Datasets:

---

Francesco-A

/

github-issues_huggingface-datasets

like 0

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[ "Now, I can't find any reference to set_progress_bar_enabled in the code.\r\n\r\nI think it have been deleted", "Hi @virilo,\r\n\r\nPlease note that since `datasets` version 2.0.0, we have aligned with `transformers` the management of the progress bar (among other things):\r\n- #3897\r\n\r\nNow, you should update your code to use `datasets.logging.disable_progress_bar`.\r\n\r\nYou have more info in our docs: [Logging methods](https://huggingface.co/docs/datasets/package_reference/logging_methods)", "One important thing for beginner like me is: from datasets.utils.logging import disable_progress_bar\r\nDo not forget the 'utils' or you will waste a long time like me...." ]

[ { "color": "E5583E", "default": false, "description": "Related to the dataset viewer on huggingface.co", "id": 3470211881, "name": "dataset-viewer", "node_id": "LA_kwDODunzps7O1zsp", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset-viewer" } ]

[ "It's a bug in the dataset viewer: the dataset cannot be downloaded in streaming mode, but since the dataset is relatively small, the dataset viewer should have fallback to normal mode. Working on a fix.", "Fixed. https://huggingface.co/datasets/samsum/viewer/samsum/train\r\n\r\n<img width=\"1563\" alt=\"Capture d’écran 2022-04-12 à 13 47 45\" src=\"https://user-images.githubusercontent.com/1676121/162953339-cd8922d7-9037-408b-b896-eac1af0bb54f.png\">\r\n\r\nThanks for reporting!" ]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[ "EDIT:\r\nI think in this case, the metadata files (either README or JSON) should not be read (i.e. `self.info.splits` should be None).\r\n\r\nOne idea: \r\n- I think ideally we should set this behavior when we pass `--save_info` to the CLI `test`\r\n- However, currently, the builder is unaware of this: `save_info` arg is not passed to it", "> I think in this case\r\n\r\n@albertvillanova You mean in cases when the script was changed? \r\n\r\nI suggest that we:\r\n* add a check on the slice (like 'split_name[n%]) kind of format here: https://github.com/huggingface/datasets/blob/main/src/datasets/splits.py#L523 to catch things like this. \r\n* Error here happens before splits verification, but in `_prepare_split`, and `_prepare_split` doesn't perform any verification and don't know about it. so we can pass this parameter and take splits from `split_generator`, not from `split.info` in case when `verify_infos` is False\r\n* we can check if split **names** from split_generators and self.info.splits are the same **before** preparing splits (if `verify_info=True`) so that we don't spend time on generating unwanted data. \r\n* provide some user-friendly warnings about `ignore_verifications` parameter so that users know that if something is not matching they can ignore it\r\n\r\nI started it here: https://github.com/huggingface/datasets/pull/5327/files\r\n\r\nWhat do you think @albertvillanova ?", "I edited my previous comment:\r\n- First I proposed setting `self.info.splits` to None when `ignore_verifications=True`\r\n - I thought it was the easiest implementation because `ignore_verifications` is passed to `DatasetBuilder.download_and_prepare`\r\n - However, afterwards, I realized this might not be a good idea for this use case:\r\n - A user wants to optimize the loading of the dataset, and passes `ignore_verifications=False` to avoid all the verifications\r\n - In this case, we want `self.info.splits` to be read from metadata file\r\n- Then, I thought that it might be better to set `self.info.splits` to None when we pass `--save_info` to the CLI test: if we are going to save the info to the metadata file, it makes no sense to read the info from the metadata file\r\n - This implementation is not so easy because the Builder knows nothing about `--save_info`\r\n\r\nI agree with you there are 2 things to be addressed here:\r\n- One is what I have just commented: `self.info.splits` should be None in this case\r\n- The other, a validation should be implemented when calling `make_file_instructions` and/or `SplitDict.__getitem__`, so that when passing \"training\" to it, we get a more descriptive error other than `TypeError: expected str, bytes or os.PathLike object, not NoneType` " ]

[ { "color": "a2eeef", "default": true, "description": "New feature or request", "id": 1935892871, "name": "enhancement", "node_id": "MDU6TGFiZWwxOTM1ODkyODcx", "url": "https://api.github.com/repos/huggingface/datasets/labels/enhancement" } ]

[ "Indeed that would be cool\r\n\r\nAlso FYI right now the easiest way to do this is\r\n```python\r\ndataset_dict[\"train\"] = dataset_dict[\"train\"].map(my_transform_for_the_train_set)\r\ndataset_dict[\"test\"] = dataset_dict[\"test\"].map(my_transform_for_the_test_set)\r\n```", "I don't feel like adding an extra param for this simple usage makes sense, considering how many args `map` already has. \r\n\r\n(Feel free to re-open this issue if you don't agree with me)", "I still think this is useful, since it's common that the data processing is different for training/dev/testing. And I don't know if the fact that `map` currently takes many arguments is a good reason not to support a useful feature." ]

[]

[]

[]

[ "Hey @albertvillanova could you provide more context, including extracts from the discussion we had ?\r\n\r\nLet's ping @Pierrci @julien-c and @n1t0 for their opinion about that", "@julien-c increased the huggingface.co production workers in order to see if it solve [the 502 you had this morning](https://app.circleci.com/pipelines/github/huggingface/datasets/7843/workflows/fc83fa32-18f5-4dc3-9e2f-ba277ae1af74)\r\n\r\nFor the decision process: be aware that moon-staging does not have persistent repos (they are deleted regularly). as a consequence, **if the moon-staging solution is validated**, you should consider a way to keep the repository that are loaded in tests. These are the ones I found : https://github.com/huggingface/datasets/blob/d488db2f64f312f88f72bbc57a09b7eddb329182/tests/test_load.py and https://github.com/huggingface/datasets/blob/40773111c3e7db8a992fa1c48af32d900a1018d6/tests/test_streaming_download_manager." ]

[ { "color": "E5583E", "default": false, "description": "Related to the dataset viewer on huggingface.co", "id": 3470211881, "name": "dataset-viewer", "node_id": "LA_kwDODunzps7O1zsp", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset-viewer" } ]

[ "Can you reopen with the correct dataset name (if relevant)?\r\n\r\nThanks", "The dataset viewer was down tonight. It works again." ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

[]

[ "Nice :) \r\n\r\ncould you add the tags and also remove all the dummy data files that are not zipped ? The diff currently shows 800 files changes xD", "Aaaaand it's rebase time - the new one is at #1264 !", "closing this one since a new PR was created" ]

[]

[ "@lhoestq please review this PR when you get free", "If the CI fails just because of `RemoteDatasetTest` errors it's fine, they're fixed on master", "> If the CI fails just because of `RemoteDatasetTest` errors it's fine, they're fixed on master\r\n\r\nThe same 3 tests are failing again :(\r\n```\r\nFAILED tests/test_dataset_common.py::RemoteDatasetTest::test_builder_class_norwegian_ner\r\nFAILED tests/test_dataset_common.py::RemoteDatasetTest::test_builder_configs_norwegian_ner\r\nFAILED tests/test_dataset_common.py::RemoteDatasetTest::test_load_dataset_norwegian_ner\r\n```", "One trick if you want to add more datasets to avoid these errors : you can just rebase the master branch of your fork from the master branch of the repo. Then each time you make a new branch from master on your fork, it will include the fix for these errors", "> One trick if you want to add more datasets to avoid these errors : you can just rebase the master branch of your fork from the master branch of the repo. Then each time you make a new branch from master on your fork, it will include the fix for these errors\r\n\r\nYes, I almost always do that, but somehow seems even this branch got old 😓 \r\nI also do the following if I directly create a new branch locally: `git checkout -b <branchname> upstream/master` so it stays up-to date irrespective of my fork, still don't know how this crept in again", "Merging this one since the CI is fixed on master" ]

[ { "color": "a2eeef", "default": true, "description": "New feature or request", "id": 1935892871, "name": "enhancement", "node_id": "MDU6TGFiZWwxOTM1ODkyODcx", "url": "https://api.github.com/repos/huggingface/datasets/labels/enhancement" } ]

[]

[ { "color": "e99695", "default": false, "description": "Requesting to add a new dataset", "id": 2067376369, "name": "dataset request", "node_id": "MDU6TGFiZWwyMDY3Mzc2MzY5", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20request" } ]

[]

[]

[]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

[]

[]

[]

[]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[ "Hi @adrianeboyd, thanks for reporting.\r\n\r\nThere is indeed a bug in that community dataset:\r\nLine:\r\n```python\r\nmetadata_and_text_files = list(zip(metadata_files, text_files))\r\n``` \r\nshould be replaced with\r\n```python\r\nmetadata_and_text_files = list(zip(sorted(metadata_files), sorted(text_files)))\r\n```\r\n\r\nI am going to contact their owners (https://huggingface.co/oscar-corpus) in order to inform them about the bug.\r\n\r\nI keep you informed.", "That fix is part of it, but it's clearly not the only issue.\r\n\r\nI also already contacted the OSCAR creators, but I reported it here because it looked like huggingface members were the main authors in the git history. Is there a better place to have reported this?", "Hello,\r\n\r\nWe've had an issue that could be linked to this one here: https://github.com/oscar-corpus/corpus/issues/15.\r\n\r\nI have been spot checking the source (`.txt`/`.jsonl`) files for a while, and have not found issues, especially in the start/end of corpora (but I conceed that more integration testing would be necessary on our side).\r\n\r\nThe text and metadata files are designed to be used in sync (with `lang_part_n.txt` and `lang_meta_part_n.jsonl` working together), while staying independent from part to part, so that anyone could randomly choose a part and work with it.\r\n\r\nThe fix @albertvillanova proposed should fix the problem, as the parts will be in sync again.\r\n\r\nLet me know if you need help or more details, I'd be glad to help!", "I'm happy to move the discussion to the other repo!\r\n\r\nMerely sorting the files only **maybe** fixes the processing of the first part. If the first part contains non-unix newlines, it will still be misaligned/truncated, and all the following parts will be truncated with incorrect text offsets and metadata due the offset and newline bugs.", "Fixed:\r\n- https://huggingface.co/datasets/oscar-corpus/OSCAR-2109/commit/3cd7e95aa1799b73c5ea8afc3989635f3e19b86b", "Hi @Uinelj, This is a total noobs question but how can I integrate that bugfix into my code? I reinstalled the datasets library this time from source. Should that have fixed the issue? I am still facing the misalignment issue. Do I need to download the dataset from scratch?", "Hi, I re-downloaded the dataset and still have the problem. See: https://github.com/oscar-corpus/corpus/issues/18", "Sorry @norakassner for the late reply.\r\n\r\nThere are indeed several issues creating the misalignment, as @adrianeboyd cleverly pointed out:\r\n- https://huggingface.co/datasets/oscar-corpus/OSCAR-2109/commit/3cd7e95aa1799b73c5ea8afc3989635f3e19b86b fixed one of them\r\n- but there are still others to be fixed", "Normally, the issues should be fixed now:\r\n- Fix offset initialization for each file: https://huggingface.co/datasets/oscar-corpus/OSCAR-2109/commit/1ad9b7bfe00798a9258a923b887bb1c8d732b833\r\n- Disable default universal newline support: https://huggingface.co/datasets/oscar-corpus/OSCAR-2109/commit/0c2f307d3167f03632f502af361ac6c3c393f510\r\n\r\nFeel free to reopen if you find additional misalignments/truncations.\r\n\r\nCC: @adrianeboyd @norakassner @Uinelj ", "Thanks for the updates!\r\n\r\nThe purist in me would still like to have the rstrip not strip additional characters from the original text (unicode whitespace mainly in practice, I think), but the differences are extremely small in practice and it doesn't actually matter for my current task:\r\n\r\n```python\r\ntext = \"\".join([text_f.readline() for _ in range(meta[\"nb_sentences\"])]).rstrip(\"\\n\")\r\n```" ]

