html_url stringlengths 51 51 | comments stringlengths 67 24.7k | title stringlengths 6 280 | body stringlengths 51 36.2k | comment_length int64 16 1.45k | text stringlengths 190 38.3k | embeddings list |
|---|---|---|---|---|---|---|
https://github.com/huggingface/datasets/issues/2547 | There are no solutions yet unfortunately.
We're still trying to figure out a way to make the loading instantaneous on such disks, I'll keep you posted | Dataset load_from_disk is too slow | @lhoestq
## Describe the bug
It's not normal that I have to wait 7-8 hours for a dataset to be loaded from disk, as there are no preprocessing steps, it's only loading it with load_from_disk. I have 96 cpus, however only 1 is used for this, which is inefficient. Moreover, its usage is at 1%... This is happening in t... | 26 | Dataset load_from_disk is too slow
@lhoestq
## Describe the bug
It's not normal that I have to wait 7-8 hours for a dataset to be loaded from disk, as there are no preprocessing steps, it's only loading it with load_from_disk. I have 96 cpus, however only 1 is used for this, which is inefficient. Moreover, its usa... | [
-0.4694788157939911,
-0.24230189621448517,
-0.11789415776729584,
0.3941159248352051,
0.2421811819076538,
0.05294790863990784,
0.19930562376976013,
0.265770822763443,
0.4733242392539978,
0.1050993874669075,
0.05338817834854126,
0.449489027261734,
-0.02447601780295372,
0.013596870005130768,
... |
https://github.com/huggingface/datasets/issues/2543 | Hi @stas00, thanks for pointing out this issue with logging.
I agree that `datasets` can sometimes be too verbose... I can create a PR and we could discuss there the choice of the log levels for different parts of the code. | switching some low-level log.info's to log.debug? | In https://github.com/huggingface/transformers/pull/12276 we are now changing the examples to have `datasets` on the same log level as `transformers`, so that one setting can do a consistent logging across all involved components.
The trouble is that now we get a ton of these:
```
06/23/2021 12:15:31 - INFO - da... | 41 | switching some low-level log.info's to log.debug?
In https://github.com/huggingface/transformers/pull/12276 we are now changing the examples to have `datasets` on the same log level as `transformers`, so that one setting can do a consistent logging across all involved components.
The trouble is that now we get a t... | [
0.1725226640701294,
-0.34119701385498047,
0.087156280875206,
0.20244204998016357,
0.25608453154563904,
0.11248648911714554,
0.5520288944244385,
0.3101122975349426,
-0.07186184078454971,
-0.30187132954597473,
-0.041637394577264786,
0.17572852969169617,
-0.2178720235824585,
0.280829340219497... |
https://github.com/huggingface/datasets/issues/2542 | Hi @VictorSanh, thank you for reporting this issue with duplicated keys.
- The issue with "adversarial_qa" was fixed 23 days ago: #2433. Current version of `datasets` (1.8.0) includes the patch.
- I am investigating the issue with `drop`. I'll ping you to keep you informed. | `datasets.keyhash.DuplicatedKeysError` for `drop` and `adversarial_qa/adversarialQA` | ## Describe the bug
Failure to generate the datasets (`drop` and subset `adversarialQA` from `adversarial_qa`) because of duplicate keys.
## Steps to reproduce the bug
```python
from datasets import load_dataset
load_dataset("drop")
load_dataset("adversarial_qa", "adversarialQA")
```
## Expected results
Th... | 45 | `datasets.keyhash.DuplicatedKeysError` for `drop` and `adversarial_qa/adversarialQA`
## Describe the bug
Failure to generate the datasets (`drop` and subset `adversarialQA` from `adversarial_qa`) because of duplicate keys.
## Steps to reproduce the bug
```python
from datasets import load_dataset
load_dataset("... | [
-0.09050873667001724,
-0.11600645631551743,
0.009455308318138123,
0.470907986164093,
0.0840873196721077,
0.028658613562583923,
0.30280959606170654,
0.23976539075374603,
0.12779703736305237,
0.181435227394104,
-0.1439216434955597,
0.5057470798492432,
-0.04829590022563934,
0.1198931038379669... |
https://github.com/huggingface/datasets/issues/2542 | Hi @VictorSanh, the issue is already fixed and merged into master branch and will be included in our next release version 1.9.0. | `datasets.keyhash.DuplicatedKeysError` for `drop` and `adversarial_qa/adversarialQA` | ## Describe the bug
Failure to generate the datasets (`drop` and subset `adversarialQA` from `adversarial_qa`) because of duplicate keys.
## Steps to reproduce the bug
```python
from datasets import load_dataset
load_dataset("drop")
load_dataset("adversarial_qa", "adversarialQA")
```
## Expected results
Th... | 22 | `datasets.keyhash.DuplicatedKeysError` for `drop` and `adversarial_qa/adversarialQA`
## Describe the bug
Failure to generate the datasets (`drop` and subset `adversarialQA` from `adversarial_qa`) because of duplicate keys.
## Steps to reproduce the bug
```python
from datasets import load_dataset
load_dataset("... | [
-0.09050873667001724,
-0.11600645631551743,
0.009455308318138123,
0.470907986164093,
0.0840873196721077,
0.028658613562583923,
0.30280959606170654,
0.23976539075374603,
0.12779703736305237,
0.181435227394104,
-0.1439216434955597,
0.5057470798492432,
-0.04829590022563934,
0.1198931038379669... |
https://github.com/huggingface/datasets/issues/2538 | Hi ! `load_dataset` downloads the full dataset once and caches it, so that subsequent calls to `load_dataset` just reloads the dataset from your disk.
Then when you specify a `split` in `load_dataset`, it will just load the requested split from the disk. If your specified split is a sliced split (e.g. `"train[:10]"`),... | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 98 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.367188960313797,
-0.14316871762275696,
-0.02160826325416565,
0.36220449209213257,
-0.06753025949001312,
0.27566948533058167,
0.5740379691123962,
0.41680413484573364,
0.2612757682800293,
-0.08351046591997147,
-0.08284446597099304,
0.14032073318958282,
-0.17392076551914215,
0.182089120149... |
https://github.com/huggingface/datasets/issues/2538 | Hi @reachtarunhere.
Besides the above insights provided by @lhoestq and @thomwolf, there is also a Dataset feature in progress (I plan to finish it this week): #2249, which will allow you, when calling `load_dataset`, to pass the option to download/preprocess/cache only some specific split(s), which will definitely ... | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 71 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.42554810643196106,
-0.10989359021186829,
-0.03485330194234848,
0.3032791018486023,
-0.070411816239357,
0.28239279985427856,
0.5759148597717285,
0.43113768100738525,
0.30114299058914185,
-0.08975893259048462,
-0.09464352577924728,
0.1700282096862793,
-0.167522132396698,
0.238328844308853... |
https://github.com/huggingface/datasets/issues/2538 | Thanks all for responding.
Hey @albertvillanova
Thanks. Yes, I would be interested.
@lhoestq I think even if a small split is specified it loads up the full dataset from the disk (please correct me if this is not the case). Because it does seem to be slow to me even on subsequent calls. There is no repeated d... | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 85 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.4159126877784729,
-0.14814624190330505,
-0.012523014098405838,
0.3571510910987854,
-0.09925118088722229,
0.2774942219257355,
0.5004060864448547,
0.4125298261642456,
0.30492353439331055,
-0.1691027283668518,
-0.09666959941387177,
0.12072671949863434,
-0.11352957785129547,
0.2518863379955... |
https://github.com/huggingface/datasets/issues/2538 | If I want to create a dataset, containing only the 10 elements of a given dataset (slice it), how do I do that? | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 23 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.39776554703712463,
-0.09477035701274872,
-0.037197716534137726,
0.28004902601242065,
-0.05208887159824371,
0.2643589377403259,
0.5830543041229248,
0.36525142192840576,
0.26721981167793274,
-0.04507526755332947,
-0.11924267560243607,
0.20403513312339783,
-0.24334481358528137,
0.330198645... |
https://github.com/huggingface/datasets/issues/2538 |
> ```python
> small_ds = ds.select(range(10))
> ```
Thanks, but this doesn't help me to save time during initial loading, right? | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 21 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.38979572057724,
-0.13344919681549072,
-0.043076299130916595,
0.2874338924884796,
-0.06734947115182877,
0.3263415992259979,
0.5881058573722839,
0.4134097993373871,
0.32278165221214294,
-0.10512988269329071,
-0.12785908579826355,
0.14691978693008423,
-0.2003723382949829,
0.295417904853820... |
https://github.com/huggingface/datasets/issues/2538 | Indeed by default load_dataset would download and prepare everything as Arrow files. And passing `split=train[:10]` memory maps only the beginning of the full dataset that has been prepared on disk.
If you don't want to download everything, you can use streaming :
```python
ids = load_dataset(..., streaming=True... | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 67 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.4390909671783447,
-0.12437446415424347,
-0.00954730436205864,
0.3589227497577667,
-0.059830423444509506,
0.3549805283546448,
0.5410900115966797,
0.45107051730155945,
0.31206005811691284,
-0.08555978536605835,
-0.09199511259794235,
0.1409197598695755,
-0.20304223895072937,
0.174974560737... |
https://github.com/huggingface/datasets/issues/2538 | Thanks @lhoestq, but I don't think it is 100% accurate, as it doesn't keep the dataset structure exactly the same.
To load the full dataset, I do:
```
data = load_dataset("json", data_files="a.json")
train_data = data["train"].shuffle()
```
But when I am changing it as per your instructions:
```
ids = load_d... | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 82 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.32418400049209595,
-0.06388327479362488,
-0.00794219970703125,
0.4085160195827484,
0.016670942306518555,
0.3319082260131836,
0.5023548603057861,
0.3716801106929779,
0.15747973322868347,
0.013587478548288345,
-0.14452959597110748,
0.26646891236305237,
-0.20985513925552368,
0.246435850858... |
https://github.com/huggingface/datasets/issues/2538 | Whoops I fixed my code sorry
```diff
- ds = Dataset.from_generator(ids["train"].take(10).__iter__)
+ ds = Dataset.from_generator(ids.take(10).__iter__)
```
in your case that means running
```python
train_data = data.shuffle()
```
without `["train"]` | Loading partial dataset when debugging | I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing as per the other issues.
Is there a wa... | 29 | Loading partial dataset when debugging
I am using PyTorch Lightning along with datasets (thanks for so many datasets already prepared and the great splits).
