Datasets:

---

Francesco-A

/

github-issues_huggingface-datasets

like 0

[ { "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" } ]

[ "I want to be able to create folders in a model.", "How to set new split names, instead of train/test/validation? For example, I have a local dataset, consists of several subsets, named \"A\", \"B\", and \"C\". How can I create a huggingface dataset, with splits A/B/C ?\r\n\r\nThe document in https://huggingface.co/docs/datasets/dataset_script only tells me how to create datasets with subsets that is hosted on another server. How to do it if my datasets are local?", "> The document in https://huggingface.co/docs/datasets/dataset_script only tells me how to create datasets with subsets that is hosted on another server. How to do it if my datasets are local?\r\n\r\nIt works the same - you just need to use local paths instead of URLs" ]

[]

[ "Hi @phiwi,\r\n\r\nThanks for contributing this nice dataset. If you have any blocking problem or question, do not hesitate to ask here. We are pleased to help you.\r\n\r\nCould you please first synchronize with our master branch? From your branch `bbaw_egyptian`, type:\r\n```\r\ngit fetch upstream master\r\ngit merge upstream/master\r\n```", "Thanks ! Can you check that you have `black==21.4b0` and run `make style` again ? This should fix the \"check_code_quality\" CI issue", "Reformatted with black.", "Hi @phiwi, there are still some minor problems in relation with the tags you used in the dataset card (README.md).\r\n\r\nHere you can find the output of the metadata validator:\r\n```\r\nWARNING:root:❌ Failed to validate 'datasets/bbaw_egyptian/README.md':\r\nCould not validate the metada, found the following errors:\r\n* field 'size_categories':\r\n\t['100K<n<1000K'] are not registered tags for 'size_categories', reference at https://github.com/huggingface/datasets/tree/master/src/datasets/utils/resources/size_categories.json\r\n* field 'task_ids':\r\n\t['machine translation'] are not registered tags for 'task_ids', reference at https://github.com/huggingface/datasets/tree/master/src/datasets/utils/resources/tasks.json\r\n* field 'languages':\r\n\t['eg'] are not registered tags for 'languages', reference at https://github.com/huggingface/datasets/tree/master/src/datasets/utils/resources/languages.json\r\n\r\n``` ", "@albertvillanova corrected :-)", "Thanks, @phiwi. Now all tests should pass green.\r\n\r\nHowever, I think there is still an issue with the language code:\r\n- the code for the Ancient Egyptian is not `ar-EG`\r\n- there is no ISO 639-1 code for the Ancient Egyptian\r\n- there is an ISO 639-2 code: `egy`; but this code will not pass the validation test because it is not in the list of valid codes\r\n\r\nI am not sure what to do in this case... Maybe @lhoestq has an idea? Maybe adding the code to the list? https://github.com/huggingface/datasets/blob/master/src/datasets/utils/resources/languages.json", "I have just checked that in the [list of valid codes](https://github.com/huggingface/datasets/blob/master/src/datasets/utils/resources/languages.json) there are already ISO 639-2 codes. Therefore, I would suggest you to add it to the list:\r\n```\r\n\"egy\": \"Egyptian (Ancient)\",\r\n```\r\nand change it in the dataset card.", "Done.", "Hope, everything is okay right now." ]

[]

[ "This dataset seems to have around 1000 configs. Therefore when creating the dummy data we end up with hundreds of MB of dummy data which we don't want to add in the repo.\r\nLet's make this PR on hold for now and find a solution after the sprint of next week", "Closing in favor of #1566 " ]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_5885). All of your documentation changes will be reflected on that endpoint.", "@lhoestq would you or another maintainer be able to review please? :)", "Why you do need to support FUSE mounted paths ?\r\n\r\n`datasets` uses data that live on disk for fast lookups - FUSE mounted disks would lead to poor performance and I wouldn't recomment using it.", "Fuse is commonly used to mount remote file systems (e.g. S3, DBFS) as a local directory. Since it's slower than using an actual local device, it's better to treat it as remote to reduce latency.", "I think people would be confused if they don't have the same dataset behavior depending on the disk type.\r\n\r\nIf they want to use a remote bucket they should use the remote URI instead, e.g. `s3://...`. Advancements on this are tracked at #5281 " ]

[]

[]

[]

[]

[]

[]

[]

