dict({"a": range(12)}).to_iterable_dataset(num_sha... | 31 | ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Describe the bug
ShuffledDataSourcesArrowExamplesIterable cannot properly resume from checkpoint
### Steps to reproduce the bug
1. The reproducible code is as follows：
```
from datasets import Dataset, concatenate_datasets, interlea... |
https://github.com/huggingface/datasets/issues/7457 | Document the HF_DATASETS_CACHE env variable | Strongly agree to this, in addition, I am also suffering to change the cache location similar to other issues (since I changed the environmental variables).
https://github.com/huggingface/datasets/issues/6886 | ### Feature request
Hello,
I have a use case where my team is sharing models and dataset in shared directory to avoid duplication.
I noticed that the [cache documentation for datasets](https://huggingface.co/docs/datasets/main/en/cache) only mention the `HF_HOME` environment variable but never the `HF_DATASETS_CACHE`... | 26 | Document the HF_DATASETS_CACHE env variable
### Feature request
Hello,
I have a use case where my team is sharing models and dataset in shared directory to avoid duplication.
I noticed that the [cache documentation for datasets](https://huggingface.co/docs/datasets/main/en/cache) only mention the `HF_HOME` environmen... |
https://github.com/huggingface/datasets/issues/7457 | Document the HF_DATASETS_CACHE env variable | sure ! you can also comment #self-assign in an issue and a bot assigns you automatically :) | ### Feature request
Hello,
I have a use case where my team is sharing models and dataset in shared directory to avoid duplication.
I noticed that the [cache documentation for datasets](https://huggingface.co/docs/datasets/main/en/cache) only mention the `HF_HOME` environment variable but never the `HF_DATASETS_CACHE`... | 17 | Document the HF_DATASETS_CACHE env variable
### Feature request
Hello,
I have a use case where my team is sharing models and dataset in shared directory to avoid duplication.
I noticed that the [cache documentation for datasets](https://huggingface.co/docs/datasets/main/en/cache) only mention the `HF_HOME` environmen... |
https://github.com/huggingface/datasets/issues/6194 | Support custom fingerprinting with `Dataset.from_generator` | The `fingerprint` parameter serves a slightly different purpose - we use it to inject a new fingerprint after transforming a `Dataset` (computed from the previous fingerprint + transform + transform args), e.g., to be able to compute the cache file for a transform. There is no concept of `fingerprint` before a `Dataset... | ### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
### Motivation
Using the `.from_generator` constructor with ... | 119 | Support custom fingerprinting with `Dataset.from_generator`
### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
###... |
https://github.com/huggingface/datasets/issues/6194 | Support custom fingerprinting with `Dataset.from_generator` | Adding +1 here:
If the generator needs to access some external resources or state, then it's not always straightforward to make it pickle-able. So I'd like to be able to override how the default cache key derivation needs to pickle the generator (and of course, I'd accept responsibility for that part of cache consis... | ### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
### Motivation
Using the `.from_generator` constructor with ... | 65 | Support custom fingerprinting with `Dataset.from_generator`
### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
###... |
https://github.com/huggingface/datasets/issues/6194 | Support custom fingerprinting with `Dataset.from_generator` | Silly hack incoming:
```python
import uuid
class _DatasetGeneratorPickleHack:
def __init__(self, generator, generator_id=None):
self.generator = generator
self.generator_id = (
generator_id if generator_id is not None else str(uuid.uuid4())
)
def __call__(self,... | ### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
### Motivation
Using the `.from_generator` constructor with ... | 82 | Support custom fingerprinting with `Dataset.from_generator`
### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
###... |
https://github.com/huggingface/datasets/issues/6194 | Support custom fingerprinting with `Dataset.from_generator` | I'd like some way to do this too. I find that sometimes the hash doesn't cover enough, and that the dataset is not regenerated even when underlying data has changed, and by supplying a custom fingerprint I could do a better job of controlling when my dataset is regenerated. | ### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
### Motivation
Using the `.from_generator` constructor with ... | 49 | Support custom fingerprinting with `Dataset.from_generator`
### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
###... |
https://github.com/huggingface/datasets/issues/6194 | Support custom fingerprinting with `Dataset.from_generator` | I ran into the same thing - my actual generator reads from a disk source that might have new data (images) available at some point and it ends up ignoring calling the generator. Thanks for the hack @mlin 👋 | ### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
### Motivation
Using the `.from_generator` constructor with ... | 39 | Support custom fingerprinting with `Dataset.from_generator`
### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
###... |
https://github.com/huggingface/datasets/issues/6194 | Support custom fingerprinting with `Dataset.from_generator` | just wanted to pitch my support for an easy control over the generator id. requiring that generators are pickleable just to get a unique id is limiting: plenty of classes (maybe even hf.datasets own) are written with no pickle support in mind. also as mentioned above the state of a generator might extend beyond its pic... | ### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
### Motivation
Using the `.from_generator` constructor with ... | 56 | Support custom fingerprinting with `Dataset.from_generator`
### Feature request
When using `Dataset.from_generator`, the generator is hashed when building the fingerprint. Similar to `.map`, it would be interesting to let the user bypass this hashing by accepting a `fingerprint` argument to `.from_generator`.
###... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | Hi ! Thanks for reporting :)
When using a PyTorch's data loader with `num_workers>1` and an iterable dataset, each worker streams the exact same data by default, resulting in duplicate data when iterating using the data loader.
We can probably fix this in `datasets` by checking `torch.utils.data.get_worker_info()... | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 55 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | > Hi ! Thanks for reporting :)
>
> When using a PyTorch's data loader with `num_workers>1` and an iterable dataset, each worker streams the exact same data by default, resulting in duplicate data when iterating using the data loader.
>
> We can probably fix this in `datasets` by checking `torch.utils.data.get_wor... | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 74 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | Isn’t that somehow a bug on PyTorch side? (Just asking because this behavior seems quite general and maybe not what would be intended) | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 23 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | From PyTorch's documentation [here](https://pytorch.org/docs/stable/data.html#dataset-types):
> When using an IterableDataset with multi-process data loading. The same dataset object is replicated on each worker process, and thus the replicas must be configured differently to avoid duplicated data. See [IterableData... | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 127 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | Hi there @lhoestq @cloudyuyuyu
I met that problem recently, and #4375 is really useful because I finally found out I am training with duplicate data.
However, in multi-GPU training, I'm using DDP mode and IterableDataset, which still yields duplicate data for each progress. And this is dangerous because users maybe ... | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 54 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | If the worker_info.id is unique per process it should work fine, could you check that they're unique ?
The code to get the worker_info in each worker is `torch.utils.data.get_worker_info()` | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 29 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | test.py
```python
import json
import os
import torch
from torch.utils.data import IterableDataset, DataLoader
from transformers import PreTrainedTokenizer, TrainingArguments
from common.arguments import DataTrainingArguments, ModelArguments
class MyIterableDataset(IterableDataset):
def __iter__(sel... | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 84 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | It looks like a bug from pytorch no ? How can we know which data should go in which process when using DDP ?
I guess we need to check `torch.distributed.get_world_size()` and `torch.distributed.get_rank()` as well. Not fan of the design here tbh, but that's how it is | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 47 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | > It looks like a bug from pytorch no ? How can we know which data should go in which process when using DDP ?