Every time I execute load_dataset for the imdb dataset it takes some time even if I specify a split involving very few samples. I guess this due to hashing ... | [
-0.3739755153656006,
-0.10816185921430588,
-0.02359113283455372,
0.35849159955978394,
-0.0022206343710422516,
0.3494642674922943,
0.5607236623764038,
0.39705127477645874,
0.2571815848350525,
-0.07330752909183502,
-0.08933363109827042,
0.15145504474639893,
-0.2315284013748169,
0.20891930162... |
https://github.com/huggingface/datasets/issues/2532 | Hi @jerryIsHere, thanks for reporting the issue. But are you sure this is a bug in HuggingFace **Datasets**? | Tokenizer's normalization preprocessor cause misalignment in return_offsets_mapping for tokenizer classification task | [This colab notebook](https://colab.research.google.com/drive/151gKyo0YIwnlznrOHst23oYH_a3mAe3Z?usp=sharing) implements a token classification input pipeline extending the logic from [this hugging example](https://huggingface.co/transformers/custom_datasets.html#tok-ner).
The pipeline works fine with most instance i... | 18 | Tokenizer's normalization preprocessor cause misalignment in return_offsets_mapping for tokenizer classification task
[This colab notebook](https://colab.research.google.com/drive/151gKyo0YIwnlznrOHst23oYH_a3mAe3Z?usp=sharing) implements a token classification input pipeline extending the logic from [this hugging exa... | [
-0.15369659662246704,
0.07901319861412048,
0.07378184050321579,
-0.03823183476924896,
0.0921575278043747,
-0.32468128204345703,
-0.08998161554336548,
0.10603179037570953,
-0.46176066994667053,
0.19867399334907532,
-0.20853736996650696,
0.30742061138153076,
0.2766766846179962,
0.03088505566... |
https://github.com/huggingface/datasets/issues/2532 | > Hi @jerryIsHere, thanks for reporting the issue. But are you sure this is a bug in HuggingFace **Datasets**?
Oh, I am sorry
I would reopen the post on huggingface/transformers | Tokenizer's normalization preprocessor cause misalignment in return_offsets_mapping for tokenizer classification task | [This colab notebook](https://colab.research.google.com/drive/151gKyo0YIwnlznrOHst23oYH_a3mAe3Z?usp=sharing) implements a token classification input pipeline extending the logic from [this hugging example](https://huggingface.co/transformers/custom_datasets.html#tok-ner).
The pipeline works fine with most instance i... | 30 | Tokenizer's normalization preprocessor cause misalignment in return_offsets_mapping for tokenizer classification task
[This colab notebook](https://colab.research.google.com/drive/151gKyo0YIwnlznrOHst23oYH_a3mAe3Z?usp=sharing) implements a token classification input pipeline extending the logic from [this hugging exa... | [
-0.15369659662246704,
0.07901319861412048,
0.07378184050321579,
-0.03823183476924896,
0.0921575278043747,
-0.32468128204345703,
-0.08998161554336548,
0.10603179037570953,
-0.46176066994667053,
0.19867399334907532,
-0.20853736996650696,
0.30742061138153076,
0.2766766846179962,
0.03088505566... |
https://github.com/huggingface/datasets/issues/2526 | I'm currently adding it, the entire dataset is quite big around 30 GB so I add splits separately. You can take a look here https://huggingface.co/datasets/merve/coco | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 25 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.21530312299728394,
-0.4916902780532837,
-0.13700996339321136,
0.21178194880485535,
0.1525026559829712,
0.036250218749046326,
0.12675932049751282,
0.11922033131122589,
-0.11205749213695526,
0.21914547681808472,
-0.532427191734314,
0.14771439135074615,
-0.26285991072654724,
0.454018682241... |
https://github.com/huggingface/datasets/issues/2526 | I talked to @lhoestq and it's best if I download this dataset through TensorFlow datasets instead, so I'll be implementing that one really soon.
@NielsRogge | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 25 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.2278994470834732,
-0.5181816816329956,
-0.1593758761882782,
0.1203790158033371,
0.17852002382278442,
0.004588179290294647,
0.19410720467567444,
0.1587904989719391,
-0.0621921569108963,
0.258683443069458,
-0.40735116600990295,
0.14291594922542572,
-0.24881844222545624,
0.3926752209663391... |
https://github.com/huggingface/datasets/issues/2526 | I started adding COCO, will be done tomorrow EOD
my work so far https://github.com/merveenoyan/datasets (my fork) | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 16 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.18309727311134338,
-0.4954967498779297,
-0.12936384975910187,
0.1958058476448059,
0.20310907065868378,
0.05224277824163437,
0.06116677075624466,
0.10782872140407562,
-0.1408272683620453,
0.21529364585876465,
-0.5061156749725342,
0.16042372584342957,
-0.22876758873462677,
0.4129449725151... |
https://github.com/huggingface/datasets/issues/2526 | Hi Merve @merveenoyan , thank you so much for your great contribution! May I ask about the current progress of your implementation? Cuz I see the pull request is still in progess here. Or can I just run the COCO scripts in your fork repo? | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 45 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.14818283915519714,
-0.6077488660812378,
-0.11354765295982361,
0.14933542907238007,
0.13040339946746826,
-0.06740139424800873,
0.06796041131019592,
0.0632958859205246,
-0.15467938780784607,
0.17484435439109802,
-0.4082760810852051,
0.1341961771249771,
-0.3136062026023865,
0.4300083816051... |
https://github.com/huggingface/datasets/issues/2526 | Hello @yixuanren I had another prioritized project about to be merged, but I'll start continuing today will finish up soon. | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 20 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.20727208256721497,
-0.5107594132423401,
-0.13226750493049622,
0.1920369267463684,
0.20458590984344482,
0.050970710813999176,
0.12114423513412476,
0.11549240350723267,
-0.09616454690694809,
0.22603656351566315,
-0.48322466015815735,
0.15590709447860718,
-0.25631478428840637,
0.4645496308... |
https://github.com/huggingface/datasets/issues/2526 | > Hello @yixuanren I had another prioritized project about to be merged, but I'll start continuing today will finish up soon.
It's really nice of you!! I see you've commited another version just now | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 34 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.1934700310230255,
-0.530968189239502,
-0.140234112739563,
0.179476797580719,
0.20484621822834015,
0.025512047111988068,
0.1151224821805954,
0.11923915892839432,
-0.11042283475399017,
0.21517851948738098,
-0.463005393743515,
0.16055312752723694,
-0.23001518845558167,
0.451765239238739,
... |
https://github.com/huggingface/datasets/issues/2526 | Hi @NielsRogge and @merveenoyan, did you find a way to load a dataset with COCO annotations to HF's hub?
I have a panoptic segmentation dataset in COCO format and would like to share it with the community.
Thanks in advance :) | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 41 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.11135811358690262,
-0.598952054977417,
-0.019046641886234283,
0.16975998878479004,
0.16823291778564453,
0.14348392188549042,
0.07303372025489807,
0.03035975992679596,
-0.02808361127972603,
0.15712139010429382,
-0.8349184989929199,
0.13507364690303802,
-0.1977088898420334,
0.382633060216... |
https://github.com/huggingface/datasets/issues/2526 | The COCO format is not supported out of the box in the HF's hub - you'd need to reformat it to an [ImageFolder](https://huggingface.co/docs/datasets/image_dataset#imagefolder) with metadata format, or write a [loading script](https://huggingface.co/docs/datasets/image_dataset#loading-script) | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 31 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.2754530608654022,
-0.5610964298248291,
-0.06040322035551071,
0.11809279024600983,
0.0983714833855629,
0.11320105940103531,
0.10179605334997177,
0.20642125606536865,
-0.022832714021205902,
0.21535828709602356,
-0.6720103025436401,
-0.008999485522508621,
-0.2272336632013321,
0.38644409179... |
https://github.com/huggingface/datasets/issues/2526 | > The COCO format is not supported out of the box in the HF's hub - you'd need to reformat it to an [ImageFolder](https://huggingface.co/docs/datasets/image_dataset#imagefolder) with metadata format, or write a [loading script](https://huggingface.co/docs/datasets/image_dataset#loading-script)
Hi @lhoestq , thank yo... | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 163 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
0.0031578317284584045,
-0.4333893656730652,
-0.008030422031879425,
0.3482617139816284,
0.1429239809513092,
0.09797373414039612,
-0.1824260652065277,
-0.12278909981250763,
-0.2389095574617386,
-0.0353531539440155,
-0.6616313457489014,
0.27105629444122314,
-0.2954804599285126,
0.152556240558... |
https://github.com/huggingface/datasets/issues/2526 | Both were uploaded using `ds.push_to_hub()` :)
You can get a Dataset from a python dictionary using `ds = Dataset.from_dict(...)` and casts the paths to images to the `Image()` type using `ds = ds.cast_column("image", Image())`.
```python
from datasets import Dataset, Image
ds = Dataset.from_dict(...)
ds = d... | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 53 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.16209599375724792,
-0.5183026790618896,
-0.10980439186096191,
0.2343139350414276,
0.15737968683242798,
0.0431944876909256,
0.17247439920902252,
0.12560522556304932,
-0.05706825852394104,
0.21617281436920166,
-0.5137798190116882,
0.2342437207698822,
-0.2883448302745819,
0.374885052442550... |
https://github.com/huggingface/datasets/issues/2526 | > Both were uploaded using `ds.push_to_hub()` :)
>
> You can get a Dataset from a python dictionary using `ds = Dataset.from_dict(...)` and casts the paths to images to the `Image()` type using `ds = ds.cast_column("image", Image())`.