[ "Does it mean that huge files might end up being downloaded? It would go against the purpose of streaming, I think. At least, this fallback should be an option that could be disabled", "Yes, it is against the purpose of streaming, but streaming is not possible if the server does not allow HTTP range requests.\n\nWe have two options: either we download the file or we throw an error.", "I think we simply cannot fallback to downloading the file if streaming fails without the user being aware of it. Some options: \r\n- make the fallback optional (using an env var? or a function param)\r\n- use the fallback only if the dataset size is under some threshold (provided we have the data in the DatasetInfo) -> it's the option I use in `datasets-preview-backend` ([here](https://github.com/huggingface/datasets-preview-backend/blob/48ac19e49c19809763e8d640986bf2c3d792faed/src/datasets_preview_backend/models/typed_row.py#L40) and [here](https://github.com/huggingface/datasets-preview-backend/blob/aa86c5493b275c9e2dbae7dab7bd469da5773a41/src/datasets_preview_backend/models/split.py#L31-L37))\r\n- throw an exception and let the user decide what to do\r\n", "IMO in general we should throw an exception and ask the user to not use streaming mode in that case.\r\n\r\nYour second point is also interesting but I feel like it could be confusing for users sometimes: it doesn't feel natural that the streaming-ability should depend on the size of the file.", "Sure, I think we should just throw an exception\r\n", "Current behavior is already throwing an Exception:\r\n```\r\nValueError: Cannot seek streaming HTTP file\r\n```\r\n\r\nWe could customize the exception class and/or the exception message.", "I'm not sure we really need to change anything. I opened the issue https://github.com/huggingface/datasets/issues/3677 because discovery was streamable and is not anymore (according to my test suite in https://github.com/huggingface/datasets-preview-backend): I was not sure if it was due to some regression in the library, or to some change in the dataset itself.", "I'm wondering why it worked before and it is no longer working...", "> We could customize the exception class and/or the exception message.\r\n\r\nYup a message that says that the host doesn't support streaming because it doesn't support HTTP Range requests would be useful !", "DONE, @lhoestq. " ]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_3923). All of your documentation changes will be reflected on that endpoint.", "Is this feature stale or needs any help to it ? If so I can quickly send a PR. Thanks\r\n\r\nCC : @lhoestq, @albertvillanova ", "These features have been merged and are already available, thanks :)", "Hello @lhoestq, I see that `IterableDataset` doesn't allow features like `take`, `len`, `slice` which can enable a lot of stuffs. Is it worth an addition ? Or is it intended that they didn't have those features ?", "IterableDataset objects don't have `len` or `slice` because they can be possibly unbounded (you don't know in advance how many items they contain). Though IterableDataset.take and IterableDataset.skip do exist." ]

[]

[ "Looks like an issue with your csv file. Did you use the right delimiter ?\r\nApparently at line 37 the CSV reader from pandas reads 2 fields instead of 1.", "Note that you can pass any argument you would pass to `pandas.read_csv` as kwargs to `load_dataset`. For example you can do\r\n```python\r\nfrom datasets import load_dataset\r\ndataset = load_dataset('csv', data_files=data_files, sep=\"\\t\")\r\n```\r\n\r\nfor example to use a tab separator.\r\n\r\nYou can see the full list of arguments here: https://github.com/huggingface/datasets/blob/master/src/datasets/packaged_modules/csv/csv.py\r\n\r\n(I've not found the list in the documentation though, we definitely must add them !)", "You can try to convert the file to (CSV UTF-8)" ]

[]

[ "Hi ! This might be related to #2153 \r\n\r\nYou're right the ArrowWriter should be initialized with `features=self.info.features` ! Good catch\r\nI'm opening a PR to fix this and also to figure out how it was not caught in the tests\r\n\r\nEDIT: opened #2201", "> Hi ! This might be related to #2153\r\n> \r\n> You're right the ArrowWriter should be initialized with `features=self.info.features` ! Good catch\r\n> I'm opening a PR to fix this and also to figure out how it was not caught in the tests\r\n> \r\n> EDIT: opened #2201\r\n\r\nGlad to hear that! Thank you for your fix, I'm new to huggingface, it's a fantastic project 😁" ]

[]

[ "The tests for the Wikipedia dataset do not pass anymore with the error:\r\n```\r\nTo be able to use this dataset, you need to install the following dependencies ['mwparserfromhell'] using 'pip install mwparserfromhell' for instance'\r\n```", "This was an issue on master. You can just rebase from master.", "Perfect! Indeed, it worked^^ Thanks @lhoestq ", "Currently the dummy_data tests are always green because in a PR the dataset is not yet synchronized with aws. This PR fixes this: https://github.com/huggingface/nlp/pull/140 . \r\n\r\nCould you test `json` locally or wait until the PR: https://github.com/huggingface/nlp/pull/140 is merged ? :-) ", "Ok, I will wait #140 to be merged and then rebase :) " ]