[]

[]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_3886). All of your documentation changes will be reflected on that endpoint.", "I made it more robust by increasing the wait time, and I also added some logs when a request is retried. Let me know if it's ok for you", "At the end you did not set the agreed max value of 60s. \r\n\r\nMoreover, with the new numbers, there is a slight contradiction: although you set max_retries=5, we will only make 4 retries at most because of the combined values of `base_wait_time` and `max_wait_time`.", "Yea I thought that in total we could wait 1min, but if we have a max_wait_time of 20sec between each request it's fine IMO\r\n\r\n> Moreover, with the new numbers, there is a slight contradiction: although you set max_retries=5, we will only make 4 retries at most because of the combined values of base_wait_time and max_wait_time.\r\n\r\nWhat makes you think this ? If the exponential wait time becomes bigger than `max_wait_time` then it still does the retry, but after a wait time of `max_wait_time`", "Sorry, I meant 4 retries **with exponential backoff**; the fifth one is with constant backoff.", "OK, and one question: do you think that the retries do not affect the time the server needs to be operational again and able to process the request? I guess that if does not affect, then the cause are other users' requests, or others; not our specific request.\r\n\r\nJust to be sure: \r\n- Then 20s at most between consecutive requests do not impact the server.\r\n- And we expect after a total of 5 retries (within a total 50s of wait time + request processing/uploading time), the server should be able to come back to normality.", "> do you think that the retries do not affect the time the server needs for being able to process the request (I guess in this case the cause are other users' requests, or other causes; not our specific request).\r\n\r\nYes I don't think the retries would affect the server, I think the cause of the 504 errors is elsewhere\r\n\r\n> Just to be sure:\r\n>\r\n> Then 20s at most between consecutive requests do not impact the server.\r\n> And we expect after a total of 5 retries (within a total 50s of wait time + request processing/uploading time), the server should be able to come back to normality.\r\n\r\nYes I think it's fine for now, we can still adapt this later if needed", "Will be curious to see the impact of this in terms of upload reliability! Don't forget to let us know when you have more data. cc @huggingface/moon-landing-back " ]

[ { "color": "a2eeef", "default": true, "description": "New feature or request", "id": 1935892871, "name": "enhancement", "node_id": "MDU6TGFiZWwxOTM1ODkyODcx", "url": "https://api.github.com/repos/huggingface/datasets/labels/enhancement" }, { "color": "c5def5", "default": false, "description": "Generic discussion on the library", "id": 2067400324, "name": "generic discussion", "node_id": "MDU6TGFiZWwyMDY3NDAwMzI0", "url": "https://api.github.com/repos/huggingface/datasets/labels/generic%20discussion" } ]

[ "Thanks a lot, @aleSuglia for the very detailed and introductive feature request.\r\nIt seems like we could build something pretty useful here indeed.\r\n\r\nOne of the questions here is that Arrow doesn't have built-in support for generic \"tensors\" in records but there might be ways to do that in a clean way. We'll probably try to tackle this during the summer.", "I was looking into Facebook MMF and apparently they decided to use LMDB to store additional features associated with every example: https://github.com/facebookresearch/mmf/blob/master/mmf/datasets/databases/features_database.py\r\n\r\n", "I saw the Mozilla common_voice dataset in model hub, which has mp3 audio recordings as part it. It's use predominantly maybe in ASR and TTS, but dataset is a Language + Voice Dataset similar to @aleSuglia's point about Language + Vision. \r\n\r\nhttps://huggingface.co/datasets/common_voice", "Hey @thomwolf, are there any updates on this? I would love to contribute if possible!\r\n\r\nThanks, \r\nAlessandro ", "Hi @aleSuglia :) In today's new release 1.17 of `datasets` we introduce a new feature type `Image` that allows to store images directly in a dataset, next to text features and labels for example. There is also an `Audio` feature type, for datasets containing audio data. For tensors there are `Array2D`, `Array3D`, etc. feature types\r\n\r\nNote that both Image and Audio feature types take care of decoding the images/audio data if needed. The returned images are PIL images, and the audio signals are decoded as numpy arrays.\r\n\r\nAnd `datasets` also leverage end-to-end zero copy from the arrow data for all of them, for maximum speed :)" ]

[]

[ "> Awesome thank you !\r\n> \r\n> The dummy data are quite big but it was expected given that the raw files are flac files.\r\n> Given that the script doesn't even read the flac files I think we can remove them. Or maybe use empty flac files (see [here](https://hydrogenaud.io/index.php?topic=118685.0) for example). What do you think ?\r\n> \r\n> We'll find a better solution to be able to have bigger dummy_data (max 1MB instead of a few KB, maybe using git LFS.\r\n\r\nHmm, I already made the dummy data as small as possible (a single flac filie per split only). I'd like to keep them at least to have complete dummy data and don't think 500KB for all datasets together is a problem (the long-range summarization datasets are similarly heavy). The moment we allow dummy data to be loaded directly for testing, we need the flac files IMO.\r\n\r\nBut I agree that longterm, we need a better solution for the dummy data (maybe stop hosting it on github to not make the repo too heavy)" ]

[]

[ "The MIT license seems to be for the crawling code, no ? Then maybe we can also redirect users to the [terms of smashwords.com](https://www.smashwords.com/about/tos) regarding copyrights, in particular the paragraph 10 for end-users. In particular it seems that end users can download and use the content \"for their personal enjoyment in any reasonable non-commercial manner in compliance with copyright law\" and the smashwords end-users agreement.\r\n\r\nIt should be the same for https://github.com/huggingface/datasets/pull/3526 as well", "May I merge this one ?", "_The documentation is not available anymore as the PR was closed or merged._" ]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[ "Hi @erikadistefano , thanks for reporting the issue.\r\n\r\nI have created a Pull Request that should fix it. \r\n\r\nOnce merged into master, feel free to update your installed `datasets` library (either by installing it from our GitHub master branch or waiting until our next release) to be able to load omp dataset." ]

[]

[]

[]

[ "Changed the pre-processing based on the comments raised in [PR-1482](https://github.com/huggingface/datasets/pull/1482).The below command is passing in my local environment:\r\n\r\n`python datasets-cli test datasets/covid_qa_ucsd/ --save_infos --all_configs --data_dir ~/Downloads/Medical-Dialogue-Dataset/CovidDailogue/`\r\n\r\n" ]

[]

[ "cc @SBrandeis", "I think this has been addressed.\r\n\r\nPlease open a new issue if you are still not getting notified." ]

[]

[ "Hi @mariosasko,\r\nJust came across this PR and I was wondering if we can use\r\n`description = \"\\n\\n\".join(OrderedDict.fromkeys([info.description for info in dataset_infos]))`\r\n\r\nThis will obviate the need for `unique` and is almost as fast as `set`. We could have used `dict` inplace of `OrderedDict` but it's available 3.7+ onwards", "Hi,\r\n\r\nlet's see what @lhoestq thinks. Although my approach adds more code, it's more readable IMO.", "Yeah, that's true. Your approach is more readable." ]

[]

[ "Hi ! Indeed we used to require mapping functions to be picklable with `pickle` or `dill` in order to cache the resulting datasets. And FAISS indexes are not picklable unfortunately.\r\n\r\nBut since #1703 this is no longer required (the caching will simply be disabled). This change will be available in the next release of `datasets`, or you can also install `datasets` from source.", "I totally forgot to answer this issue, I'm so sorry. \r\n\r\nI was able to get it working by installing `datasets` from source. Huge thanks!" ]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[ "Thanks for reporting @liaeh, we are investigating this. " ]

[]

[ "Thanks for reporting ! I'm looking into it." ]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[]

[]

[ "Note:\r\n```\r\nnlp-cli test datasets/nlp/<your-dataset-folder> --save_checksums --all_configs\r\n```\r\ndirectly saves the checksums in the right place, and runs for all the dataset configurations.", "@patrickvonplaten can you provide the add the link to the PR for the dummy data? ", "https://github.com/huggingface/nlp/pull/15 - But it's probably best to checkout into this branch and look how the dummy data strtucture is for `squad` for example.", "are lock files supposed to stay ?", "> are lock files supposed to stay ?\r\n\r\nNot sure! I think the checksum command creates them, so I just uploaded them as well.", "We can trash the `lock` file, they are dummy file that are only used to avoid concurrent access when the library is run.\r\nYou can read the filelock readme and code, it's a very simple single-file library: https://github.com/benediktschmitt/py-filelock", "The testing design was slightly changed as explained in https://github.com/huggingface/nlp/pull/51 . \r\nIf creating the dummy folder is too confusing it helps to upload everything else to AWS, then run the test and check the INFO when testing on how to create the dummy folder structure.", "Closing because we can now work on master" ]