>
> ```python
> from datasets import Dataset, Image
>
> ds = Dataset.from_d... | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 456 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.05919724330306053,
-0.4270419776439667,
-0.025775648653507233,
0.23544448614120483,
0.1592135727405548,
0.0224774032831192,
0.15408384799957275,
0.197895348072052,
-0.25009244680404663,
0.16166016459465027,
-0.5740943551063538,
0.024561066180467606,
-0.10765625536441803,
0.2952242493629... |
https://github.com/huggingface/datasets/issues/2526 | You can pass this data in .from_dict() - no need to cast anything for this column
```python
ds = Dataset.from_dict({
"image": [...],
"label": [...],
"segments_info": [...],
)}
```
where `segments_info` is the list of the segment_infos of all the examples in the dataset, and therefore is a list o... | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 52 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.19400638341903687,
-0.5657399296760559,
-0.12166833132505417,
0.22642728686332703,
0.16179406642913818,
0.06718018651008606,
0.19251829385757446,
0.1666170358657837,
-0.023890066891908646,
0.17521804571151733,
-0.5087730288505554,
0.23101961612701416,
-0.2850366532802582,
0.368548363447... |
https://github.com/huggingface/datasets/issues/2526 | > You can pass this data in .from_dict() - no need to cast anything for this column
>
> ```python
> ds = Dataset.from_dict({
> "image": [...],
> "label": [...],
> "segments_info": [...],
> )}
> ```
>
> where `segments_info` is the list of the segment_infos of all the examples in the dataset, and... | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 89 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.021371185779571533,
-0.5979077219963074,
-0.1236754059791565,
0.3704049587249756,
0.23139257729053497,
0.1123054027557373,
0.05855163559317589,
0.08938145637512207,
-0.043444279581308365,
0.07443318516016006,
-0.5112690925598145,
0.42554619908332825,
-0.24832186102867126,
0.293189644813... |
https://github.com/huggingface/datasets/issues/2526 | You read the JSON file and transform the data yourself. I don't think there's an automatic converter somewhere | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 18 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.1299549788236618,
-0.3851433992385864,
-0.11860890686511993,
0.19418677687644958,
0.20388300716876984,
0.025883540511131287,
0.0030269185081124306,
0.181259885430336,
-0.0772172212600708,
0.20925650000572205,
-0.4467281401157379,
0.2712952792644501,
-0.27274614572525024,
0.4661293029785... |
https://github.com/huggingface/datasets/issues/2526 | > You read the JSON file and transform the data yourself. I don't think there's an automatic converter somewhere
Perfect, I've done it and succesfully uploaded a new dataset [here](https://huggingface.co/datasets/lombardata/panoptic_2023_06_22), but I've (I hope) a last problem.
The dataset has currently 302 images... | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 141 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
0.061464570462703705,
-0.47556155920028687,
-0.033519625663757324,
0.4929114878177643,
0.19361799955368042,
0.18587136268615723,
-0.010941646993160248,
0.27351805567741394,
-0.16582933068275452,
0.24338729679584503,
-0.2654048502445221,
-0.026507502421736717,
-0.05747365206480026,
0.306972... |
https://github.com/huggingface/datasets/issues/2526 | Awesome ! Your dataset looks all good 🤗
The `null` in the viewer is a bug on our side, let me investigate | Add COCO datasets | ## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in HuggingFace datasets... | 22 | Add COCO datasets
## Adding a Dataset
- **Name:** COCO
- **Description:** COCO is a large-scale object detection, segmentation, and captioning dataset.
- **Paper + website:** https://cocodataset.org/#home
- **Data:** https://cocodataset.org/#download
- **Motivation:** It would be great to have COCO available in ... | [
-0.27886953949928284,
-0.4296073317527771,
-0.05939604714512825,
0.444915235042572,
0.1920853555202484,
0.03030010312795639,
0.22380408644676208,
0.19492442905902863,
-0.09795382618904114,
0.22592610120773315,
-0.34727656841278076,
0.14553657174110413,
-0.28121069073677063,
0.2961997985839... |
https://github.com/huggingface/datasets/issues/2522 | Hi,
this issue has been already reported in the dataset repo (https://github.com/dair-ai/emotion_dataset/issues/2), so this is a bug on their side. | Documentation Mistakes in Dataset: emotion | As per documentation,
Dataset: emotion
Homepage: https://github.com/dair-ai/emotion_dataset
Dataset: https://github.com/huggingface/datasets/blob/master/datasets/emotion/emotion.py
Permalink: https://huggingface.co/datasets/viewer/?dataset=emotion
Emotion is a dataset of English Twitter messages with eight b... | 20 | Documentation Mistakes in Dataset: emotion
As per documentation,
Dataset: emotion
Homepage: https://github.com/dair-ai/emotion_dataset
Dataset: https://github.com/huggingface/datasets/blob/master/datasets/emotion/emotion.py
Permalink: https://huggingface.co/datasets/viewer/?dataset=emotion
Emotion is a dat... | [
0.22127103805541992,
-0.3492066264152527,
-0.10875460505485535,
0.5762393474578857,
0.2677905559539795,
0.17377620935440063,
0.294209361076355,
0.11103951930999756,
-0.12832501530647278,
0.20382319390773773,
-0.12744291126728058,
0.04295263811945915,
-0.20233356952667236,
-0.11315067112445... |
https://github.com/huggingface/datasets/issues/2522 | The documentation has another bug in the dataset card [here](https://huggingface.co/datasets/emotion).
In the dataset summary **six** emotions are mentioned: *"six basic emotions: anger, fear, joy, love, sadness, and surprise"*, however, in the datafields section we have only **five**:
```
label: a classification... | Documentation Mistakes in Dataset: emotion | As per documentation,
Dataset: emotion
Homepage: https://github.com/dair-ai/emotion_dataset
Dataset: https://github.com/huggingface/datasets/blob/master/datasets/emotion/emotion.py
Permalink: https://huggingface.co/datasets/viewer/?dataset=emotion
Emotion is a dataset of English Twitter messages with eight b... | 57 | Documentation Mistakes in Dataset: emotion
As per documentation,
Dataset: emotion
Homepage: https://github.com/dair-ai/emotion_dataset
Dataset: https://github.com/huggingface/datasets/blob/master/datasets/emotion/emotion.py
Permalink: https://huggingface.co/datasets/viewer/?dataset=emotion
Emotion is a dat... | [
0.22789007425308228,
-0.3761104345321655,
-0.06796948611736298,
0.5594586730003357,
0.20203638076782227,
0.2261340618133545,
0.1969553530216217,
0.022524066269397736,
0.0012623891234397888,
0.1313016563653946,
-0.06918162107467651,
0.011118803173303604,
-0.2228650152683258,
-0.050262227654... |
https://github.com/huggingface/datasets/issues/2522 | @GDGauravDutta the dataset author replied in their issue: https://github.com/dair-ai/emotion_dataset/issues/2
> The dataset released is a preprocessed six emotion variant released for educational and research purposes.
@albertovilla the dataset card was fixed with 6 emotions. | Documentation Mistakes in Dataset: emotion | As per documentation,
Dataset: emotion
Homepage: https://github.com/dair-ai/emotion_dataset
Dataset: https://github.com/huggingface/datasets/blob/master/datasets/emotion/emotion.py
Permalink: https://huggingface.co/datasets/viewer/?dataset=emotion
Emotion is a dataset of English Twitter messages with eight b... | 34 | Documentation Mistakes in Dataset: emotion
As per documentation,
Dataset: emotion
Homepage: https://github.com/dair-ai/emotion_dataset
Dataset: https://github.com/huggingface/datasets/blob/master/datasets/emotion/emotion.py
Permalink: https://huggingface.co/datasets/viewer/?dataset=emotion
Emotion is a dat... | [
0.16476678848266602,
-0.36794978380203247,
-0.14471957087516785,
0.5058468580245972,
0.20540262758731842,
0.10941104590892792,
0.2985422909259796,
0.09094996750354767,
-0.07220041751861572,
0.18671053647994995,
-0.11138271540403366,
0.05344632267951965,
-0.19112031161785126,
-0.10717709362... |
https://github.com/huggingface/datasets/issues/2520 | The `task_templates` API is deprecated in favor of the `train-eval-index` YAML field, so I'm closing this issue. | Datasets with tricky task templates | I'm collecting a list of datasets here that don't follow the "standard" taxonomy and require further investigation to implement task templates for.
## Text classification
* [hatexplain](https://huggingface.co/datasets/hatexplain): ostensibly a form of text classification, but not in the standard `(text, target)` ... | 17 | Datasets with tricky task templates
I'm collecting a list of datasets here that don't follow the "standard" taxonomy and require further investigation to implement task templates for.
## Text classification
* [hatexplain](https://huggingface.co/datasets/hatexplain): ostensibly a form of text classification, but... | [
-0.03939249739050865,
-0.3244529366493225,
-0.036528486758470535,
0.11130198836326599,
-0.0356159582734108,
0.13298724591732025,
0.28443917632102966,
0.1137927696108818,
0.08774848282337189,
-0.2107875943183899,
0.11740924417972565,
0.29740259051322937,
-0.3008587956428528,
0.2331094890832... |
https://github.com/huggingface/datasets/issues/2516 | Hi ! `map` calls `__getstate__` using `dill` to hash your map function. This is used by the caching mechanism to recover previously computed results. That's why you don't see any `__setstate__` call.
Why do you change an attribute of your tokenizer when `__getstate__` is called ? | datasets.map pickle issue resulting in invalid mapping function | I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it's correct but it fails when I use it inside a function which is m... | 46 | datasets.map pickle issue resulting in invalid mapping function
I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it... | [
-0.22488752007484436,
0.1931205689907074,
0.1127529963850975,
0.02877509593963623,
0.034195929765701294,
-0.3237702250480652,
0.13853655755519867,
0.19508479535579681,
0.3468087911605835,
-0.1000182032585144,
0.19218169152736664,
0.725385308265686,
-0.19296038150787354,
0.20866991579532623... |
https://github.com/huggingface/datasets/issues/2516 | @lhoestq because if I try to pickle my custom tokenizer (it contains a pure python pretokenization step in an otherwise rust backed tokenizer) I get
> Exception: Error while attempting to pickle Tokenizer: Custom PreTokenizer cannot be serialized
So I remove the Custom PreTokenizer in `__getstate__` and then rest... | datasets.map pickle issue resulting in invalid mapping function | I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it's correct but it fails when I use it inside a function which is m... | 121 | datasets.map pickle issue resulting in invalid mapping function
I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it... | [
-0.22488752007484436,
0.1931205689907074,
0.1127529963850975,
0.02877509593963623,
0.034195929765701294,
-0.3237702250480652,
0.13853655755519867,
0.19508479535579681,
0.3468087911605835,
-0.1000182032585144,
0.19218169152736664,
0.725385308265686,
-0.19296038150787354,
0.20866991579532623... |
https://github.com/huggingface/datasets/issues/2516 | Actually, maybe I need to deep copy `self.__dict__`? That way `self` isn't modified. That was my intention and I thought it was working - I'll double-check after the weekend. | datasets.map pickle issue resulting in invalid mapping function | I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it's correct but it fails when I use it inside a function which is m... | 29 | datasets.map pickle issue resulting in invalid mapping function
I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it... | [
-0.22488752007484436,
0.1931205689907074,
0.1127529963850975,
0.02877509593963623,
0.034195929765701294,
-0.3237702250480652,
0.13853655755519867,
0.19508479535579681,
0.3468087911605835,
-0.1000182032585144,
0.19218169152736664,
0.725385308265686,
-0.19296038150787354,
0.20866991579532623... |
https://github.com/huggingface/datasets/issues/2516 | Doing a deep copy results in the warning:
> 06/20/2021 16:02:15 - WARNING - datasets.fingerprint - Parameter 'function'=<function tokenize_function at 0x7f1e95f05d40> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. Make sure your transforms a... | datasets.map pickle issue resulting in invalid mapping function | I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it's correct but it fails when I use it inside a function which is m... | 114 | datasets.map pickle issue resulting in invalid mapping function
I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it... | [
-0.22488752007484436,
0.1931205689907074,
0.1127529963850975,
0.02877509593963623,
0.034195929765701294,
-0.3237702250480652,
0.13853655755519867,
0.19508479535579681,
0.3468087911605835,
-0.1000182032585144,
0.19218169152736664,
0.725385308265686,
-0.19296038150787354,
0.20866991579532623... |
https://github.com/huggingface/datasets/issues/2516 | Looks like there is still an object that is not pickable in your `tokenize_function` function.