>
> I guess we need to check `torch.distributed.get_world_size()` and `torch.distributed.get_rank()` as well. Not fan of the design here tbh, but that's how it is
Maybe we should document it? | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 55 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/3423 | data duplicate when setting num_works > 1 with streaming data | hmm actually let me open a new issue on DDP - original post was for single node | ## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy as np
import os
from datasets import load_dataset
from tor... | 17 | data duplicate when setting num_works > 1 with streaming data
## Describe the bug
The data is repeated num_works times when we load_dataset with streaming and set num_works > 1 when construct dataloader
## Steps to reproduce the bug
```python
# Sample code to reproduce the bug
import pandas as pd
import numpy a... |
https://github.com/huggingface/datasets/issues/6051 | Skipping shard in the remote repo and resume upload | Hi! `_select_contiguous` fetches a (zero-copy) slice of the dataset's Arrow table to build a shard, so I don't think this part is the problem. To me, the issue seems to be the step where we embed external image files' bytes (a lot of file reads). You can use `.map` with multiprocessing to perform this step before `push... | ### Describe the bug
For some reason when I try to resume the upload of my dataset, it is very slow to reach the index of the shard from which to resume the uploading.
From my understanding, the problem is in this part of the code:
arrow_dataset.py
```python
for index, shard in logging.tqdm(
enume... | 111 | Skipping shard in the remote repo and resume upload
### Describe the bug
For some reason when I try to resume the upload of my dataset, it is very slow to reach the index of the shard from which to resume the uploading.
From my understanding, the problem is in this part of the code:
arrow_dataset.py
```python
... |
https://github.com/huggingface/datasets/issues/6051 | Skipping shard in the remote repo and resume upload | Hi, thanks, this solution saves some time.
But can't we avoid embedding all external image files bytes with each push, skipping the images that have already been pushed into the repo?
Edit: Ok I missed the part of cache it manually on the disk the first time, this solves the problem. Thank you | ### Describe the bug
For some reason when I try to resume the upload of my dataset, it is very slow to reach the index of the shard from which to resume the uploading.
From my understanding, the problem is in this part of the code:
arrow_dataset.py
```python
for index, shard in logging.tqdm(
enume... | 53 | Skipping shard in the remote repo and resume upload
### Describe the bug
For some reason when I try to resume the upload of my dataset, it is very slow to reach the index of the shard from which to resume the uploading.
From my understanding, the problem is in this part of the code:
arrow_dataset.py
```python
... |
https://github.com/huggingface/datasets/issues/7536 | [Errno 13] Permission denied: on `.incomplete` file | It must be an issue with umask being used by multiple threads indeed. Maybe we can try to make a thread safe function to apply the umask (using filelock for example) | ### Describe the bug
When downloading a dataset, we frequently hit the below Permission Denied error. This looks to happen (at least) across datasets in HF, S3, and GCS.
It looks like the `temp_file` being passed [here](https://github.com/huggingface/datasets/blob/main/src/datasets/utils/file_utils.py#L412) can somet... | 31 | [Errno 13] Permission denied: on `.incomplete` file
### Describe the bug
When downloading a dataset, we frequently hit the below Permission Denied error. This looks to happen (at least) across datasets in HF, S3, and GCS.
It looks like the `temp_file` being passed [here](https://github.com/huggingface/datasets/blob/m... |
https://github.com/huggingface/datasets/issues/7536 | [Errno 13] Permission denied: on `.incomplete` file | > It must be an issue with umask being used by multiple threads indeed. Maybe we can try to make a thread safe function to apply the umask (using filelock for example)
@lhoestq is this something which can go in a 3.5.1 release? | ### Describe the bug
When downloading a dataset, we frequently hit the below Permission Denied error. This looks to happen (at least) across datasets in HF, S3, and GCS.
It looks like the `temp_file` being passed [here](https://github.com/huggingface/datasets/blob/main/src/datasets/utils/file_utils.py#L412) can somet... | 43 | [Errno 13] Permission denied: on `.incomplete` file
### Describe the bug
When downloading a dataset, we frequently hit the below Permission Denied error. This looks to happen (at least) across datasets in HF, S3, and GCS.
It looks like the `temp_file` being passed [here](https://github.com/huggingface/datasets/blob/m... |
https://github.com/huggingface/datasets/issues/4241 | NonMatchingChecksumError when attempting to download GLUE | Hi :)
I think your issue may be related to the older `nlp` library. I was able to download `glue` with the latest version of `datasets`. Can you try updating with:
```py
pip install -U datasets
```
Then you can download:
```py
from datasets import load_dataset
ds = load_dataset("glue", "rte")
``` | ## Describe the bug
I am trying to download the GLUE dataset from the NLP module but get an error (see below).