[]

[ "I think this is the way to go but I’m afraid this might be a little slow. I was thinking that we could use a high quality very fast non crypto hash like xxhash for these stuff (hashing data files)", "Yep good idea, I'll take a look", "I tested the hashing speed [here](https://colab.research.google.com/drive/1hlhP84kLIHmOzMRQN1h8x10hKWpXXyud?usp=sharing).\r\nI was able to get 8x speed with `xxhashlib` (42ms vs 345ms for 100MiB of data).\r\nWhat do you think @thomwolf ?", "I added xxhash and some tests" ]

[ { "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! I was looking into this and hope you can clarify a point. Your my_dataset variable would be of type DatasetDict which means the alternative you've described (dict comprehension) is what makes sense. \r\nIs there a reason why you wouldn't want to convert my_dataset to a pandas df if you'd like to use it like one? Please let me know if I'm missing something.", "Hi! Sorry for the delay.\r\n\r\nIn this case, the dataset would be a `datasets.Dataset` and we want to select multiple columns, the `idx` and `label` columns for example.\r\n\r\nMy issue is that my dataset is too big for memory if I load everything into pandas." ]

[ { "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" } ]

[ "Some of the shard files now have 0 number of rows.\r\n\r\nWe have opened an issue in the Apache Beam repo:\r\n- https://github.com/apache/beam/issues/25041" ]

[]

[]

[]

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

[]

[ "Hi @lhoestq @yjernite, requesting you to review this for any changes needed. Thanks! :)", "Hi @lhoestq , I have updated the PR" ]

[]

[ "When should we expect to see our dataset appear in the search dropdown at huggingface.co?", "Hi @eusip,\r\n\r\n> When should we expect to see our dataset appear in the search dropdown at huggingface.co?\r\n\r\nwhen this PR is merged.", "Thanks!", "I've implemented all the changes requested by @lhoestq but I made the mistake of trying to change the remote branch name. \r\n\r\nHopefully the changes are seen on your end as both branches `silicone` and `main` should be up-to-date.", "It looks like the PR includes changes about many other files than the ones for Silicone (+30,000 line changes)\r\n\r\nMaybe you can try to create another branch and another PR ?", "> It looks like the PR includes changes about many other files than the ones for Silicone (+30,000 line changes)\r\n> \r\n> Maybe you can try to create another branch and another PR ?\r\n\r\nSure. I will make a new pull request." ]

[]

[]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "What do you think @albertvillanova ?" ]

[]

[ "Hi @iliemihai, it looks like this PR holds changes from your previous PR #1493 .\r\nWould you mind removing them from the branch please ?", "@SBrandeis I am sorry. Yes I will remove them. Thank you :D ", "Hi @lhoestq @SBrandeis @iliemihai\r\n\r\nIs this still in progress or can I take over this one?\r\n\r\nThanks,\r\nGunjan", "Hi,\r\nWhile trying to add this dataset, I found some potential issues. \r\nThe homepage mentioned is : https://github.com/katakonst/sentiment-analysis-tensorflow/tree/master/datasets/ro/, where the dataset is different from the URLs: https://raw.githubusercontent.com/dumitrescustefan/Romanian-Transformers/examples/examples/sentiment_analysis/ro/train.csv. It is unclear which dataset is \"correct\". I checked the total examples (train+test) in both places and they do not match.", "We should use the data from dumitrescustefan and set the homepage to his repo IMO, since he's first author of the paper of the dataset.", "Hi @lhoestq,\r\n\r\nCool, I'll get working on it.\r\n\r\nThanks", "Hi @lhoestq, \r\n\r\nThis PR can be closed.", "Closing in favor of #2011 \r\nThanks again for adding it !" ]

[]

[ "merging since the CI just had a connection error" ]

[]