You can test if an object can be pickled and hashed by using
```python
from datasets.fingerprint import Hasher
Hasher.hash(my_object)
```
Under the hood it pickles the object to compute its hash, so it calls `__g... | datasets.map pickle issue resulting in invalid mapping function | I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it's correct but it fails when I use it inside a function which is m... | 52 | datasets.map pickle issue resulting in invalid mapping function
I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it... | [
-0.22488752007484436,
0.1931205689907074,
0.1127529963850975,
0.02877509593963623,
0.034195929765701294,
-0.3237702250480652,
0.13853655755519867,
0.19508479535579681,
0.3468087911605835,
-0.1000182032585144,
0.19218169152736664,
0.725385308265686,
-0.19296038150787354,
0.20866991579532623... |
https://github.com/huggingface/datasets/issues/2516 | I figured it out, the problem is deep copy itself uses pickle (unless you implement `__deepcopy__`). So when I changed `__getstate__` it started throwing an error.
I'm sure there's a better way of doing this, but in order to return the `__dict__` without the non-pikelable pre-tokeniser and without modifying self I r... | datasets.map pickle issue resulting in invalid mapping function | I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it's correct but it fails when I use it inside a function which is m... | 126 | datasets.map pickle issue resulting in invalid mapping function
I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it... | [
-0.22488752007484436,
0.1931205689907074,
0.1127529963850975,
0.02877509593963623,
0.034195929765701294,
-0.3237702250480652,
0.13853655755519867,
0.19508479535579681,
0.3468087911605835,
-0.1000182032585144,
0.19218169152736664,
0.725385308265686,
-0.19296038150787354,
0.20866991579532623... |
https://github.com/huggingface/datasets/issues/2516 | I'm glad you figured something out :)
Regarding hashing: we're not using hashing for the same purpose as the python `__hash__` purpose (which is in general for dictionary lookups). For example it is allowed for python hashing to not return the same hash across sessions, while our hashing must return the same hashes ... | datasets.map pickle issue resulting in invalid mapping function | I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it's correct but it fails when I use it inside a function which is m... | 61 | datasets.map pickle issue resulting in invalid mapping function
I trained my own tokenizer, and I needed to use a python custom class. Because of this I have to detach the custom step before saving and reattach after restore. I did this using the standard pickle `__get_state__` / `__set_state__` mechanism. I think it... | [
-0.22488752007484436,
0.1931205689907074,
0.1127529963850975,
0.02877509593963623,
0.034195929765701294,
-0.3237702250480652,
0.13853655755519867,
0.19508479535579681,
0.3468087911605835,
-0.1000182032585144,
0.19218169152736664,
0.725385308265686,
-0.19296038150787354,
0.20866991579532623... |
https://github.com/huggingface/datasets/issues/2514 | Hi ! For now this is probably the best option.
We might add a feature like this in the feature as well.
Do you know any deduplication method that works on arbitrary big datasets without filling up RAM ?
Otherwise we can have do the deduplication in memory like pandas but I feel like this is going to be limiting fo... | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 63 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.14068497717380524,
-0.1654823273420334,
-0.16508832573890686,
0.15400491654872894,
0.04894325137138367,
0.2394908219575882,
0.2769090533256531,
0.06343920528888702,
-0.3376512825489044,
0.019205082207918167,
-0.034110575914382935,
0.23469814658164978,
0.11909039318561554,
0.2062142640352... |
https://github.com/huggingface/datasets/issues/2514 | Yes, I'd like to work on this feature once I'm done with #2500, but first I have to do some research, and see if the implementation wouldn't be too complex.
In the meantime, maybe [this lib](https://github.com/TomScheffers/pyarrow_ops) can help. However, note that this lib operates directly on pyarrow tables and rel... | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 80 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.10441111773252487,
0.01838865876197815,
-0.13539132475852966,
0.09495442360639572,
0.033134885132312775,
0.14071276783943176,
0.3450477123260498,
0.2536356449127197,
-0.6958326697349548,
0.0018976181745529175,
-0.09365198016166687,
0.37214305996894836,
0.03419318422675133,
0.165043324232... |
https://github.com/huggingface/datasets/issues/2514 | > Hi ! For now this is probably the best option.
> We might add a feature like this in the feature as well.
>
> Do you know any deduplication method that works on arbitrary big datasets without filling up RAM ?
> Otherwise we can have do the deduplication in memory like pandas but I feel like this is going to be l... | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 119 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.10129868239164352,
-0.19105321168899536,
-0.13927015662193298,
0.28647005558013916,
0.0751885175704956,
0.2036612331867218,
0.27155786752700806,
0.09866943210363388,
-0.33350056409835815,
-0.01633448153734207,
-0.0008100755512714386,
0.2826298177242279,
0.04018646106123924,
0.19777157902... |
https://github.com/huggingface/datasets/issues/2514 | Hello,
I'm also interested in this feature.
Has there been progress on this issue?
Could we use a similar trick as above, but with a better hashing algorithm like SHA?
We could also use a [bloom filter](https://en.wikipedia.org/wiki/Bloom_filter), should we care a lot about collision in this case? | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 47 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.20651397109031677,
-0.060459405183792114,
-0.14436282217502594,
0.16920100152492523,
0.034545958042144775,
0.20407815277576447,
0.4904038906097412,
0.1726582944393158,
-0.3908334970474243,
-0.07263363152742386,
0.012869387865066528,
0.17522726953029633,
-0.006163710728287697,
0.258381009... |
https://github.com/huggingface/datasets/issues/2514 | For reference, we can get a solution fairly easily if we assume that we can hold in memory all unique values.
```python
from datasets import Dataset
from itertools import cycle
from functools import partial
memory = set()
def is_unique(elem:Any , column: str, memory: set) -> bool:
if elem[column] in mem... | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 117 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.09119115024805069,
-0.16821198165416718,
-0.11286728084087372,
0.1837807595729828,
-0.0014151688665151596,
0.3015866279602051,
0.3887415826320648,
0.017941679805517197,
-0.30535268783569336,
-0.08604380488395691,
-0.14281922578811646,
0.25312894582748413,
-0.062169551849365234,
0.1245175... |
https://github.com/huggingface/datasets/issues/2514 | An approach that works assuming you can hold the all the unique document hashes in memory:
```python
from datasets import load_dataset
def get_hash(example):
"""Get hash of content field."""
return {"hash": hash(example["content"])} # can use any hashing function here
def check_uniques(example, un... | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 105 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.12614496052265167,
-0.04981991648674011,
-0.10539953410625458,
0.2112497240304947,
0.14171594381332397,
0.37468910217285156,
0.32474109530448914,
0.19652754068374634,
-0.2755315899848938,
-0.18274490535259247,
-0.042024943977594376,
0.31486138701438904,
-0.15311168134212494,
0.0484706833... |
https://github.com/huggingface/datasets/issues/2514 | @lvwerra hey, could you tell me how reliable is this deduplication method. i am currently using the same deduplication strategy to deduplicate a large text corpus to pretrain LLMs ~ 11B to 20B. just needed to ensure if this strategy would be fine on large datasets for LLMs pretraining. | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 49 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.043261416256427765,
-0.16642586886882782,
-0.08219671249389648,
0.2686764895915985,
0.00791722722351551,
0.2554418742656708,
0.3373904526233673,
0.0805903971195221,
-0.4248509407043457,
-0.06280083954334259,
-0.031623318791389465,
0.2793353497982025,
-0.036746736615896225,
0.052438087761... |
https://github.com/huggingface/datasets/issues/2514 | Hi @StephennFernandes I'm also trying to pretrain an llm, and need to do deduplication for my dataset,
which method you applied please? | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 22 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.10408513993024826,
-0.1962968409061432,
-0.09547524154186249,
0.1945139616727829,
-0.01837163232266903,
0.28303074836730957,
0.390622079372406,
0.011555496603250504,
-0.41531312465667725,
-0.0391712412238121,
0.020090743899345398,
0.26913005113601685,
-0.03262265771627426,
0.059373810887... |
https://github.com/huggingface/datasets/issues/2514 | Hey @Manel-Hik
The following is a simpler yet really effective deduplication code that i has used in the past.
given that I have limited training corpus for the languages I wanted to train i relied on this code. https://huggingface.co/datasets/Finnish-NLP/mc4_fi_cleaned/blob/main/deduplicate.py
for more robust an... | Can datasets remove duplicated rows? | **Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
**Describe the solution you'd like*... | 54 | Can datasets remove duplicated rows?
**Is your feature request related to a problem? Please describe.**
i find myself more and more relying on datasets just to do all the preprocessing. One thing however, for removing duplicated rows, I couldn't find out how and am always converting datasets to pandas to do that..
... | [
0.1537148505449295,
-0.09739527851343155,
-0.15554805099964142,
0.17679104208946228,
0.034648459404706955,
0.20085906982421875,
0.37883663177490234,
0.09440972656011581,
-0.3979538083076477,
-0.06671015918254852,
-0.05599019676446915,
0.320603609085083,
0.10343741625547409,
0.2095335125923... |
https://github.com/huggingface/datasets/issues/2511 | Update on this: I'm computing the checksums of the data files. It will be available soon | Add C4 | ## Adding a Dataset
- **Name:** *C4*
- **Description:** *https://github.com/allenai/allennlp/discussions/5056*
- **Paper:** *https://arxiv.org/abs/1910.10683*
- **Data:** *https://huggingface.co/datasets/allenai/c4*
- **Motivation:** *Used a lot for pretraining*
Instructions to add a new dataset can be found [h... | 16 | Add C4
## Adding a Dataset
- **Name:** *C4*
- **Description:** *https://github.com/allenai/allennlp/discussions/5056*
- **Paper:** *https://arxiv.org/abs/1910.10683*
- **Data:** *https://huggingface.co/datasets/allenai/c4*
- **Motivation:** *Used a lot for pretraining*
Instructions to add a new dataset can be... | [
-0.25861161947250366,
-0.23689307272434235,
-0.21562549471855164,
0.07441946864128113,
0.20855362713336945,
-0.08748025447130203,
0.10772189497947693,
0.3031298518180847,
0.06430117040872574,
0.3231356739997864,
-0.21327093243598938,
0.06667788326740265,
-0.1542893648147583,
0.433284401893... |
https://github.com/huggingface/datasets/issues/2508 | Hi ! Is this folder structure a standard, a bit like imagenet ?