[]

[]

[ { "color": "a2eeef", "default": true, "description": "New feature or request", "id": 1935892871, "name": "enhancement", "node_id": "MDU6TGFiZWwxOTM1ODkyODcx", "url": "https://api.github.com/repos/huggingface/datasets/labels/enhancement" } ]

[ "Hi ! Data streaming uses `fsspec` to read the data files progressively. IIRC the block size for buffering is 5MiB by default. So every time you finish iterating over a block, it downloads the next one. You can still try to increase the `fsspec` block size for buffering if it can help. To do so you just need to increase `fsspec.spec.AbstractBufferedFile.DEFAULT_BLOCK_SIZE `\r\n\r\nCurrently this is unfortunately done in a single thread, so it blocks the processing to download and uncompress the next block. At one point it would be nice to be able to do that in parallel !", "Hi! Thanks for the help, I will try it :)" ]

[]

[ "```\r\n(env) cwarny@Cedrics-Air datasets % RUN_SLOW=1 pytest tests/test_dataset_common.py::LocalDatasetTest::test_load_real_dataset_mbpp\r\n================================================================================================ test session starts =================================================================================================\r\nplatform darwin -- Python 3.8.13, pytest-7.1.3, pluggy-1.0.0\r\nrootdir: /Users/cwarny/datasets, configfile: setup.cfg\r\ncollected 1 item \r\n\r\ntests/test_dataset_common.py . [100%]\r\n\r\n================================================================================================= 1 passed in 1.12s ==================================================================================================\r\n(env) cwarny@Cedrics-Air datasets % RUN_SLOW=1 pytest tests/test_dataset_common.py::LocalDatasetTest::test_load_dataset_all_configs_mbpp \r\n================================================================================================ test session starts =================================================================================================\r\nplatform darwin -- Python 3.8.13, pytest-7.1.3, pluggy-1.0.0\r\nrootdir: /Users/cwarny/datasets, configfile: setup.cfg\r\ncollected 1 item \r\n\r\ntests/test_dataset_common.py . [100%]\r\n\r\n================================================================================================= 1 passed in 0.35s ==================================================================================================\r\n\r\n```", "_The documentation is not available anymore as the PR was closed or merged._", "Hi @cwarny ! Thanks for adding the correct splits :)\r\n\r\nYou can fix the CI error by running `make style` - this should reformat the dataset script", "done" ]

[ { "color": "a2eeef", "default": true, "description": "New feature or request", "id": 1935892871, "name": "enhancement", "node_id": "MDU6TGFiZWwxOTM1ODkyODcx", "url": "https://api.github.com/repos/huggingface/datasets/labels/enhancement" } ]

[ "Hi! Our versioning is based on Git revisions (the `revision` param in `load_dataset`), so you can just replace the old URL with the new one and open a PR :). I can also give you some pointers if needed.", "@mariosasko Awesome thanks! I will do that. Looks like this new version of the data is not available as a zip but as three files (train/dev/test). How is this usually handled in HF Datasets, should `_URL` be a dict with keys `train`, `val`, `test` perhaps?", "Yes! Let me help you with more detailed instructions.\r\n\r\nIn the first step, we need to update the URLs. One of the possible dictionary structures is as follows:\r\n```python\r\n_URLs = {\r\n \"train\": {\"src\": \"https://drive.google.com/uc?export=download&id=1wHAWDOwOoQWSj7HYpyJ3Aeud8WhhaJ7P\", \"tgt\": \"https://drive.google.com/uc?export=download&id=1QVgswwhVTkd3VLCzajK6eVkcrSWEK6kq\"}\r\n \"val\": ...\r\n \"test\": ...\r\n}\r\n```\r\n\r\n(You can use this page to generate direct download links: https://sites.google.com/site/gdocs2direct/)\r\n\r\nThen we move to the `split_generators` method:\r\n```python\r\ndef _split_generators(self, dl_manager):\r\n \"\"\"Returns SplitGenerators.\"\"\"\r\n files = dl_manager.download(_URLs)\r\n return [\r\n datasets.SplitGenerator(\r\n name=datasets.Split.TRAIN,\r\n gen_kwargs={\"src_file\": files[\"train\"][\"src\"], \"tgt_file\": files[\"train\"][\"tgt\"]},\r\n ),\r\n ... # same for val and test\r\n ]\r\n```\r\nFinally, we adjust the signature of `_generate_examples`:\r\n```python\r\ndef _generate_examples(self, src_file, tgt_file):\r\n \"\"\"Yields examples.\"\"\"\r\n with open(src_file, encoding=\"utf-8\") as src_f, open(\r\n tgt_file, encoding=\"utf-8\"\r\n ) as tgt_f:\r\n ... # the rest is the same\r\n```\r\n\r\nAnd that's it!\r\n\r\nPS: Let me know if you need help updating the dummy data and regenerating the metadata file.", "Awesome! Thanks for the detailed help, that was straightforward with your instruction. However, I think I am being blocked by this issue: https://github.com/huggingface/datasets/issues/4428", "Feel free to open a PR, and I can fix this manually.", "Awsome, done in #4451!" ]

[ { "color": "e99695", "default": false, "description": "Requesting to add a new dataset", "id": 2067376369, "name": "dataset request", "node_id": "MDU6TGFiZWwyMDY3Mzc2MzY5", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20request" } ]

[]

[]

[ "THanks!\r\nDone!" ]

[]

[ "Hi ! This is not possible just with `load_dataset`.\r\n\r\nYou can do something like this instead:\r\n```python\r\nseed=42\r\ndata_files_dict = {\r\n \"train\": [train_file1, train_file2],\r\n \"test\": [test_file1, test_file2],\r\n \"val\": [val_file1, val_file2]\r\n}\r\ndataset = datasets.load_dataset(\r\n \"csv\",\r\n data_files=data_files_dict,\r\n).shuffle(seed=seed)\r\n\r\nsample_dataset = {splitname: split.select(range(8)) for splitname, split in dataset.items()}\r\n```\r\n\r\nAnother alternative is loading each file separately with `split=\"train[:8]\"` and then use `concatenate_datasets` to merge the sample of each file." ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

[]

[ "Looks like a duplicate of #120.\r\nThis is already fixed on master. We'll do a new release on pypi soon", "Good catch, it looks fixed.\r\n" ]

[]

[ "> What if people want to run some tests without having zstandard ?\r\n> Usually what we do is add a decorator @require_zstandard for example\r\n\r\n@lhoestq I think I'm missing something here...\r\n\r\nTests are a *development* tool (to ensure we deliver a good quality lib), not something we offer to the end users of the lib. Users of the lib just `pip install datasets` and no tests are delivered with the lib (`tests` directory is outside the `src` code dir). \r\n\r\nOn the contrary, developers (contributors) of the lib do need to be able to run tests (TDD). And because of that, they are required to install datasets differently: `pip install -e .[dev]`, so that all required developing (and testing) dependencies are properly installed (included `zstandard`).\r\n\r\nApart from `zsatandard`, there are many other dev/test required dependencies for running tests, and we do not have a `@require_toto` for each and every of these dependencies in our tests: \r\n- `pytest` and `absl-py` (they are not dependencies in install_requires, but only in TEST_REQUIRE extras_require), \r\n- `boto3` (in test_filesystem.py), \r\n- `seqeval` (in test_metric_common.py), \r\n- `bs4` (used by eli5 and tested in test_hf_gcp.py)\r\n- ...\r\n\r\nSo IMHO, to run tests you should previously install datasets with dev or tests dependencies: either `pip install -e .[dev]` or `pip install -e .[tests]` (the latter to be used in CI testing-only part of the development cycle). And the tests should be written accordingly, assuming all tests dependencies are installed.", "Hi !\r\nI was saying that because the other dependencies you mentioned are only required for _some_ tests. While here zstd is required for _all_ tests since it's imported in the conftest.py\r\nFeel free to keep it as it is right now, or maybe move the fixture to test_file_utils.py to allow users without zstd to run tests for their builders, dataset card etc. without issues", "Thank you ! I think we can merge now", "@lhoestq does this mean that the pile could have streaming support in the future? Afaik streaming doesnt support zstandard compressed type", "> @lhoestq does this mean that the pile could have streaming support in the future? Afaik streaming doesnt support zstandard compressed type\r\n\r\njust for reference, i tried to stream one of the `.zst` files from [the pile](https://the-eye.eu/public/AI/pile/) using\r\n\r\n```python\r\ndata_files = [\"https://the-eye.eu/public/AI/pile/train/00.jsonl.zst\"]\r\nstreamed_dataset = load_dataset('json', split='train', data_files=data_files, streaming=True)\r\n```\r\n\r\nand got the following error:\r\n\r\n```\r\nUsing custom data configuration default-4e71acadc389c254\r\n---------------------------------------------------------------------------\r\nNotImplementedError Traceback (most recent call last)\r\n/tmp/ipykernel_1187680/10848115.py in <module>\r\n 1 data_files = [\"https://the-eye.eu/public/AI/pile/train/00.jsonl.zst\"]\r\n 2 \r\n----> 3 streamed_dataset = load_dataset('json', split='train', data_files=data_files, streaming=True)\r\n 4 \r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/load.py in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, ignore_verifications, keep_in_memory, save_infos, script_version, use_auth_token, task, streaming, **config_kwargs)\r\n 835 # this extends the open and os.path.join functions for data streaming\r\n 836 extend_module_for_streaming(builder_instance.__module__, use_auth_token=use_auth_token)\r\n--> 837 return builder_instance.as_streaming_dataset(\r\n 838 split=split,\r\n 839 use_auth_token=use_auth_token,\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/builder.py in as_streaming_dataset(self, split, base_path, use_auth_token)\r\n 922 data_dir=self.config.data_dir,\r\n 923 )\r\n--> 924 splits_generators = {sg.name: sg for sg in self._split_generators(dl_manager)}\r\n 925 # By default, return all splits\r\n 926 if split is None:\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/packaged_modules/json/json.py in _split_generators(self, dl_manager)\r\n 50 if not self.config.data_files:\r\n 51 raise ValueError(f\"At least one data file must be specified, but got data_files={self.config.data_files}\")\r\n---> 52 data_files = dl_manager.download_and_extract(self.config.data_files)\r\n 53 if isinstance(data_files, (str, list, tuple)):\r\n 54 files = data_files\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/utils/streaming_download_manager.py in download_and_extract(self, url_or_urls)\r\n 140 \r\n 141 def download_and_extract(self, url_or_urls):\r\n--> 142 return self.extract(self.download(url_or_urls))\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/utils/streaming_download_manager.py in extract(self, path_or_paths)\r\n 115 \r\n 116 def extract(self, path_or_paths):\r\n--> 117 urlpaths = map_nested(self._extract, path_or_paths, map_tuple=True)\r\n 118 return urlpaths\r\n 119 \r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/utils/py_utils.py in map_nested(function, data_struct, dict_only, map_list, map_tuple, map_numpy, num_proc, types)\r\n 202 num_proc = 1\r\n 203 if num_proc <= 1 or len(iterable) <= num_proc:\r\n--> 204 mapped = [\r\n 205 _single_map_nested((function, obj, types, None, True))\r\n 206 for obj in utils.tqdm(iterable, disable=disable_tqdm)\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/utils/py_utils.py in <listcomp>(.0)\r\n 203 if num_proc <= 1 or len(iterable) <= num_proc:\r\n 204 mapped = [\r\n--> 205 _single_map_nested((function, obj, types, None, True))\r\n 206 for obj in utils.tqdm(iterable, disable=disable_tqdm)\r\n 207 ]\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/utils/py_utils.py in _single_map_nested(args)\r\n 141 # Singleton first to spare some computation\r\n 142 if not isinstance(data_struct, dict) and not isinstance(data_struct, types):\r\n--> 143 return function(data_struct)\r\n 144 \r\n 145 # Reduce logging to keep things readable in multiprocessing with tqdm\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/utils/streaming_download_manager.py in _extract(self, urlpath)\r\n 119 \r\n 120 def _extract(self, urlpath):\r\n--> 121 protocol = self._get_extraction_protocol(urlpath)\r\n 122 if protocol is None:\r\n 123 # no extraction\r\n\r\n~/miniconda3/envs/hf/lib/python3.8/site-packages/datasets/utils/streaming_download_manager.py in _get_extraction_protocol(self, urlpath)\r\n 137 elif path.endswith(\".zip\"):\r\n 138 return \"zip\"\r\n--> 139 raise NotImplementedError(f\"Extraction protocol for file at {urlpath} is not implemented yet\")\r\n 140 \r\n 141 def download_and_extract(self, url_or_urls):\r\n\r\nNotImplementedError: Extraction protocol for file at https://the-eye.eu/public/AI/pile/train/00.jsonl.zst is not implemented yet\r\n```\r\n\r\ni'm not sure whether @Shashi456 is referring to a fundamental limitation with \"streaming\" zstandard compression files or simply that we need to support the protocol in the streaming api of `datasets`\r\n\r\n", "@lewtun our streaming mode patches the Python `open` function. I could have a look tomorrow if it is easily implementable for this case.", "@lewtun, I have tested and yes, it is easily implementable. I've created a draft Pull Request with an implementation proposal: #2786.", "thanks a lot @albertvillanova - now i can stream the pile :)" ]