## Steps to reproduce the bug
```python
import nlp
nlp.__version__ # '0.2.0'
nlp.load_dataset('glue', name="rte", download_mode="force_redownload")
```
## Expected results
I expect the dataset to ... | 51 | NonMatchingChecksumError when attempting to download GLUE
## Describe the bug
I am trying to download the GLUE dataset from the NLP module but get an error (see below).
## Steps to reproduce the bug
```python
import nlp
nlp.__version__ # '0.2.0'
nlp.load_dataset('glue', name="rte", download_mode="force_redownlo... |
https://github.com/huggingface/datasets/issues/4241 | NonMatchingChecksumError when attempting to download GLUE | This appears to work. Thank you!
On Wed, Apr 27, 2022, 1:18 PM Steven Liu ***@***.***> wrote:
> Hi :)
>
> I think your issue may be related to the older nlp library. I was able to
> download glue with the latest version of datasets. Can you try updating
> with:
>
> pip install -U datasets
>
> Then you can download:
>... | ## Describe the bug
I am trying to download the GLUE dataset from the NLP module but get an error (see below).
## Steps to reproduce the bug
```python
import nlp
nlp.__version__ # '0.2.0'
nlp.load_dataset('glue', name="rte", download_mode="force_redownload")
```
## Expected results
I expect the dataset to ... | 110 | NonMatchingChecksumError when attempting to download GLUE
## Describe the bug
I am trying to download the GLUE dataset from the NLP module but get an error (see below).
## Steps to reproduce the bug
```python
import nlp
nlp.__version__ # '0.2.0'
nlp.load_dataset('glue', name="rte", download_mode="force_redownlo... |
https://github.com/huggingface/datasets/issues/7508 | Iterating over Image feature columns is extremely slow | Hi ! Could it be because the `Image()` type in dataset does `image = Image.open(image_path)` and also `image.load()` which actually loads the image data in memory ? This is needed to avoid too many open files issues, see https://github.com/huggingface/datasets/issues/3985 | We are trying to load datasets where the image column stores `PIL.PngImagePlugin.PngImageFile` images. However, iterating over these datasets is extremely slow.
What I have found:
1. It is the presence of the image column that causes the slowdown. Removing the column from the dataset results in blazingly fast (as expe... | 39 | Iterating over Image feature columns is extremely slow
We are trying to load datasets where the image column stores `PIL.PngImagePlugin.PngImageFile` images. However, iterating over these datasets is extremely slow.
What I have found:
1. It is the presence of the image column that causes the slowdown. Removing the col... |
https://github.com/huggingface/datasets/issues/7508 | Iterating over Image feature columns is extremely slow | Yes, that seems to be it. For my purposes, I've cast the column to `Image(decode=False)`, and only load the images when necessary, which is much much faster | We are trying to load datasets where the image column stores `PIL.PngImagePlugin.PngImageFile` images. However, iterating over these datasets is extremely slow.
What I have found:
1. It is the presence of the image column that causes the slowdown. Removing the column from the dataset results in blazingly fast (as expe... | 27 | Iterating over Image feature columns is extremely slow
We are trying to load datasets where the image column stores `PIL.PngImagePlugin.PngImageFile` images. However, iterating over these datasets is extremely slow.
What I have found:
1. It is the presence of the image column that causes the slowdown. Removing the col... |
https://github.com/huggingface/datasets/issues/3729 | Wrong number of examples when loading a text dataset | Hi @kg-nlp, thanks for reporting.