In this case maybe we can consider having a dataset loader for cifar-like, imagenet-like, squad-like, conll-like etc. datasets ?
```python
from datasets import load_dataset
my_custom_cifar = load_dataset("cifar_like", data_dir="path/to/data/dir")
``... | Load Image Classification Dataset from Local | **Is your feature request related to a problem? Please describe.**
Yes - we would like to load an image classification dataset with datasets without having to write a custom data loader.
**Describe the solution you'd like**
Given a folder structure with images of each class in each folder, the ability to load th... | 48 | Load Image Classification Dataset from Local
**Is your feature request related to a problem? Please describe.**
Yes - we would like to load an image classification dataset with datasets without having to write a custom data loader.
**Describe the solution you'd like**
Given a folder structure with images of e... | [
-0.1711522787809372,
-0.166746586561203,
0.03135290741920471,
0.4532008767127991,
0.2632547616958618,
-0.08111825585365295,
0.22850829362869263,
0.051784537732601166,
0.3194648325443268,
0.2681330442428589,
-0.19803519546985626,
0.09343153238296509,
-0.33474603295326233,
0.3943921327590942... |
https://github.com/huggingface/datasets/issues/2508 | @lhoestq I think we'll want a generic `image-folder` dataset (same as 'imagenet-like'). This is like `torchvision.datasets.ImageFolder`, and is something vision folks are used to seeing. | Load Image Classification Dataset from Local | **Is your feature request related to a problem? Please describe.**
Yes - we would like to load an image classification dataset with datasets without having to write a custom data loader.
**Describe the solution you'd like**
Given a folder structure with images of each class in each folder, the ability to load th... | 25 | Load Image Classification Dataset from Local
**Is your feature request related to a problem? Please describe.**
Yes - we would like to load an image classification dataset with datasets without having to write a custom data loader.
**Describe the solution you'd like**
Given a folder structure with images of e... | [
-0.2203829288482666,
-0.17356407642364502,
0.03799234330654144,
0.46598947048187256,
0.19796618819236755,
-0.050957344472408295,
0.1801861822605133,
0.07256574928760529,
0.311346173286438,
0.24052414298057556,
-0.1453476846218109,
0.1418924182653427,
-0.29275497794151306,
0.284378975629806... |
https://github.com/huggingface/datasets/issues/2508 | Opening this back up, since I'm planning on tackling this. Already posted a quick version of it on my account on the hub.
```python
from datasets import load_dataset
ds = load_dataset('nateraw/image-folder', data_files='PetImages/')
``` | Load Image Classification Dataset from Local | **Is your feature request related to a problem? Please describe.**
Yes - we would like to load an image classification dataset with datasets without having to write a custom data loader.
**Describe the solution you'd like**
Given a folder structure with images of each class in each folder, the ability to load th... | 33 | Load Image Classification Dataset from Local
**Is your feature request related to a problem? Please describe.**
Yes - we would like to load an image classification dataset with datasets without having to write a custom data loader.
**Describe the solution you'd like**
Given a folder structure with images of e... | [
-0.20274963974952698,
-0.1886906921863556,
0.013178244233131409,
0.4011634588241577,
0.25930503010749817,
-0.027974359691143036,
0.19759730994701385,
0.14511771500110626,
0.3119896650314331,
0.2857656478881836,
-0.1935354471206665,
0.12354758381843567,
-0.30474144220352173,
0.3270212411880... |
https://github.com/huggingface/datasets/issues/2503 | @arnaudstiegler I have just checked that these mismatches are already present in the original dataset: https://github.com/megagonlabs/SubjQA
We are going to contact the dataset owners to report this. | SubjQA wrong boolean values in entries | ## Describe the bug
SubjQA seems to have a boolean that's consistently wrong.
It defines:
- question_subj_level: The subjectiviy level of the question (on a 1 to 5 scale with 1 being the most subjective).
- is_ques_subjective: A boolean subjectivity label derived from question_subj_level (i.e., scores below 4 are... | 27 | SubjQA wrong boolean values in entries
## Describe the bug
SubjQA seems to have a boolean that's consistently wrong.
It defines:
- question_subj_level: The subjectiviy level of the question (on a 1 to 5 scale with 1 being the most subjective).
- is_ques_subjective: A boolean subjectivity label derived from ques... | [
0.12356594204902649,
0.0693044364452362,
0.027031637728214264,
0.1645200103521347,
-0.3365962505340576,
-0.006898999214172363,
0.08650708198547363,
0.04523950070142746,
-0.09279756993055344,
0.18643899261951447,
-0.09195969998836517,
0.4202954173088074,
0.21497012674808502,
0.2549099624156... |
https://github.com/huggingface/datasets/issues/2503 | I have:
- opened an issue in their repo: https://github.com/megagonlabs/SubjQA/issues/3
- written an email to all the paper authors | SubjQA wrong boolean values in entries | ## Describe the bug
SubjQA seems to have a boolean that's consistently wrong.
It defines:
- question_subj_level: The subjectiviy level of the question (on a 1 to 5 scale with 1 being the most subjective).
- is_ques_subjective: A boolean subjectivity label derived from question_subj_level (i.e., scores below 4 are... | 19 | SubjQA wrong boolean values in entries
## Describe the bug
SubjQA seems to have a boolean that's consistently wrong.
It defines:
- question_subj_level: The subjectiviy level of the question (on a 1 to 5 scale with 1 being the most subjective).
- is_ques_subjective: A boolean subjectivity label derived from ques... | [
0.17107358574867249,
0.05873541533946991,
0.04421219602227211,
0.09914503246545792,
-0.31854403018951416,
-0.10435991734266281,
0.09934559464454651,
0.024407971650362015,
-0.14664696156978607,
0.22348669171333313,
-0.06140259653329849,
0.3840121328830719,
0.2275163233280182,
0.227973610162... |
https://github.com/huggingface/datasets/issues/2499 | Thanks @VictorSanh!
There's also a [notebook](https://aka.ms/python_puzzles) and [demo](https://aka.ms/python_puzzles_study) available now to try out some of the puzzles | Python Programming Puzzles | ## Adding a Dataset
- **Name:** Python Programming Puzzles
- **Description:** Programming challenge called programming puzzles, as an objective and comprehensive evaluation of program synthesis
- **Paper:** https://arxiv.org/pdf/2106.05784.pdf
- **Data:** https://github.com/microsoft/PythonProgrammingPuzzles ([Scro... | 17 | Python Programming Puzzles
## Adding a Dataset
- **Name:** Python Programming Puzzles
- **Description:** Programming challenge called programming puzzles, as an objective and comprehensive evaluation of program synthesis
- **Paper:** https://arxiv.org/pdf/2106.05784.pdf
- **Data:** https://github.com/microsoft/P... | [
-0.06903722882270813,
-0.15612730383872986,
-0.27598676085472107,
-0.05275075137615204,
-0.0006479248404502869,
0.07016494870185852,
0.019562741741538048,
0.2562192380428314,
0.003462783992290497,
0.127641499042511,
0.03861631453037262,
0.30958127975463867,
-0.38762837648391724,
0.35757476... |
https://github.com/huggingface/datasets/issues/2498 | That’s interesting thanks, let’s see what we can do. Can you detail your last sentence? I’m not sure I understand it well. | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 22 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | Hi ! I just re-ran a quick benchmark and using `to_numpy()` seems to be faster now:
```python
import pyarrow as pa # I used pyarrow 3.0.0
import numpy as np
n, max_length = 1_000, 512
low, high, size = 0, 2 << 16, (n, max_length)
table = pa.Table.from_pydict({
"input_ids": np.random.default_rng(42).in... | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 150 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | Sounds like a plan @lhoestq If you create a PR I'll pick it up and try it out right away! | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 20 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | I’m not exactly sure how to read the graph but it seems that to_categorical take a lot of time here. Could you share more informations on the features/stats of your datasets so we could maybe design a synthetic datasets that looks more similar for debugging testing? | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 46 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | > I’m not exactly sure how to read the graph but it seems that to_categorical take a lot of time here. Could you share more informations on the features/stats of your datasets so we could maybe design a synthetic datasets that looks more similar for debugging testing?
@thomwolf starting from the top, each rectangle ... | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 140 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | @lhoestq the proposed branch is faster, but overall training speedup is a few percentage points. I couldn't figure out how to include the GitHub branch into setup.py, so I couldn't start NVidia optimized Docker-based pre-training run. But on bare metal, there is a slight improvement. I'll do some more performance trac... | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 51 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | Hi @vblagoje, to install Datasets from @lhoestq PR reference #2505, you can use:
```shell
pip install git+ssh://git@github.com/huggingface/datasets.git@refs/pull/2505/head#egg=datasets
``` | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 18 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | Hey @albertvillanova yes thank you, I am aware, I can easily pull it from a terminal command line but then I can't automate docker image builds as dependencies are picked up from setup.py and for some reason setup.py doesn't accept this string format. | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 43 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | @vblagoje in that case, you can add this to your `setup.py`:
```python
install_requires=[
"datasets @ git+ssh://git@github.com/huggingface/datasets.git@refs/pull/2505/head",
``` | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 17 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | @lhoestq @thomwolf @albertvillanova The new approach is definitely faster, dataloader now takes less than 3% cumulative time (pink rectangle two rectangles to the right of tensor.py backward invocation)
![Screen Shot 2021-06-16 at 3 05 06 PM](https://user-images.githubusercontent.com/458335/122224432-19de4700-ce82-1... | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 80 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | > ```python
> datasets @ git+ssh://git@github.com/huggingface/datasets.git@refs/pull/2505/head
> ```
@albertvillanova how would I replace datasets dependency in https://github.com/huggingface/transformers/blob/master/setup.py as the above approach is not working. | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 24 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | @vblagoje I tested my proposed approach before posting it here and it worked for me.