[ { "color": "e99695", "default": false, "description": "Requesting to add a new dataset", "id": 2067376369, "name": "dataset request", "node_id": "MDU6TGFiZWwyMDY3Mzc2MzY5", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20request" }, { "color": "d93f0b", "default": false, "description": "", "id": 2725241052, "name": "speech", "node_id": "MDU6TGFiZWwyNzI1MjQxMDUy", "url": "https://api.github.com/repos/huggingface/datasets/labels/speech" } ]

[ "@patrickvonplaten Could you please help me with how the output text has to be represented in the data? TIMIT has Words, Phonemes and texts. Also has lot on info on the speaker and the dialect. Could you please help me? An example of how to arrange it would be super helpful!\r\n\r\n", "Hey @vrindaprabhu - sure I'll help you :-) Could you open a first PR for TIMIT where you copy-paste more or less the `librispeech_asr` script: https://github.com/huggingface/datasets/blob/28be129db862ec89a87ac9349c64df6b6118aff4/datasets/librispeech_asr/librispeech_asr.py#L93 (obviously replacing all the naming and links correctly...) and then you can list all possible outputs in the features dict: https://github.com/huggingface/datasets/blob/28be129db862ec89a87ac9349c64df6b6118aff4/datasets/librispeech_asr/librispeech_asr.py#L104 (words, phonemes should probably be of kind `datasets.Sequence(datasets.Value(\"string\"))` and texts I think should be of type `\"text\": datasets.Value(\"string\")`.\r\n\r\nWhen you've opened a first PR, I think it'll be much easier for us to take a look together :-) ", "I am sorry! I created the PR [#1903](https://github.com/huggingface/datasets/pull/1903#). Requesting your comments! CircleCI tests are failing, will address them along with your comments!" ]

[]

[]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[]

[]

[]

[]

[ "thanks ! merging now" ]

[]

[]

[]

[ "@mariosasko This is kinda cool! " ]

[]

[ "Which OS are you using ?@abhi1nandy2", "> Which OS are you using ?\r\n\r\nPRETTY_NAME=\"Debian GNU/Linux 9 (stretch)\"\r\nNAME=\"Debian GNU/Linux\"\r\nVERSION_ID=\"9\"\r\nVERSION=\"9 (stretch)\"\r\nVERSION_CODENAME=stretch\r\nID=debian\r\nHOME_URL=\"https://www.debian.org/\"\r\nSUPPORT_URL=\"https://www.debian.org/support\"\r\nBUG_REPORT_URL=\"https://bugs.debian.org/\"", "Do you mind sharing the data you used (or part of it), so I can try to reproduce ?\r\nOr at least some info about the text file you're using ? (size, n of lines, encoding)", "Lot of data, difficult to share. There are 46 shards, each having about 256000 lines. using `file` command gives this - `ASCII text, with very long lines`.", "Ok I see, no problem :) \r\nI'll see what I can do\r\n\r\nCould you just test with one single dummy text file with a few lines to see if you're having the issue ?\r\nAlso which version of `datasets` do you have ?" ]

[]

[ "Seems like the issue still exists,:\r\n`Downloading and preparing dataset clue/chid (download: 127.15 MiB, generated: 259.71 MiB, post-processed: Unknown size, total: 386.86 MiB) to /mnt/cache/tanhaochen/.cache/huggingface/datasets/clue/chid/1.0.0/e55b490cb7809dcd8db31b9a87119f2e2ec87cdc060da8a9ac070b070ca3e379...\r\nTraceback (most recent call last):\r\n File \"/mnt/cache/tanhaochen/PromptCLUE/test_datasets.py\", line 3, in <module>\r\n cluewsc2020 = datasets.load_dataset(\"clue\",\"chid\")\r\n File \"/mnt/cache/tanhaochen/dependencies/datasets/src/datasets/load.py\", line 1667, in load_dataset\r\n builder_instance.download_and_prepare(\r\n File \"/mnt/cache/tanhaochen/dependencies/datasets/src/datasets/builder.py\", line 593, in download_and_prepare\r\n self._download_and_prepare(\r\n File \"/mnt/cache/tanhaochen/dependencies/datasets/src/datasets/builder.py\", line 663, in _download_and_prepare\r\n verify_checksums(\r\n File \"/mnt/cache/tanhaochen/dependencies/datasets/src/datasets/utils/info_utils.py\", line 40, in verify_checksums\r\n raise NonMatchingChecksumError(error_msg + str(bad_urls))\r\ndatasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:\r\n['https://storage.googleapis.com/cluebenchmark/tasks/chid_public.zip']\r\n`", "Hi,\r\n\r\nthe fix hasn't been merged yet (it should be merged early next week)." ]

[]