That is weird... I guess we would need some sample data file where this behavior appears to reproduce the bug for further investigation... | ## Describe the bug
when I use load_dataset to read a txt file I find that the number of the samples is incorrect
## Steps to reproduce the bug
```
fr = open('train.txt','r',encoding='utf-8').readlines()
print(len(fr)) # 1199637
datasets = load_dataset('text', data_files={'train': ['train.txt']}, streaming... | 28 | Wrong number of examples when loading a text dataset
## Describe the bug
when I use load_dataset to read a txt file I find that the number of the samples is incorrect
## Steps to reproduce the bug
```
fr = open('train.txt','r',encoding='utf-8').readlines()
print(len(fr)) # 1199637
datasets = load_dataset('... |
https://github.com/huggingface/datasets/issues/3729 | Wrong number of examples when loading a text dataset | ok, I found the reason why that two results are not same.
there is /u2029 in the text, the datasets will split sentence according to the /u2029,but when I use open function will not do that .
so I want to know which function shell do that
thanks | ## Describe the bug
when I use load_dataset to read a txt file I find that the number of the samples is incorrect
## Steps to reproduce the bug
```
fr = open('train.txt','r',encoding='utf-8').readlines()
print(len(fr)) # 1199637
datasets = load_dataset('text', data_files={'train': ['train.txt']}, streaming... | 48 | Wrong number of examples when loading a text dataset
## Describe the bug
when I use load_dataset to read a txt file I find that the number of the samples is incorrect
## Steps to reproduce the bug
```
fr = open('train.txt','r',encoding='utf-8').readlines()
print(len(fr)) # 1199637
datasets = load_dataset('... |
https://github.com/huggingface/datasets/issues/4181 | Support streaming FLEURS dataset | Yes, you just have to use `dl_manager.iter_archive` instead of `dl_manager.download_and_extract`.
That's because `download_and_extract` doesn't support TAR archives in streaming mode. | ## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.tar.gz' is not implemented in str... | 20 | Support streaming FLEURS dataset
## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.... |
https://github.com/huggingface/datasets/issues/4181 | Support streaming FLEURS dataset | Tried to make it streamable, but I don't think it's really possible. @lhoestq @polinaeterna maybe you guys can check:
https://huggingface.co/datasets/google/fleurs/commit/dcf80160cd77977490a8d32b370c027107f2407b
real quick.
I think the problem is that we cannot ensure that the metadata file is found before th... | ## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.tar.gz' is not implemented in str... | 47 | Support streaming FLEURS dataset
## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.... |
https://github.com/huggingface/datasets/issues/4181 | Support streaming FLEURS dataset | @patrickvonplaten I think the metadata file should be found first because the audio files are contained in a folder next to the metadata files (just as in common voice), so the metadata files should be "on top of the list" as they are closer to the root in the directories hierarchy | ## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.tar.gz' is not implemented in str... | 51 | Support streaming FLEURS dataset
## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.... |
https://github.com/huggingface/datasets/issues/4181 | Support streaming FLEURS dataset | The order of the files is determined when the TAR archive is created, depending on the commands the creator ran.
If the metadata file is not at the beginning of the file, that makes streaming completely inefficient. In this case the TAR archive needs to be recreated in an appropriate order. | ## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.tar.gz' is not implemented in str... | 51 | Support streaming FLEURS dataset
## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.... |
https://github.com/huggingface/datasets/issues/4181 | Support streaming FLEURS dataset | Actually we could maybe just host the metadata file ourselves and then stream the audio data only. Don't think that this would be a problem for the FLEURS authors (I can ask them :-)) | ## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.tar.gz' is not implemented in str... | 34 | Support streaming FLEURS dataset
## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.... |
https://github.com/huggingface/datasets/issues/4181 | Support streaming FLEURS dataset | I made a PR to their repo to support streaming (by uploading the metadata file to the Hub). See:
- https://huggingface.co/datasets/google/fleurs/discussions/4 | ## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.tar.gz' is not implemented in str... | 21 | Support streaming FLEURS dataset
## Dataset viewer issue for '*name of the dataset*'
https://huggingface.co/datasets/google/fleurs
```
Status code: 400
Exception: NotImplementedError
Message: Extraction protocol for TAR archives like 'https://storage.googleapis.com/xtreme_translations/FLEURS/af_za.... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.