Is it not working in your case because of the SSH protocol? In that case you could try the same approach but using HTTPS:
```
"datasets @ git+https://github.com/huggingface/datasets.git@refs/pull/2505/head",
``` | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 44 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | @albertvillanova of course it works. Apologies. I needed to change datasets in all deps references , like [here](https://github.com/huggingface/transformers/blob/master/setup.py#L235) for example. | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 20 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2498 | Is time spent casting an issue here? See https://github.com/huggingface/datasets/issues/4676 that Datasets can spend huge amounts of time repeatedly casting to Python objects. | Improve torch formatting performance | **Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ecosystem. We use encoded HF Wikipedia an... | 22 | Improve torch formatting performance
**Is your feature request related to a problem? Please describe.**
It would be great, if possible, to further improve read performance of raw encoded datasets and their subsequent conversion to torch tensors.
A bit more background. I am working on LM pre-training using HF ec... | [
-0.35013505816459656,
-0.24029682576656342,
-0.07366813719272614,
0.1903572380542755,
0.10275399684906006,
0.14770692586898804,
0.25998902320861816,
0.6868168115615845,
-0.12113024294376373,
-0.0641198456287384,
-0.356641948223114,
0.2431408166885376,
-0.11459861695766449,
-0.0332220494747... |
https://github.com/huggingface/datasets/issues/2494 | Hi @albertvillanova, I hope you are doing well.
I am interested in this issue, is this still unresolved and open ?
The link you have provided in the above message directs to a webpage that does not exist.
Thanks and Regards | Improve docs on Enhancing performance | In the ["Enhancing performance"](https://huggingface.co/docs/datasets/loading_datasets.html#enhancing-performance) section of docs, add specific use cases:
- How to make datasets the fastest
- How to make datasets take the less RAM
- How to make datasets take the less hard drive mem
cc: @thomwolf
| 41 | Improve docs on Enhancing performance
In the ["Enhancing performance"](https://huggingface.co/docs/datasets/loading_datasets.html#enhancing-performance) section of docs, add specific use cases:
- How to make datasets the fastest
- How to make datasets take the less RAM
- How to make datasets take the less hard dri... | [
0.14003565907478333,
-0.2542183995246887,
-0.1804462969303131,
0.5229964256286621,
-0.008909366093575954,
0.21884125471115112,
-0.02222549170255661,
0.09890597313642502,
-0.054345838725566864,
0.20887111127376556,
0.2272411584854126,
0.029515430331230164,
0.0963437408208847,
0.227055311203... |
https://github.com/huggingface/datasets/issues/2481 | My suggestion for this would be to have this enabled by default.
Plus I don't know if there should be a dedicated issue to that is another functionality. But I propose layered building rather than all at once. That is:
1. uncompress a handful of files via a generator enough to generate one arrow file
2. process ... | Delete extracted files to save disk space | As discussed with @stas00 and @lhoestq, allowing the deletion of extracted files would save a great amount of disk space to typical user. | 164 | Delete extracted files to save disk space
As discussed with @stas00 and @lhoestq, allowing the deletion of extracted files would save a great amount of disk space to typical user.
My suggestion for this would be to have this enabled by default.
Plus I don't know if there should be a dedicated issue to that is an... | [
-0.13562098145484924,
-0.30157917737960815,
-0.1714525818824768,
0.15534181892871857,
-0.07602386176586151,
-0.040315352380275726,
-0.12200194597244263,
0.5078598856925964,
-0.07595346868038177,
0.5109493136405945,
0.18390482664108276,
0.39844661951065063,
-0.372112512588501,
0.25196111202... |
https://github.com/huggingface/datasets/issues/2480 | For example to be able to send uncompressed and temp build files to another volume/partition, so that the user gets the minimal disk usage on their primary setup - and ends up with just the downloaded compressed data + arrow files, but outsourcing the huge files and building to another partition. e.g. on JZ there is a... | Set download/extracted paths configurable | As discussed with @stas00 and @lhoestq, setting these paths configurable may allow to overcome disk space limitation on different partitions/drives.
TODO:
- [x] Set configurable extracted datasets path: #2487
- [x] Set configurable downloaded datasets path: #2488
- [ ] Set configurable "incomplete" datasets path? | 85 | Set download/extracted paths configurable
As discussed with @stas00 and @lhoestq, setting these paths configurable may allow to overcome disk space limitation on different partitions/drives.
TODO:
- [x] Set configurable extracted datasets path: #2487
- [x] Set configurable downloaded datasets path: #2488
- [ ] ... | [
-0.2962554693222046,
-0.10550066828727722,
-0.18218055367469788,
0.1895645409822464,
0.11870540678501129,
-0.0878327265381813,
-0.04248432070016861,
0.29166164994239807,
0.04623398929834366,
0.34794938564300537,
-0.18032431602478027,
0.19076304137706757,
0.014195285737514496,
0.39972886443... |
https://github.com/huggingface/datasets/issues/2478 | I've aligned the release script with Transformers in #6004, so I think this issue can be closed. | Create release script | Create a script so that releases can be done automatically (as done in `transformers`). | 17 | Create release script
Create a script so that releases can be done automatically (as done in `transformers`).
I've aligned the release script with Transformers in #6004, so I think this issue can be closed. | [
-0.2713417410850525,
-0.17844514548778534,
-0.14083795249462128,
-0.14698253571987152,
0.19010137021541595,
-0.39825642108917236,
0.30418235063552856,
0.23012995719909668,
-0.02865901216864586,
0.2744125723838806,
0.08765643835067749,
0.19482098519802094,
-0.1964903324842453,
0.11970333009... |
https://github.com/huggingface/datasets/issues/2474 | Hi ! `load_from_disk` doesn't move the data. If you specify a local path to your mounted drive, then the dataset is going to be loaded directly from the arrow file in this directory. The cache files that result from `map` operations are also stored in the same directory by default.
However note than writing data to ... | cache_dir parameter for load_from_disk ? | **Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _load_from_disk_ function, the data gets cache... | 84 | cache_dir parameter for load_from_disk ?
**Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _lo... | [
-0.2717251181602478,
-0.12313547730445862,
-0.07020491361618042,
0.12663310766220093,
0.2155744582414627,
-0.02143780142068863,
0.28476476669311523,
0.08484739065170288,
0.42967867851257324,
0.3610472083091736,
-0.176080584526062,
0.24749420583248138,
-0.23836660385131836,
0.24224062263965... |
https://github.com/huggingface/datasets/issues/2474 | Thanks for your answer! I am a little surprised since I just want to read the dataset.
After debugging a bit, I noticed that the VM’s disk fills up when the tables (generator) are converted to a list:
https://github.com/huggingface/datasets/blob/5ba149773d23369617563d752aca922081277ec2/src/datasets/table.py#L850
... | cache_dir parameter for load_from_disk ? | **Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _load_from_disk_ function, the data gets cache... | 69 | cache_dir parameter for load_from_disk ?
**Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _lo... | [
-0.3292524814605713,
-0.13604718446731567,
-0.07416902482509613,
0.33951494097709656,
0.30544790625572205,
0.024210765957832336,
0.11280092597007751,
0.11024829745292664,
0.3817082643508911,
0.4735015630722046,
-0.11442671716213226,
0.33082467317581177,
-0.38737359642982483,
0.262867569923... |
https://github.com/huggingface/datasets/issues/2474 | Indeed reading the data shouldn't increase the VM's disk. Not sure what google colab does under the hood for that to happen | cache_dir parameter for load_from_disk ? | **Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _load_from_disk_ function, the data gets cache... | 22 | cache_dir parameter for load_from_disk ?
**Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _lo... | [
-0.29479527473449707,
-0.20461322367191315,
-0.1245887279510498,
0.24072860181331635,
0.19853802025318146,
0.0009989142417907715,
0.20941965281963348,
0.08362610638141632,
0.355987548828125,
0.48150819540023804,
-0.15545223653316498,
0.2485089898109436,
-0.3452035188674927,
0.2828158736228... |
https://github.com/huggingface/datasets/issues/2474 | Apparently, Colab uses a local cache of the data files read/written from Google Drive. See:
- https://github.com/googlecolab/colabtools/issues/2087#issuecomment-860818457
- https://github.com/googlecolab/colabtools/issues/1915#issuecomment-804234540
- https://github.com/googlecolab/colabtools/issues/2147#issuecommen... | cache_dir parameter for load_from_disk ? | **Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _load_from_disk_ function, the data gets cache... | 21 | cache_dir parameter for load_from_disk ?
**Is your feature request related to a problem? Please describe.**
When using Google Colab big datasets can be an issue, as they won't fit on the VM's disk. Therefore mounting google drive could be a possible solution. Unfortunatly when loading my own dataset by using the _lo... | [
-0.32445400953292847,
-0.10295620560646057,
-0.07441139966249466,
0.14649097621440887,
0.17781157791614532,
-0.02876509726047516,
0.32057327032089233,
0.07577325403690338,
0.26880136132240295,
0.4276999235153198,
-0.0916886180639267,
0.31767046451568604,
-0.2790684103965759,
0.263169348239... |
https://github.com/huggingface/datasets/issues/2472 | I have received a reply from Zenodo support:
> We are currently investigating and fixing this issue related to GitHub releases. As soon as we have solved it we will reach back to you. | Fix automatic generation of Zenodo DOI | After the last release of Datasets (1.8.0), the automatic generation of the Zenodo DOI failed: it appears in yellow as "Received", instead of in green as "Published".
I have contacted Zenodo support to fix this issue.
TODO:
- [x] Check with Zenodo to fix the issue
- [x] Check BibTeX entry is right | 34 | Fix automatic generation of Zenodo DOI
After the last release of Datasets (1.8.0), the automatic generation of the Zenodo DOI failed: it appears in yellow as "Received", instead of in green as "Published".
I have contacted Zenodo support to fix this issue.
TODO:
- [x] Check with Zenodo to fix the issue
- [x] ... | [
-0.0967244952917099,
0.37050050497055054,
-0.037610672414302826,
0.08053374290466309,
0.148946613073349,
-0.10635081678628922,
0.38638532161712646,
0.4031185209751129,
-0.061123933643102646,
0.25448524951934814,
-0.03677353262901306,
0.1469542682170868,
0.20907346904277802,
-0.082582816481... |
https://github.com/huggingface/datasets/issues/2472 | Other repo maintainers had the same problem with Zenodo.
There is an open issue on their GitHub repo: zenodo/zenodo#2181 | Fix automatic generation of Zenodo DOI | After the last release of Datasets (1.8.0), the automatic generation of the Zenodo DOI failed: it appears in yellow as "Received", instead of in green as "Published".
I have contacted Zenodo support to fix this issue.