[ "> Awesome good job !\r\n> \r\n> Could you also add a dataset card using the template guide here : https://github.com/huggingface/datasets/blob/master/templates/README_guide.md\r\n> If you can't fill some fields then just leave `[N/A]`\r\n\r\nQuick amendment: `[N/A]` is for fields that are not relevant: if you can't find the information just leave `[More Information Needed]`", "@lhoestq since datset cards are optional for this sprint I'll add those later. Good for merge.", "Indeed we only require the tags to be added now (the yaml part at the top of the dataset card).\r\nCould you add them please ?\r\nYou can find more infos here : https://github.com/huggingface/datasets/blob/master/ADD_NEW_DATASET.md#tag-the-dataset-and-write-the-dataset-card", "@lhoestq added tags, I'll fill rest of the info after current sprint :)", "The tests are failing tests for other datasets, not this one.", "@lhoestq could you tell me why these tests are failing, they don't seem related to this PR. " ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "<details>\n<summary>Show benchmarks</summary>\n\nPyArrow==8.0.0\n\n<details>\n<summary>Show updated benchmarks!</summary>\n\n### Benchmark: benchmark_array_xd.json\n\n| metric | read_batch_formatted_as_numpy after write_array2d | read_batch_formatted_as_numpy after write_flattened_sequence | read_batch_formatted_as_numpy after write_nested_sequence | read_batch_unformated after write_array2d | read_batch_unformated after write_flattened_sequence | read_batch_unformated after write_nested_sequence | read_col_formatted_as_numpy after write_array2d | read_col_formatted_as_numpy after write_flattened_sequence | read_col_formatted_as_numpy after write_nested_sequence | read_col_unformated after write_array2d | read_col_unformated after write_flattened_sequence | read_col_unformated after write_nested_sequence | read_formatted_as_numpy after write_array2d | read_formatted_as_numpy after write_flattened_sequence | read_formatted_as_numpy after write_nested_sequence | read_unformated after write_array2d | read_unformated after write_flattened_sequence | read_unformated after write_nested_sequence | write_array2d | write_flattened_sequence | write_nested_sequence |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.009725 / 0.011353 (-0.001628) | 0.006014 / 0.011008 (-0.004994) | 0.136039 / 0.038508 (0.097531) | 0.049685 / 0.023109 (0.026576) | 0.492967 / 0.275898 (0.217068) | 0.553775 / 0.323480 (0.230295) | 0.007421 / 0.007986 (-0.000564) | 0.004686 / 0.004328 (0.000357) | 0.106639 / 0.004250 (0.102389) | 0.073483 / 0.037052 (0.036431) | 0.507194 / 0.258489 (0.248705) | 0.535760 / 0.293841 (0.241919) | 0.049666 / 0.128546 (-0.078880) | 0.014139 / 0.075646 (-0.061507) | 0.435459 / 0.419271 (0.016188) | 0.076026 / 0.043533 (0.032493) | 0.454542 / 0.255139 (0.199403) | 0.512724 / 0.283200 (0.229524) | 0.034969 / 0.141683 (-0.106713) | 1.881048 / 1.452155 (0.428893) | 1.959915 / 1.492716 (0.467199) |\n\n### Benchmark: benchmark_getitem\\_100B.json\n\n| metric | get_batch_of\\_1024\\_random_rows | get_batch_of\\_1024\\_rows | get_first_row | get_last_row |\n|--------|---|---|---|---|\n| new / old (diff) | 0.265322 / 0.018006 (0.247316) | 0.573963 / 0.000490 (0.573474) | 0.017493 / 0.000200 (0.017293) | 0.000637 / 0.000054 (0.000582) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.028712 / 0.037411 (-0.008699) | 0.149554 / 0.014526 (0.135029) | 0.130013 / 0.176557 (-0.046544) | 0.203408 / 0.737135 (-0.533727) | 0.144778 / 0.296338 (-0.151561) |\n\n### Benchmark: benchmark_iterating.json\n\n| metric | read 5000 | read 50000 | read_batch 50000 10 | read_batch 50000 100 | read_batch 50000 1000 | read_formatted numpy 5000 | read_formatted pandas 5000 | read_formatted tensorflow 5000 | read_formatted torch 5000 | read_formatted_batch numpy 5000 10 | read_formatted_batch numpy 5000 1000 | shuffled read 5000 | shuffled read 50000 | shuffled read_batch 50000 10 | shuffled read_batch 50000 100 | shuffled read_batch 50000 1000 | shuffled read_formatted numpy 5000 | shuffled read_formatted_batch numpy 5000 10 | shuffled read_formatted_batch numpy 5000 1000 |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.664198 / 0.215209 (0.448989) | 6.418054 / 2.077655 (4.340399) | 2.602338 / 1.504120 (1.098219) | 2.212992 / 1.541195 (0.671797) | 2.214309 / 1.468490 (0.745819) | 0.914772 / 4.584777 (-3.670005) | 5.824831 / 3.745712 (2.079119) | 2.865381 / 5.269862 (-2.404481) | 1.906020 / 4.565676 (-2.659657) | 0.106947 / 0.424275 (-0.317328) | 0.013467 / 0.007607 (0.005860) | 0.834556 / 0.226044 (0.608512) | 8.237078 / 2.268929 (5.968150) | 3.380919 / 55.444624 (-52.063705) | 2.656713 / 6.876477 (-4.219764) | 2.834941 / 2.142072 (0.692869) | 1.151241 / 4.805227 (-3.653986) | 0.220860 / 6.500664 (-6.279804) | 0.080781 / 0.075469 (0.005312) |\n\n### Benchmark: benchmark_map_filter.json\n\n| metric | filter | map fast-tokenizer batched | map identity | map identity batched | map no-op batched | map no-op batched numpy | map no-op batched pandas | map no-op batched pytorch | map no-op batched tensorflow |\n|--------|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 1.655128 / 1.841788 (-0.186660) | 18.696108 / 8.074308 (10.621800) | 22.882108 / 10.191392 (12.690716) | 0.236041 / 0.680424 (-0.444383) | 0.031073 / 0.534201 (-0.503128) | 0.525263 / 0.579283 (-0.054021) | 0.632933 / 0.434364 (0.198569) | 0.707228 / 0.540337 (0.166890) | 0.753508 / 1.386936 (-0.633428) |\n\n</details>\nPyArrow==latest\n\n<details>\n<summary>Show updated benchmarks!</summary>\n\n### Benchmark: benchmark_array_xd.json\n\n| metric | read_batch_formatted_as_numpy after write_array2d | read_batch_formatted_as_numpy after write_flattened_sequence | read_batch_formatted_as_numpy after write_nested_sequence | read_batch_unformated after write_array2d | read_batch_unformated after write_flattened_sequence | read_batch_unformated after write_nested_sequence | read_col_formatted_as_numpy after write_array2d | read_col_formatted_as_numpy after write_flattened_sequence | read_col_formatted_as_numpy after write_nested_sequence | read_col_unformated after write_array2d | read_col_unformated after write_flattened_sequence | read_col_unformated after write_nested_sequence | read_formatted_as_numpy after write_array2d | read_formatted_as_numpy after write_flattened_sequence | read_formatted_as_numpy after write_nested_sequence | read_unformated after write_array2d | read_unformated after write_flattened_sequence | read_unformated after write_nested_sequence | write_array2d | write_flattened_sequence | write_nested_sequence |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.009875 / 0.011353 (-0.001478) | 0.005135 / 0.011008 (-0.005873) | 0.101307 / 0.038508 (0.062799) | 0.044895 / 0.023109 (0.021786) | 0.497824 / 0.275898 (0.221926) | 0.573098 / 0.323480 (0.249618) | 0.006669 / 0.007986 (-0.001317) | 0.004289 / 0.004328 (-0.000039) | 0.105824 / 0.004250 (0.101573) | 0.061002 / 0.037052 (0.023950) | 0.510127 / 0.258489 (0.251638) | 0.581387 / 0.293841 (0.287546) | 0.052843 / 0.128546 (-0.075703) | 0.015506 / 0.075646 (-0.060140) | 0.116057 / 0.419271 (-0.303215) | 0.063444 / 0.043533 (0.019912) | 0.479366 / 0.255139 (0.224227) | 0.518419 / 0.283200 (0.235220) | 0.034876 / 0.141683 (-0.106806) | 2.018446 / 1.452155 (0.566292) | 1.960755 / 1.492716 (0.468039) |\n\n### Benchmark: benchmark_getitem\\_100B.json\n\n| metric | get_batch_of\\_1024\\_random_rows | get_batch_of\\_1024\\_rows | get_first_row | get_last_row |\n|--------|---|---|---|---|\n| new / old (diff) | 0.269077 / 0.018006 (0.251070) | 0.606059 / 0.000490 (0.605569) | 0.000488 / 0.000200 (0.000288) | 0.000093 / 0.000054 (0.000038) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.032465 / 0.037411 (-0.004946) | 0.136517 / 0.014526 (0.121991) | 0.147740 / 0.176557 (-0.028816) | 0.193802 / 0.737135 (-0.543334) | 0.151876 / 0.296338 (-0.144462) |\n\n### Benchmark: benchmark_iterating.json\n\n| metric | read 5000 | read 50000 | read_batch 50000 10 | read_batch 50000 100 | read_batch 50000 1000 | read_formatted numpy 5000 | read_formatted pandas 5000 | read_formatted tensorflow 5000 | read_formatted torch 5000 | read_formatted_batch numpy 5000 10 | read_formatted_batch numpy 5000 1000 | shuffled read 5000 | shuffled read 50000 | shuffled read_batch 50000 10 | shuffled read_batch 50000 100 | shuffled read_batch 50000 1000 | shuffled read_formatted numpy 5000 | shuffled read_formatted_batch numpy 5000 10 | shuffled read_formatted_batch numpy 5000 1000 |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.709866 / 0.215209 (0.494657) | 6.848193 / 2.077655 (4.770538) | 3.310853 / 1.504120 (1.806733) | 2.940813 / 1.541195 (1.399619) | 2.934934 / 1.468490 (1.466444) | 0.927104 / 4.584777 (-3.657673) | 5.921607 / 3.745712 (2.175895) | 4.926558 / 5.269862 (-0.343303) | 2.853269 / 4.565676 (-1.712407) | 0.120278 / 0.424275 (-0.303998) | 0.015468 / 0.007607 (0.007861) | 0.820509 / 0.226044 (0.594464) | 8.263136 / 2.268929 (5.994208) | 3.780214 / 55.444624 (-51.664410) | 3.108482 / 6.876477 (-3.767995) | 3.101544 / 2.142072 (0.959471) | 1.165539 / 4.805227 (-3.639688) | 0.229215 / 6.500664 (-6.271449) | 0.079862 / 0.075469 (0.004393) |\n\n### Benchmark: benchmark_map_filter.json\n\n| metric | filter | map fast-tokenizer batched | map identity | map identity batched | map no-op batched | map no-op batched numpy | map no-op batched pandas | map no-op batched pytorch | map no-op batched tensorflow |\n|--------|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 1.775071 / 1.841788 (-0.066717) | 19.327621 / 8.074308 (11.253313) | 23.057537 / 10.191392 (12.866145) | 0.250649 / 0.680424 (-0.429775) | 0.029767 / 0.534201 (-0.504434) | 0.554774 / 0.579283 (-0.024509) | 0.651919 / 0.434364 (0.217555) | 0.651641 / 0.540337 (0.111304) | 0.762386 / 1.386936 (-0.624550) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#fdc3ce7060366f480621e8640903c9ab476164e7 \"CML watermark\")\n", "<details>\n<summary>Show benchmarks</summary>\n\nPyArrow==8.0.0\n\n<details>\n<summary>Show updated benchmarks!