TODO:
- [x] Check with Zenodo to fix the issue
- [x] Check BibTeX entry is right | 19 | Fix automatic generation of Zenodo DOI
After the last release of Datasets (1.8.0), the automatic generation of the Zenodo DOI failed: it appears in yellow as "Received", instead of in green as "Published".
I have contacted Zenodo support to fix this issue.
TODO:
- [x] Check with Zenodo to fix the issue
- [x] ... | [
-0.09328241646289825,
0.3942432701587677,
-0.03474089503288269,
0.09913042187690735,
0.1498735100030899,
-0.10668719559907913,
0.3806303143501282,
0.38822177052497864,
-0.07586833834648132,
0.25355538725852966,
-0.06456340849399567,
0.2056424766778946,
0.20682501792907715,
-0.0944910645484... |
https://github.com/huggingface/datasets/issues/2472 | I have received the following request from Zenodo support:
> Could you send us the link to the repository as well as the release tag?
My reply:
> Sure, here it is:
> - Link to the repository: https://github.com/huggingface/datasets
> - Link to the repository at the release tag: https://github.com/huggingface/dat... | Fix automatic generation of Zenodo DOI | After the last release of Datasets (1.8.0), the automatic generation of the Zenodo DOI failed: it appears in yellow as "Received", instead of in green as "Published".
I have contacted Zenodo support to fix this issue.
TODO:
- [x] Check with Zenodo to fix the issue
- [x] Check BibTeX entry is right | 55 | Fix automatic generation of Zenodo DOI
After the last release of Datasets (1.8.0), the automatic generation of the Zenodo DOI failed: it appears in yellow as "Received", instead of in green as "Published".
I have contacted Zenodo support to fix this issue.
TODO:
- [x] Check with Zenodo to fix the issue
- [x] ... | [
0.018832698464393616,
0.26431697607040405,
-0.053757596760988235,
0.1654243767261505,
0.1444077491760254,
-0.0486195832490921,
0.295654833316803,
0.3008950352668762,
0.012550584971904755,
0.2421855479478836,
-0.012595836073160172,
0.15270034968852997,
0.15315325558185577,
-0.01056037843227... |
https://github.com/huggingface/datasets/issues/2470 | Hi ! It looks like the issue comes from pyarrow. What version of pyarrow are you using ? How did you install it ? | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`. | ## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure whether the issue is on my end, because it's difficult for me to debug! Any ti... | 24 | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`.
## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure ... | [
-0.4569716453552246,
-0.21739079058170319,
0.022877953946590424,
0.42194560170173645,
0.31130778789520264,
-0.22648212313652039,
0.2361345887184143,
0.15250523388385773,
0.04312615841627121,
0.36432933807373047,
0.5303294658660889,
0.2979789674282074,
-0.33465102314949036,
-0.0465701818466... |
https://github.com/huggingface/datasets/issues/2470 | Thank you for the quick reply! I have `pyarrow==4.0.0`, and I am installing with `pip`. It's not one of my explicit dependencies, so I assume it came along with something else. | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`. | ## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure whether the issue is on my end, because it's difficult for me to debug! Any ti... | 31 | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`.
## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure ... | [
-0.4569716453552246,
-0.21739079058170319,
0.022877953946590424,
0.42194560170173645,
0.31130778789520264,
-0.22648212313652039,
0.2361345887184143,
0.15250523388385773,
0.04312615841627121,
0.36432933807373047,
0.5303294658660889,
0.2979789674282074,
-0.33465102314949036,
-0.0465701818466... |
https://github.com/huggingface/datasets/issues/2470 | Could you trying reinstalling pyarrow with pip ?
I'm not sure why it would check in your multicurtural-sc directory for source files. | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`. | ## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure whether the issue is on my end, because it's difficult for me to debug! Any ti... | 22 | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`.
## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure ... | [
-0.4569716453552246,
-0.21739079058170319,
0.022877953946590424,
0.42194560170173645,
0.31130778789520264,
-0.22648212313652039,
0.2361345887184143,
0.15250523388385773,
0.04312615841627121,
0.36432933807373047,
0.5303294658660889,
0.2979789674282074,
-0.33465102314949036,
-0.0465701818466... |
https://github.com/huggingface/datasets/issues/2470 | Sure! I tried reinstalling to get latest. pip was mad because it looks like Datasets currently wants <4.0.0 (which is interesting, because apparently I ended up with 4.0.0 already?), but I gave it a shot anyway:
```bash
$ pip install --upgrade --force-reinstall pyarrow
Collecting pyarrow
Downloading pyarrow-4.0... | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`. | ## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure whether the issue is on my end, because it's difficult for me to debug! Any ti... | 305 | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`.
## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure ... | [
-0.4569716453552246,
-0.21739079058170319,
0.022877953946590424,
0.42194560170173645,
0.31130778789520264,
-0.22648212313652039,
0.2361345887184143,
0.15250523388385773,
0.04312615841627121,
0.36432933807373047,
0.5303294658660889,
0.2979789674282074,
-0.33465102314949036,
-0.0465701818466... |
https://github.com/huggingface/datasets/issues/2470 | Good catch ! Not sure why it could raise such a weird issue from pyarrow though
We should definitely reduce num_proc to the length of the dataset if needed and log a warning. | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`. | ## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure whether the issue is on my end, because it's difficult for me to debug! Any ti... | 33 | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`.
## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure ... | [
-0.4569716453552246,
-0.21739079058170319,
0.022877953946590424,
0.42194560170173645,
0.31130778789520264,
-0.22648212313652039,
0.2361345887184143,
0.15250523388385773,
0.04312615841627121,
0.36432933807373047,
0.5303294658660889,
0.2979789674282074,
-0.33465102314949036,
-0.0465701818466... |
https://github.com/huggingface/datasets/issues/2470 | This has been fixed in #2566, thanks @connor-mccarthy !
We'll make a new release soon that includes the fix ;) | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`. | ## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure whether the issue is on my end, because it's difficult for me to debug! Any ti... | 20 | Crash when `num_proc` > dataset length for `map()` on a `datasets.Dataset`.
## Describe the bug
Crash if when using `num_proc` > 1 (I used 16) for `map()` on a `datasets.Dataset`.
I believe I've had cases where `num_proc` > 1 works before, but now it seems either inconsistent, or depends on my data. I'm not sure ... | [
-0.4569716453552246,
-0.21739079058170319,
0.022877953946590424,
0.42194560170173645,
0.31130778789520264,
-0.22648212313652039,
0.2361345887184143,
0.15250523388385773,
0.04312615841627121,
0.36432933807373047,
0.5303294658660889,
0.2979789674282074,
-0.33465102314949036,
-0.0465701818466... |
https://github.com/huggingface/datasets/issues/2462 | Unless there is high demande I don't think we will end up implementing this. This is a lot of work with very few advantages | Merge DatasetDict and Dataset | As discussed in #2424 and #2437 (please see there for detailed conversation):
- It would be desirable to improve UX with respect the confusion between DatasetDict and Dataset.
- The difference between Dataset and DatasetDict is an additional abstraction complexity that confuses "typical" end users.
- A user expects... | 24 | Merge DatasetDict and Dataset
As discussed in #2424 and #2437 (please see there for detailed conversation):
- It would be desirable to improve UX with respect the confusion between DatasetDict and Dataset.
- The difference between Dataset and DatasetDict is an additional abstraction complexity that confuses "typica... | [
-0.24885764718055725,
0.2865617871284485,
-0.1583000123500824,
-0.08640151470899582,
0.1658450812101364,
0.13266025483608246,
0.2925472855567932,
0.5712538957595825,
0.303635835647583,
0.15918242931365967,
-0.04712817817926407,
0.3749987781047821,
-0.21041971445083618,
0.5265560150146484,
... |
https://github.com/huggingface/datasets/issues/2450 | Hi ! The `blue` metric doesn't exist, but the `bleu` metric does.
You can get the full list of metrics [here](https://github.com/huggingface/datasets/tree/master/metrics) or by running
```python
from datasets import list_metrics
print(list_metrics())
``` | BLUE file not found | Hi, I'm having the following issue when I try to load the `blue` metric.
```shell
import datasets
metric = datasets.load_metric('blue')
Traceback (most recent call last):
File "/home/irfan/environments/Perplexity_Transformers/lib/python3.6/site-packages/datasets/load.py", line 320, in prepare_module
local... | 31 | BLUE file not found
Hi, I'm having the following issue when I try to load the `blue` metric.
```shell
import datasets
metric = datasets.load_metric('blue')
Traceback (most recent call last):
File "/home/irfan/environments/Perplexity_Transformers/lib/python3.6/site-packages/datasets/load.py", line 320, in pre... | [
-0.20386210083961487,
-0.2862851917743683,
-0.06850095838308334,
0.4027130603790283,
0.3037267327308655,
0.06551104038953781,
0.2102445513010025,
0.3386648893356323,
0.08859750628471375,
0.18913912773132324,
-0.22808998823165894,
-0.15020716190338135,
0.08000030368566513,
-0.38681069016456... |
https://github.com/huggingface/datasets/issues/2447 | Hi ! I'm pretty sure that the answers are not made available for the test set on purpose because it is part of the DynaBench benchmark, for which you can submit your predictions on the website.
In any case we should mention this in the dataset card of this dataset. | dataset adversarial_qa has no answers in the "test" set | ## Describe the bug
When loading the adversarial_qa dataset the 'test' portion has no answers. Only the 'train' and 'validation' portions do. This occurs with all four of the configs ('adversarialQA', 'dbidaf', 'dbert', 'droberta')
## Steps to reproduce the bug
```
from datasets import load_dataset
examples ... | 50 | dataset adversarial_qa has no answers in the "test" set
## Describe the bug
When loading the adversarial_qa dataset the 'test' portion has no answers. Only the 'train' and 'validation' portions do. This occurs with all four of the configs ('adversarialQA', 'dbidaf', 'dbert', 'droberta')
## Steps to reproduce th... | [
-0.10978080332279205,
0.17811372876167297,
-0.13903790712356567,
0.31031355261802673,
0.16347694396972656,
-0.08224822580814362,
0.3094848096370697,
0.4499255120754242,
-0.051962461322546005,
0.11818760633468628,
0.16112393140792847,
0.3690260648727417,
-0.1847234070301056,
-0.247967511415... |
https://github.com/huggingface/datasets/issues/2447 | Makes sense, but not intuitive for someone searching through the datasets. Thanks for adding the note to clarify. | dataset adversarial_qa has no answers in the "test" set | ## Describe the bug
When loading the adversarial_qa dataset the 'test' portion has no answers. Only the 'train' and 'validation' portions do. This occurs with all four of the configs ('adversarialQA', 'dbidaf', 'dbert', 'droberta')
## Steps to reproduce the bug
```
from datasets import load_dataset
examples ... | 18 | dataset adversarial_qa has no answers in the "test" set
## Describe the bug
When loading the adversarial_qa dataset the 'test' portion has no answers. Only the 'train' and 'validation' portions do. This occurs with all four of the configs ('adversarialQA', 'dbidaf', 'dbert', 'droberta')
## Steps to reproduce th... | [
-0.038019124418497086,
0.17751628160476685,
-0.10937897861003876,
0.3759942650794983,
0.19822123646736145,
-0.006238430738449097,
0.36139416694641113,
0.3695281744003296,
-0.06413215398788452,
0.03522852063179016,
0.1037120372056961,
0.379220575094223,
-0.1990857869386673,
-0.1972342431545... |
https://github.com/huggingface/datasets/issues/2446 | ```
File "/home/sasha/.local/share/virtualenvs/lib-ogGKnCK_/lib/python3.7/site-packages/streamlit/script_runner.py", line 332, in _run_script
exec(code, module.__dict__)
File "/home/sasha/nlp-viewer/run.py", line 233, in <module>
configs = get_confs(option)
File "/home/sasha/.local/share/virtualenvs/lib-og... | `yelp_polarity` is broken | 
| 118 | `yelp_polarity` is broken
```
File "/home/sasha/.local/share/virtualenvs/lib-ogGKnCK_/lib/python3.7/site-packages/streamlit/script_runner.py", line 332, in _run_script
exec(code, module.__dict__)
... | [
-0.1024581789970398,
-0.4164356589317322,
-0.15971586108207703,
0.11063823103904724,
0.2700803279876709,
-0.13424381613731384,
0.062041815370321274,
0.3248958885669708,
0.11502442508935928,
0.020860828459262848,
-0.07330349832773209,
0.09813584387302399,
-0.07752877473831177,
0.08816040307... |
https://github.com/huggingface/datasets/issues/2444 | Hi,
This is a known issue. More info on this issue can be found in #2061. If you are looking for an open-source contribution, there are step-by-step instructions in the linked issue that you can follow to fix it. | Sentence Boundaries missing in Dataset: xtreme / udpos | I was browsing through annotation guidelines, as suggested by the datasets introduction.
The guidlines saids "There must be exactly one blank line after every sentence, including the last sentence in the file. Empty sentences are not allowed." in the [Sentence Boundaries and Comments section](https://universaldepend... | 39 | Sentence Boundaries missing in Dataset: xtreme / udpos
I was browsing through annotation guidelines, as suggested by the datasets introduction.
The guidlines saids "There must be exactly one blank line after every sentence, including the last sentence in the file. Empty sentences are not allowed." in the [Sentence... | [
0.294378399848938,
-0.3165704011917114,
-0.00033968687057495117,
-0.04292578995227814,
0.20856916904449463,
-0.2181830257177353,
0.2861594259738922,
-0.2059638351202011,
-0.08608429878950119,
0.0388183519244194,
-0.13336583971977234,
-0.03214403614401817,
0.09912853688001633,
0.01232764124... |
https://github.com/huggingface/datasets/issues/2443 | Hi ! That would be nice indeed to at least have a warning, since we don't handle the max path length limit.
Also if we could have an error instead of an infinite loop I'm sure windows users will appreciate that | Some tests hang on Windows | Currently, several tests hang on Windows if the max path limit of 260 characters is not disabled. This happens due to the changes introduced by #2223 that cause an infinite loop in `WindowsFileLock` described in #2220. This can be very tricky to debug, so I think now is a good time to address these issues/PRs. IMO thr... | 41 | Some tests hang on Windows
Currently, several tests hang on Windows if the max path limit of 260 characters is not disabled. This happens due to the changes introduced by #2223 that cause an infinite loop in `WindowsFileLock` described in #2220. This can be very tricky to debug, so I think now is a good time to addr... | [
-0.2391643077135086,
-0.021761521697044373,
-0.1179872453212738,
0.001324240118265152,
0.3143852651119232,
-0.050242818892002106,
0.18320363759994507,
0.15665362775325775,
0.16434034705162048,
0.3078768253326416,
0.7059504985809326,
-0.09653893113136292,
-0.47966915369033813,
0.04251411557... |
https://github.com/huggingface/datasets/issues/2443 | Unfortunately, I know this problem very well... 😅
I remember having proposed to throw an error instead of hanging in an infinite loop #2220: 60c7d1b6b71469599a27147a08100f594e7a3f84, 8c8ab60018b00463edf1eca500e434ff061546fc
but @lhoestq told me:
> Note that the filelock module comes from this project that hasn'... | Some tests hang on Windows | Currently, several tests hang on Windows if the max path limit of 260 characters is not disabled. This happens due to the changes introduced by #2223 that cause an infinite loop in `WindowsFileLock` described in #2220. This can be very tricky to debug, so I think now is a good time to address these issues/PRs. IMO thr... | 85 | Some tests hang on Windows
Currently, several tests hang on Windows if the max path limit of 260 characters is not disabled. This happens due to the changes introduced by #2223 that cause an infinite loop in `WindowsFileLock` described in #2220. This can be very tricky to debug, so I think now is a good time to addr... | [
-0.13163551688194275,
-0.014070495963096619,
-0.11839597672224045,
0.009343014098703861,
0.21633321046829224,
0.009574316442012787,
0.25838252902030945,
0.13995510339736938,
0.21939101815223694,
0.1932307779788971,
0.6185334324836731,
-0.20448359847068787,
-0.4562077820301056,
-0.072764605... |
https://github.com/huggingface/datasets/issues/2443 | @albertvillanova Thanks for additional info on this issue.
Yes, I think the best option is to throw an error instead of suppressing it in a loop. I've considered 2 more options, but I don't really like them:
1. create a temporary file with a filename longer than 255 characters on import; if this fails, long paths a... | Some tests hang on Windows | Currently, several tests hang on Windows if the max path limit of 260 characters is not disabled. This happens due to the changes introduced by #2223 that cause an infinite loop in `WindowsFileLock` described in #2220. This can be very tricky to debug, so I think now is a good time to address these issues/PRs. IMO thr... | 109 | Some tests hang on Windows
Currently, several tests hang on Windows if the max path limit of 260 characters is not disabled. This happens due to the changes introduced by #2223 that cause an infinite loop in `WindowsFileLock` described in #2220. This can be very tricky to debug, so I think now is a good time to addr... | [
-0.21726931631565094,
0.011509314179420471,
-0.15492044389247894,
0.09689950197935104,
0.2777681052684784,
-0.11498236656188965,
0.29523104429244995,
0.1591925323009491,
0.20752447843551636,
0.33040669560432434,
0.5141476392745972,
-0.141556054353714,
-0.37055206298828125,
0.00127326697111... |
https://github.com/huggingface/datasets/issues/2441 | Hi ! In your dataset script you must be yielding examples like
```python
for line in file:
...
yield key, {...}
```
Since `datasets` 1.7.0 we enforce the keys to be unique.
However it looks like your examples generator creates duplicate keys: at least two examples have key 0.
You can fix that by mak... | DuplicatedKeysError on personal dataset | ## Describe the bug
Ever since today, I have been getting a DuplicatedKeysError while trying to load my dataset from my own script.
Error returned when running this line: `dataset = load_dataset('/content/drive/MyDrive/Thesis/Datasets/book_preprocessing/goodreads_maharjan_trimmed_and_nered/goodreadsnered.py')`
Note ... | 104 | DuplicatedKeysError on personal dataset
## Describe the bug
Ever since today, I have been getting a DuplicatedKeysError while trying to load my dataset from my own script.
Error returned when running this line: `dataset = load_dataset('/content/drive/MyDrive/Thesis/Datasets/book_preprocessing/goodreads_maharjan_tri... | [
-0.18643437325954437,
-0.05652356147766113,
0.07348434627056122,
0.2193855345249176,
0.14254815876483917,
0.029286324977874756,
0.5267680287361145,
0.2207099050283432,
0.05882291495800018,
0.08725564181804657,
-0.1790749430656433,
0.2924933433532715,
-0.13538840413093567,
-0.12589076161384... |
https://github.com/huggingface/datasets/issues/2440 | The tagger also doesn't insert the value for the `size_categories` field automatically, so this should be fixed too | Remove `extended` field from dataset tagger | ## Describe the bug
While working on #2435 I used the [dataset tagger](https://huggingface.co/datasets/tagging/) to generate the missing tags for the YAML metadata of each README.md file. However, it seems that our CI raises an error when the `extended` field is included:
```
dataset_name = 'arcd'
@pytest.m... | 18 | Remove `extended` field from dataset tagger
## Describe the bug
While working on #2435 I used the [dataset tagger](https://huggingface.co/datasets/tagging/) to generate the missing tags for the YAML metadata of each README.md file. However, it seems that our CI raises an error when the `extended` field is included:
... | [
-0.047057975083589554,
-0.027882933616638184,
-0.006331034004688263,
0.22309041023254395,
0.3980686068534851,
0.3846164345741272,
0.3813190460205078,
0.20380324125289917,
0.133192777633667,
0.30706682801246643,
0.34983932971954346,
0.3464972972869873,
-0.33761686086654663,
0.01012215763330... |
https://github.com/huggingface/datasets/issues/2440 | Thanks for reporting. Indeed the `extended` tag doesn't exist. Not sure why we had that in the tagger.
The repo of the tagger is here if someone wants to give this a try: https://github.com/huggingface/datasets-tagging
Otherwise I can probably fix it next week | Remove `extended` field from dataset tagger | ## Describe the bug
While working on #2435 I used the [dataset tagger](https://huggingface.co/datasets/tagging/) to generate the missing tags for the YAML metadata of each README.md file. However, it seems that our CI raises an error when the `extended` field is included:
```
dataset_name = 'arcd'
@pytest.m... | 42 | Remove `extended` field from dataset tagger
## Describe the bug
While working on #2435 I used the [dataset tagger](https://huggingface.co/datasets/tagging/) to generate the missing tags for the YAML metadata of each README.md file. However, it seems that our CI raises an error when the `extended` field is included:
... | [
0.00031366944313049316,
-0.09879451245069504,
-0.0033113211393356323,
0.19121256470680237,
0.42418888211250305,
0.4057803153991699,
0.37730616331100464,
0.23212504386901855,
0.11827585101127625,
0.28089970350265503,
0.31300562620162964,
0.42946308851242065,
-0.29180118441581726,
0.00031004... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.