</summary>\n\n### Benchmark: benchmark_array_xd.json\n\n| metric | read_batch_formatted_as_numpy after write_array2d | read_batch_formatted_as_numpy after write_flattened_sequence | read_batch_formatted_as_numpy after write_nested_sequence | read_batch_unformated after write_array2d | read_batch_unformated after write_flattened_sequence | read_batch_unformated after write_nested_sequence | read_col_formatted_as_numpy after write_array2d | read_col_formatted_as_numpy after write_flattened_sequence | read_col_formatted_as_numpy after write_nested_sequence | read_col_unformated after write_array2d | read_col_unformated after write_flattened_sequence | read_col_unformated after write_nested_sequence | read_formatted_as_numpy after write_array2d | read_formatted_as_numpy after write_flattened_sequence | read_formatted_as_numpy after write_nested_sequence | read_unformated after write_array2d | read_unformated after write_flattened_sequence | read_unformated after write_nested_sequence | write_array2d | write_flattened_sequence | write_nested_sequence |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.005997 / 0.011353 (-0.005356) | 0.003892 / 0.011008 (-0.007116) | 0.098020 / 0.038508 (0.059512) | 0.042584 / 0.023109 (0.019475) | 0.317909 / 0.275898 (0.042011) | 0.395042 / 0.323480 (0.071563) | 0.005358 / 0.007986 (-0.002628) | 0.003266 / 0.004328 (-0.001062) | 0.076698 / 0.004250 (0.072447) | 0.062331 / 0.037052 (0.025279) | 0.334900 / 0.258489 (0.076411) | 0.379355 / 0.293841 (0.085514) | 0.030815 / 0.128546 (-0.097731) | 0.008596 / 0.075646 (-0.067050) | 0.327739 / 0.419271 (-0.091533) | 0.054061 / 0.043533 (0.010528) | 0.311044 / 0.255139 (0.055905) | 0.336705 / 0.283200 (0.053506) | 0.022785 / 0.141683 (-0.118898) | 1.516793 / 1.452155 (0.064639) | 1.590435 / 1.492716 (0.097719) |\n\n### Benchmark: benchmark_getitem\\_100B.json\n\n| metric | get_batch_of\\_1024\\_random_rows | get_batch_of\\_1024\\_rows | get_first_row | get_last_row |\n|--------|---|---|---|---|\n| new / old (diff) | 0.289157 / 0.018006 (0.271151) | 0.531074 / 0.000490 (0.530585) | 0.004672 / 0.000200 (0.004472) | 0.000095 / 0.000054 (0.000040) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.026173 / 0.037411 (-0.011238) | 0.105723 / 0.014526 (0.091197) | 0.118010 / 0.176557 (-0.058547) | 0.178062 / 0.737135 (-0.559073) | 0.120059 / 0.296338 (-0.176279) |\n\n### Benchmark: benchmark_iterating.json\n\n| metric | read 5000 | read 50000 | read_batch 50000 10 | read_batch 50000 100 | read_batch 50000 1000 | read_formatted numpy 5000 | read_formatted pandas 5000 | read_formatted tensorflow 5000 | read_formatted torch 5000 | read_formatted_batch numpy 5000 10 | read_formatted_batch numpy 5000 1000 | shuffled read 5000 | shuffled read 50000 | shuffled read_batch 50000 10 | shuffled read_batch 50000 100 | shuffled read_batch 50000 1000 | shuffled read_formatted numpy 5000 | shuffled read_formatted_batch numpy 5000 10 | shuffled read_formatted_batch numpy 5000 1000 |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.410870 / 0.215209 (0.195661) | 4.042183 / 2.077655 (1.964528) | 1.830059 / 1.504120 (0.325939) | 1.638996 / 1.541195 (0.097802) | 1.701368 / 1.468490 (0.232878) | 0.529915 / 4.584777 (-4.054861) | 3.693308 / 3.745712 (-0.052404) | 1.827875 / 5.269862 (-3.441986) | 1.063237 / 4.565676 (-3.502440) | 0.065368 / 0.424275 (-0.358907) | 0.010986 / 0.007607 (0.003379) | 0.509399 / 0.226044 (0.283354) | 5.092739 / 2.268929 (2.823810) | 2.293490 / 55.444624 (-53.151135) | 1.958742 / 6.876477 (-4.917735) | 2.024985 / 2.142072 (-0.117088) | 0.646978 / 4.805227 (-4.158249) | 0.138616 / 6.500664 (-6.362048) | 0.062101 / 0.075469 (-0.013368) |\n\n### Benchmark: benchmark_map_filter.json\n\n| metric | filter | map fast-tokenizer batched | map identity | map identity batched | map no-op batched | map no-op batched numpy | map no-op batched pandas | map no-op batched pytorch | map no-op batched tensorflow |\n|--------|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 1.202016 / 1.841788 (-0.639772) | 14.493204 / 8.074308 (6.418896) | 12.992160 / 10.191392 (2.800768) | 0.188922 / 0.680424 (-0.491502) | 0.017594 / 0.534201 (-0.516606) | 0.399917 / 0.579283 (-0.179367) | 0.429760 / 0.434364 (-0.004604) | 0.497906 / 0.540337 (-0.042431) | 0.608745 / 1.386936 (-0.778191) |\n\n</details>\nPyArrow==latest\n\n<details>\n<summary>Show updated benchmarks!</summary>\n\n### Benchmark: benchmark_array_xd.json\n\n| metric | read_batch_formatted_as_numpy after write_array2d | read_batch_formatted_as_numpy after write_flattened_sequence | read_batch_formatted_as_numpy after write_nested_sequence | read_batch_unformated after write_array2d | read_batch_unformated after write_flattened_sequence | read_batch_unformated after write_nested_sequence | read_col_formatted_as_numpy after write_array2d | read_col_formatted_as_numpy after write_flattened_sequence | read_col_formatted_as_numpy after write_nested_sequence | read_col_unformated after write_array2d | read_col_unformated after write_flattened_sequence | read_col_unformated after write_nested_sequence | read_formatted_as_numpy after write_array2d | read_formatted_as_numpy after write_flattened_sequence | read_formatted_as_numpy after write_nested_sequence | read_unformated after write_array2d | read_unformated after write_flattened_sequence | read_unformated after write_nested_sequence | write_array2d | write_flattened_sequence | write_nested_sequence |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.006164 / 0.011353 (-0.005189) | 0.003980 / 0.011008 (-0.007028) | 0.074676 / 0.038508 (0.036168) | 0.041337 / 0.023109 (0.018228) | 0.400981 / 0.275898 (0.125083) | 0.448791 / 0.323480 (0.125312) | 0.004063 / 0.007986 (-0.003923) | 0.004443 / 0.004328 (0.000114) | 0.075011 / 0.004250 (0.070760) | 0.056494 / 0.037052 (0.019441) | 0.402054 / 0.258489 (0.143565) | 0.446122 / 0.293841 (0.152281) | 0.031752 / 0.128546 (-0.096794) | 0.008835 / 0.075646 (-0.066811) | 0.081226 / 0.419271 (-0.338046) | 0.051501 / 0.043533 (0.007969) | 0.383674 / 0.255139 (0.128535) | 0.405524 / 0.283200 (0.122325) | 0.025929 / 0.141683 (-0.115754) | 1.492985 / 1.452155 (0.040830) | 1.541601 / 1.492716 (0.048885) |\n\n### Benchmark: benchmark_getitem\\_100B.json\n\n| metric | get_batch_of\\_1024\\_random_rows | get_batch_of\\_1024\\_rows | get_first_row | get_last_row |\n|--------|---|---|---|---|\n| new / old (diff) | 0.305149 / 0.018006 (0.287142) | 0.497259 / 0.000490 (0.496770) | 0.000420 / 0.000200 (0.000220) | 0.000056 / 0.000054 (0.000002) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.027933 / 0.037411 (-0.009479) | 0.111900 / 0.014526 (0.097374) | 0.124879 / 0.176557 (-0.051678) | 0.178952 / 0.737135 (-0.558184) | 0.127698 / 0.296338 (-0.168640) |\n\n### Benchmark: benchmark_iterating.json\n\n| metric | read 5000 | read 50000 | read_batch 50000 10 | read_batch 50000 100 | read_batch 50000 1000 | read_formatted numpy 5000 | read_formatted pandas 5000 | read_formatted tensorflow 5000 | read_formatted torch 5000 | read_formatted_batch numpy 5000 10 | read_formatted_batch numpy 5000 1000 | shuffled read 5000 | shuffled read 50000 | shuffled read_batch 50000 10 | shuffled read_batch 50000 100 | shuffled read_batch 50000 1000 | shuffled read_formatted numpy 5000 | shuffled read_formatted_batch numpy 5000 10 | shuffled read_formatted_batch numpy 5000 1000 |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.448525 / 0.215209 (0.233316) | 4.486791 / 2.077655 (2.409137) | 2.256687 / 1.504120 (0.752567) | 2.061078 / 1.541195 (0.519884) | 2.078924 / 1.468490 (0.610434) | 0.534412 / 4.584777 (-4.050365) | 3.721098 / 3.745712 (-0.024614) | 1.818735 / 5.269862 (-3.451127) | 1.104198 / 4.565676 (-3.461479) | 0.066277 / 0.424275 (-0.357998) | 0.011441 / 0.007607 (0.003834) | 0.550140 / 0.226044 (0.324095) | 5.498079 / 2.268929 (3.229150) | 2.717398 / 55.444624 (-52.727227) | 2.410194 / 6.876477 (-4.466283) | 2.405304 / 2.142072 (0.263231) | 0.665432 / 4.805227 (-4.139796) | 0.141488 / 6.500664 (-6.359177) | 0.064051 / 0.075469 (-0.011419) |\n\n### Benchmark: benchmark_map_filter.json\n\n| metric | filter | map fast-tokenizer batched | map identity | map identity batched | map no-op batched | map no-op batched numpy | map no-op batched pandas | map no-op batched pytorch | map no-op batched tensorflow |\n|--------|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 1.272334 / 1.841788 (-0.569454) | 14.901608 / 8.074308 (6.827300) | 14.287857 / 10.191392 (4.096465) | 0.165337 / 0.680424 (-0.515086) | 0.017402 / 0.534201 (-0.516799) | 0.398120 / 0.579283 (-0.181163) | 0.416539 / 0.434364 (-0.017825) | 0.463890 / 0.540337 (-0.076447) | 0.567909 / 1.386936 (-0.819027) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#504ec0f2e00ee38e0993ed1e4f1e10f1eefaea0d \"CML watermark\")\n", "<details>\n<summary>Show benchmarks</summary>\n\nPyArrow==8.0.0\n\n<details>\n<summary>Show updated benchmarks!</summary>\n\n### Benchmark: benchmark_array_xd.json\n\n| metric | read_batch_formatted_as_numpy after write_array2d | read_batch_formatted_as_numpy after write_flattened_sequence | read_batch_formatted_as_numpy after write_nested_sequence | read_batch_unformated after write_array2d | read_batch_unformated after write_flattened_sequence | read_batch_unformated after write_nested_sequence | read_col_formatted_as_numpy after write_array2d | read_col_formatted_as_numpy after write_flattened_sequence | read_col_formatted_as_numpy after write_nested_sequence | read_col_unformated after write_array2d | read_col_unformated after write_flattened_sequence | read_col_unformated after write_nested_sequence | read_formatted_as_numpy after write_array2d | read_formatted_as_numpy after write_flattened_sequence | read_formatted_as_numpy after write_nested_sequence | read_unformated after write_array2d | read_unformated after write_flattened_sequence | read_unformated after write_nested_sequence | write_array2d | write_flattened_sequence | write_nested_sequence |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.009434 / 0.011353 (-0.001919) | 0.005567 / 0.011008 (-0.005441) | 0.122652 / 0.038508 (0.084144) | 0.050177 / 0.023109 (0.027067) | 0.384292 / 0.275898 (0.108394) | 0.446608 / 0.323480 (0.123128) | 0.006502 / 0.007986 (-0.001484) | 0.004523 / 0.004328 (0.000194) | 0.100581 / 0.004250 (0.096331) | 0.073615 / 0.037052 (0.036563) | 0.420179 / 0.258489 (0.161690) | 0.474631 / 0.293841 (0.180790) | 0.047942 / 0.128546 (-0.080604) | 0.013864 / 0.075646 (-0.061783) | 0.419384 / 0.419271 (0.000112) | 0.088317 / 0.043533 (0.044784) | 0.379620 / 0.255139 (0.124481) | 0.412639 / 0.283200 (0.129440) | 0.048947 / 0.141683 (-0.092736) | 1.823498 / 1.452155 (0.371343) | 1.966629 / 1.492716 (0.473913) |\n\n### Benchmark: benchmark_getitem\\_100B.json\n\n| metric | get_batch_of\\_1024\\_random_rows | get_batch_of\\_1024\\_rows | get_first_row | get_last_row |\n|--------|---|---|---|---|\n| new / old (diff) | 0.300669 / 0.018006 (0.282663) | 0.593499 / 0.000490 (0.593009) | 0.007247 / 0.000200 (0.007047) | 0.000114 / 0.000054 (0.000059) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.030556 / 0.037411 (-0.006856) | 0.119252 / 0.014526 (0.104726) | 0.131403 / 0.176557 (-0.045153) | 0.201845 / 0.737135 (-0.535291) | 0.139350 / 0.296338 (-0.156989) |\n\n### Benchmark: benchmark_iterating.json\n\n| metric | read 5000 | read 50000 | read_batch 50000 10 | read_batch 50000 100 | read_batch 50000 1000 | read_formatted numpy 5000 | read_formatted pandas 5000 | read_formatted tensorflow 5000 | read_formatted torch 5000 | read_formatted_batch numpy 5000 10 | read_formatted_batch numpy 5000 1000 | shuffled read 5000 | shuffled read 50000 | shuffled read_batch 50000 10 | shuffled read_batch 50000 100 | shuffled read_batch 50000 1000 | shuffled read_formatted numpy 5000 | shuffled read_formatted_batch numpy 5000 10 | shuffled read_formatted_batch numpy 5000 1000 |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.652400 / 0.215209 (0.437191) | 6.536540 / 2.077655 (4.458886) | 2.644565 / 1.504120 (1.140445) | 2.245181 / 1.541195 (0.703986) | 2.316030 / 1.468490 (0.847540) | 0.922535 / 4.584777 (-3.662242) | 5.469065 / 3.745712 (1.723353) | 2.800489 / 5.269862 (-2.469373) | 1.749042 / 4.565676 (-2.816635) | 0.108444 / 0.424275 (-0.315831) | 0.015651 / 0.007607 (0.008044) | 0.846085 / 0.226044 (0.620041) | 8.018460 / 2.268929 (5.749531) | 3.338710 / 55.444624 (-52.105914) | 2.675998 / 6.876477 (-4.200479) | 2.918550 / 2.142072 (0.776478) | 1.135145 / 4.805227 (-3.670082) | 0.215165 / 6.500664 (-6.285499) | 0.082066 / 0.075469 (0.006597) |\n\n### Benchmark: benchmark_map_filter.json\n\n| metric | filter | map fast-tokenizer batched | map identity | map identity batched | map no-op batched | map no-op batched numpy | map no-op batched pandas | map no-op batched pytorch | map no-op batched tensorflow |\n|--------|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 1.561661 / 1.841788 (-0.280127) | 18.519035 / 8.074308 (10.444727) | 19.046300 / 10.191392 (8.854908) | 0.236890 / 0.680424 (-0.443534) | 0.027681 / 0.534201 (-0.506520) | 0.511998 / 0.579283 (-0.067285) | 0.591627 / 0.434364 (0.157264) | 0.562021 / 0.540337 (0.021683) | 0.679354 / 1.386936 (-0.707582) |\n\n</details>\nPyArrow==latest\n\n<details>\n<summary>Show updated benchmarks!</summary>\n\n### Benchmark: benchmark_array_xd.json\n\n| metric | read_batch_formatted_as_numpy after write_array2d | read_batch_formatted_as_numpy after write_flattened_sequence | read_batch_formatted_as_numpy after write_nested_sequence | read_batch_unformated after write_array2d | read_batch_unformated after write_flattened_sequence | read_batch_unformated after write_nested_sequence | read_col_formatted_as_numpy after write_array2d | read_col_formatted_as_numpy after write_flattened_sequence | read_col_formatted_as_numpy after write_nested_sequence | read_col_unformated after write_array2d | read_col_unformated after write_flattened_sequence | read_col_unformated after write_nested_sequence | read_formatted_as_numpy after write_array2d | read_formatted_as_numpy after write_flattened_sequence | read_formatted_as_numpy after write_nested_sequence | read_unformated after write_array2d | read_unformated after write_flattened_sequence | read_unformated after write_nested_sequence | write_array2d | write_flattened_sequence | write_nested_sequence |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.009643 / 0.011353 (-0.001710) | 0.005768 / 0.011008 (-0.005241) | 0.104430 / 0.038508 (0.065922) | 0.050044 / 0.023109 (0.026935) | 0.464117 / 0.275898 (0.188219) | 0.518439 / 0.323480 (0.194959) | 0.006935 / 0.007986 (-0.001051) | 0.004316 / 0.004328 (-0.000013) | 0.094330 / 0.004250 (0.090080) | 0.071451 / 0.037052 (0.034399) | 0.492248 / 0.258489 (0.233759) | 0.555740 / 0.293841 (0.261899) | 0.047836 / 0.128546 (-0.080711) | 0.014788 / 0.075646 (-0.060859) | 0.107590 / 0.419271 (-0.311682) | 0.064396 / 0.043533 (0.020863) | 0.451529 / 0.255139 (0.196390) | 0.475025 / 0.283200 (0.191826) | 0.040006 / 0.141683 (-0.101677) | 1.797107 / 1.452155 (0.344953) | 1.879261 / 1.492716 (0.386545) |\n\n### Benchmark: benchmark_getitem\\_100B.json\n\n| metric | get_batch_of\\_1024\\_random_rows | get_batch_of\\_1024\\_rows | get_first_row | get_last_row |\n|--------|---|---|---|---|\n| new / old (diff) | 0.298458 / 0.018006 (0.280451) | 0.613022 / 0.000490 (0.612532) | 0.003582 / 0.000200 (0.003382) | 0.000106 / 0.000054 (0.000052) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.030179 / 0.037411 (-0.007232) | 0.123286 / 0.014526 (0.108760) | 0.132070 / 0.176557 (-0.044486) | 0.190883 / 0.737135 (-0.546252) | 0.138526 / 0.296338 (-0.157812) |\n\n### Benchmark: benchmark_iterating.json\n\n| metric | read 5000 | read 50000 | read_batch 50000 10 | read_batch 50000 100 | read_batch 50000 1000 | read_formatted numpy 5000 | read_formatted pandas 5000 | read_formatted tensorflow 5000 | read_formatted torch 5000 | read_formatted_batch numpy 5000 10 | read_formatted_batch numpy 5000 1000 | shuffled read 5000 | shuffled read 50000 | shuffled read_batch 50000 10 | shuffled read_batch 50000 100 | shuffled read_batch 50000 1000 | shuffled read_formatted numpy 5000 | shuffled read_formatted_batch numpy 5000 10 | shuffled read_formatted_batch numpy 5000 1000 |\n|--------|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 0.666908 / 0.215209 (0.451699) | 6.489035 / 2.077655 (4.411381) | 2.897027 / 1.504120 (1.392907) | 2.565150 / 1.541195 (1.023956) | 2.504827 / 1.468490 (1.036336) | 0.916112 / 4.584777 (-3.668665) | 5.651751 / 3.745712 (1.906039) | 2.743382 / 5.269862 (-2.526479) | 1.773338 / 4.565676 (-2.792338) | 0.128764 / 0.424275 (-0.295511) | 0.013140 / 0.007607 (0.005533) | 0.803281 / 0.226044 (0.577236) | 8.258874 / 2.268929 (5.989945) | 3.633260 / 55.444624 (-51.811364) | 2.878827 / 6.876477 (-3.997649) | 2.977178 / 2.142072 (0.835106) | 1.130467 / 4.805227 (-3.674760) | 0.226381 / 6.500664 (-6.274283) | 0.081550 / 0.075469 (0.006081) |\n\n### Benchmark: benchmark_map_filter.json\n\n| metric | filter | map fast-tokenizer batched | map identity | map identity batched | map no-op batched | map no-op batched numpy | map no-op batched pandas | map no-op batched pytorch | map no-op batched tensorflow |\n|--------|---|---|---|---|---|---|---|---|---|\n| new / old (diff) | 1.842927 / 1.841788 (0.001139) | 18.411520 / 8.074308 (10.337212) | 21.118228 / 10.191392 (10.926836) | 0.231526 / 0.680424 (-0.448898) | 0.029300 / 0.534201 (-0.504901) | 0.527450 / 0.579283 (-0.051834) | 0.618873 / 0.434364 (0.184509) | 0.593314 / 0.540337 (0.052976) | 0.734430 / 1.386936 (-0.652506) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#0d2b8854c265b4dc202e480427890f472b34ea15 \"CML watermark\")\n" ]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" } ]

[ "Hi ! Thanks for reporting. I opened a PR to fix this issue: #2201", "Nice question which helped me a lot! I have wasted a lot of time to the `DatasetDict` creation from a csv file. Hope the document of this module add some simple examples.", "Hi :) We're indeed working on tutorials that we will add to the docs !" ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Awesome thanks @mariosasko!" ]

[]

[]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_6016). All of your documentation changes will be reflected on that endpoint.", "It we could have something similar to Polars, that would be great.\r\n\r\nThis is what Polars outputs: \r\n* `__repr__`/`__str__` :\r\n```\r\nshape: (67_349, 3)\r\n┌───────┬───────────────────────────────────┬───────┐\r\n│ idx ┆ sentence ┆ label │\r\n│ --- ┆ --- ┆ --- │\r\n│ i32 ┆ str ┆ i64 │\r\n╞═══════╪═══════════════════════════════════╪═══════╡\r\n│ 0 ┆ hide new secretions from the par… ┆ 0 │\r\n│ 1 ┆ contains no wit , only labored g… ┆ 0 │\r\n│ 2 ┆ that loves its characters and co… ┆ 1 │\r\n│ 3 ┆ remains utterly satisfied to rem… ┆ 0 │\r\n│ … ┆ … ┆ … │\r\n│ 67345 ┆ anguish , anger and frustration ┆ 0 │\r\n│ 67346 ┆ at achieving the modest , crowd-… ┆ 1 │\r\n│ 67347 ┆ a patient viewer ┆ 1 │\r\n│ 67348 ┆ this new jangle of noise , mayhe… ┆ 0 │\r\n└───────┴───────────────────────────────────┴───────┘\r\n```\r\n\r\n* `_repr_html_`:\r\n<img width=\"251\" alt=\"Screenshot 2023-07-12 at 18 25 58\" src=\"https://github.com/huggingface/datasets/assets/47462742/5d04519d-f302-4411-9fbc-7445bdf53b23\">\r\n\r\n" ]

[ { "color": "a2eeef", "default": true, "description": "New feature or request", "id": 1935892871, "name": "enhancement", "node_id": "MDU6TGFiZWwxOTM1ODkyODcx", "url": "https://api.github.com/repos/huggingface/datasets/labels/enhancement" } ]

[ "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.", "Hi ! I just re-ran a quick benchmark and using `to_numpy()` seems to be faster now:\r\n\r\n```python\r\nimport pyarrow as pa # I used pyarrow 3.0.0\r\nimport numpy as np\r\n\r\nn, max_length = 1_000, 512\r\nlow, high, size = 0, 2 << 16, (n, max_length)\r\n\r\ntable = pa.Table.from_pydict({\r\n \"input_ids\": np.random.default_rng(42).integers(low=low, high=high, size=size).tolist()\r\n})\r\n\r\n\r\n%%timeit\r\n_ = table.to_pandas()[\"input_ids\"].to_numpy()\r\n# 1.44 ms ± 80.1 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)\r\n\r\n%%timeit\r\n_ = table[\"input_ids\"].to_pandas().to_numpy()\r\n# 461 µs ± 14.2 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)\r\n\r\n%%timeit\r\n_ = table[\"input_ids\"].to_numpy()\r\n# 317 µs ± 5.06 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)\r\n```\r\n\r\nCurrently the conversion from arrow to numpy is done in the NumpyArrowExtractor here:\r\n\r\nhttps://github.com/huggingface/datasets/blob/d6d0ede9486ffad7944642ca9a326e058b676788/src/datasets/formatting/formatting.py#L143-L166\r\n\r\nLet's update the NumpyArrowExtractor to call `to_numpy` directly and see how our github benchmarks evolve ?__", "Sounds like a plan @lhoestq If you create a PR I'll pick it up and try it out right away! ", "@lhoestq I can also prepare the PR, just lmk. ", "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?", "I created https://github.com/huggingface/datasets/pull/2505 if you want to play with it @vblagoje ", "> 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?\r\n\r\n@thomwolf starting from the top, each rectangle represents the cumulative amount of it takes to execute the method call. Therefore, format_batch in torch_formatter.py takes ~20 sec, and the largest portion of that call is taken by to_pandas call and the smaller portion (grey rectangle) by the other method invocation(s) in format_batch (series_to_numpy etc). \r\n\r\nFeatures of the dataset are BERT pre-training model input columns i.e:\r\n```\r\nf = Features({ \r\n \"input_ids\": Sequence(feature=Value(dtype=\"int32\")), \r\n \"attention_mask\": Sequence(feature=Value(dtype=\"int8\")), \r\n \"token_type_ids\": Sequence(feature=Value(dtype=\"int8\")), \r\n \"labels\": Sequence(feature=Value(dtype=\"int32\")), \r\n \"next_sentence_label\": Value(dtype=\"int8\")\r\n})\r\n```\r\n\r\nI'll work with @lhoestq till we get to the bottom of this one. \r\n ", "@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 traces. ", "Hi @vblagoje, to install Datasets from @lhoestq PR reference #2505, you can use:\r\n```shell\r\npip install git+ssh://[email protected]/huggingface/datasets.git@refs/pull/2505/head#egg=datasets\r\n```", "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.", "@vblagoje in that case, you can add this to your `setup.py`:\r\n```python\r\n install_requires=[\r\n \"datasets @ git+ssh://[email protected]/huggingface/datasets.git@refs/pull/2505/head\",\r\n```", "@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)\r\n\r\n\r\n\r\nWhen we drill down into dataloader next invocation we get:\r\n\r\n\r\n\r\nAnd finally format_batch:\r\n\r\n\r\n\r\n\r\nNot sure this could be further improved but this is definitely a decent step forward.\r\n\r\n", "> ```python\r\n> datasets @ git+ssh://[email protected]/huggingface/datasets.git@refs/pull/2505/head\r\n> ```\r\n\r\n@albertvillanova how would I replace datasets dependency in https://github.com/huggingface/transformers/blob/master/setup.py as the above approach is not working. ", "@vblagoje I tested my proposed approach before posting it here and it worked for me. \r\n\r\nIs it not working in your case because of the SSH protocol? In that case you could try the same approach but using HTTPS:\r\n```\r\n\"datasets @ git+https://github.com/huggingface/datasets.git@refs/pull/2505/head\",\r\n``` ", "Also note the blanks before and after the `@`.", "@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. ", "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." ]

[ { "color": "d73a4a", "default": true, "description": "Something isn't working", "id": 1935892857, "name": "bug", "node_id": "MDU6TGFiZWwxOTM1ODkyODU3", "url": "https://api.github.com/repos/huggingface/datasets/labels/bug" }, { "color": "E5583E", "default": false, "description": "Related to the dataset viewer on huggingface.co", "id": 3470211881, "name": "dataset-viewer", "node_id": "LA_kwDODunzps7O1zsp", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset-viewer" } ]

[]

[ { "color": "E5583E", "default": false, "description": "Related to the dataset viewer on huggingface.co", "id": 3470211881, "name": "dataset-viewer", "node_id": "LA_kwDODunzps7O1zsp", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset-viewer" } ]

[ "Hi ! I think the issue comes from this [line](https://huggingface.co/datasets/liweili/c4_200m/blob/main/c4_200m.py#L87):\r\n```python\r\npath = filepath + \"/*.tsv*\"\r\n```\r\n\r\nYou can fix this by doing this instead:\r\n```python\r\npath = os.path.join(filepath, \"/*.tsv*\")\r\n```\r\n\r\nHere is why:\r\n\r\nLocally you can append `\"/*.tsv*\"` to your local path, however it doesn't work in streaming mode, and the dataset viewer does use the streaming mode.\r\nIn streaming mode, the download and extract part is done lazily. It means that instead of using local paths, it's still passing around URLs and [chained URLs](https://filesystem-spec.readthedocs.io/en/latest/features.html#url-chaining)\r\n\r\nTherefore in streaming mode, `filepath` is not a local path, but instead is equal to\r\n```python\r\nzip://::https://huggingface.co/datasets/liweili/c4_200m/resolve/main/data.zip\r\n```\r\nThe `zip://` part means that we navigate inside the remote ZIP file.\r\n\r\nYou must use `os.path.join` to navigate inside it and get your TSV files:\r\n```python\r\n>>> os.path.join(filepath, \"/*.tsv*\")\r\nzip://*.tsv*::https://huggingface.co/datasets/liweili/c4_200m/resolve/main/data.zip\r\n```\r\n\r\n`datasets` extends `os.path.join`, `glob.glob`, etc. in your dataset scripts to work with remote files.", "hi @lhoestq ! thanks for the tip! i've updated the line of code but it's still not working. am i doing something else wrong? thank you!", "Hi ! Your dataset code is all good now :)\r\n```python\r\nIn [1]: from datasets import load_dataset\r\n\r\nIn [2]: d = load_dataset(\"liweili/c4_200m\", streaming=True)\r\nDownloading: 100%|█████████████████████████████████████████████| 2.79k/2.79k [00:00<00:00, 4.83MB/s]\r\nUsing custom data configuration default\r\n\r\nIn [3]: next(iter(d[\"train\"]))\r\nOut[3]: \r\n{'input': 'Bitcoin is for $7,094 this morning, which CoinDesk says.',\r\n 'output': 'Bitcoin goes for $7,094 this morning, according to CoinDesk.'}\r\n```\r\nThough the viewer doesn't seem to be updated, I'll take a look at what's wrong", "thank you @lhoestq! 😄 ", "It's working\r\n\r\n<img width=\"1424\" alt=\"Capture d’écran 2021-12-21 à 11 24 29\" src=\"https://user-images.githubusercontent.com/1676121/146914238-24bf87c0-c68d-4699-8d6c-fa3065656d1d.png\">\r\n\r\n" ]

[]

[ "The `audiofolder` loader is not available in version `2.3.2`, hence the error. Please run the `pip install -U datasets` command to update the `datasets` installation to make `load_dataset(\"audiofolder\", ...)` work." ]

[]