[ "_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) | 0.006543 / 0.011353 (-0.004809) | 0.004085 / 0.011008 (-0.006924) | 0.083989 / 0.038508 (0.045481) | 0.074733 / 0.023109 (0.051623) | 0.310839 / 0.275898 (0.034941) | 0.333540 / 0.323480 (0.010060) | 0.005566 / 0.007986 (-0.002419) | 0.003461 / 0.004328 (-0.000868) | 0.065194 / 0.004250 (0.060943) | 0.057007 / 0.037052 (0.019954) | 0.325633 / 0.258489 (0.067144) | 0.351665 / 0.293841 (0.057824) | 0.030561 / 0.128546 (-0.097985) | 0.008579 / 0.075646 (-0.067068) | 0.287457 / 0.419271 (-0.131815) | 0.063554 / 0.043533 (0.020021) | 0.309182 / 0.255139 (0.054043) | 0.327809 / 0.283200 (0.044609) | 0.034470 / 0.141683 (-0.107213) | 1.452098 / 1.452155 (-0.000057) | 1.527130 / 1.492716 (0.034414) |\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) | 0.241736 / 0.018006 (0.223729) | 0.552432 / 0.000490 (0.551943) | 0.004085 / 0.000200 (0.003885) | 0.000089 / 0.000054 (0.000035) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.027290 / 0.037411 (-0.010121) | 0.081250 / 0.014526 (0.066724) | 0.094739 / 0.176557 (-0.081818) | 0.150424 / 0.737135 (-0.586711) | 0.095488 / 0.296338 (-0.200851) |\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) | 0.377245 / 0.215209 (0.162036) | 3.781021 / 2.077655 (1.703366) | 1.820092 / 1.504120 (0.315972) | 1.654420 / 1.541195 (0.113225) | 1.751256 / 1.468490 (0.282766) | 0.475161 / 4.584777 (-4.109616) | 3.603462 / 3.745712 (-0.142251) | 5.437837 / 5.269862 (0.167975) | 3.305598 / 4.565676 (-1.260079) | 0.055856 / 0.424275 (-0.368419) | 0.007259 / 0.007607 (-0.000348) | 0.454205 / 0.226044 (0.228161) | 4.544157 / 2.268929 (2.275229) | 2.296776 / 55.444624 (-53.147848) | 1.951017 / 6.876477 (-4.925459) | 2.128759 / 2.142072 (-0.013313) | 0.590354 / 4.805227 (-4.214873) | 0.129974 / 6.500664 (-6.370690) | 0.059506 / 0.075469 (-0.015963) |\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) | 1.285866 / 1.841788 (-0.555921) | 19.419446 / 8.074308 (11.345138) | 13.985108 / 10.191392 (3.793716) | 0.146803 / 0.680424 (-0.533620) | 0.018176 / 0.534201 (-0.516025) | 0.392345 / 0.579283 (-0.186938) | 0.405394 / 0.434364 (-0.028970) | 0.454649 / 0.540337 (-0.085688) | 0.633075 / 1.386936 (-0.753861) |\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) | 0.006497 / 0.011353 (-0.004855) | 0.004092 / 0.011008 (-0.006916) | 0.064908 / 0.038508 (0.026400) | 0.073494 / 0.023109 (0.050385) | 0.382227 / 0.275898 (0.106329) | 0.407320 / 0.323480 (0.083840) | 0.005653 / 0.007986 (-0.002332) | 0.003500 / 0.004328 (-0.000829) | 0.064570 / 0.004250 (0.060320) | 0.058733 / 0.037052 (0.021681) | 0.385702 / 0.258489 (0.127213) | 0.426463 / 0.293841 (0.132622) | 0.031073 / 0.128546 (-0.097473) | 0.008710 / 0.075646 (-0.066936) | 0.071378 / 0.419271 (-0.347893) | 0.050141 / 0.043533 (0.006608) | 0.377769 / 0.255139 (0.122630) | 0.395016 / 0.283200 (0.111816) | 0.025158 / 0.141683 (-0.116525) | 1.470503 / 1.452155 (0.018348) | 1.532742 / 1.492716 (0.040026) |\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) | 0.214249 / 0.018006 (0.196243) | 0.583580 / 0.000490 (0.583090) | 0.004027 / 0.000200 (0.003828) | 0.000104 / 0.000054 (0.000050) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.030186 / 0.037411 (-0.007226) | 0.086927 / 0.014526 (0.072401) | 0.102060 / 0.176557 (-0.074497) | 0.156281 / 0.737135 (-0.580855) | 0.100825 / 0.296338 (-0.195514) |\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) | 0.419942 / 0.215209 (0.204733) | 4.183797 / 2.077655 (2.106142) | 2.205079 / 1.504120 (0.700959) | 2.071219 / 1.541195 (0.530024) | 2.194047 / 1.468490 (0.725557) | 0.478768 / 4.584777 (-4.106009) | 3.584864 / 3.745712 (-0.160848) | 3.371635 / 5.269862 (-1.898227) | 2.022134 / 4.565676 (-2.543542) | 0.056553 / 0.424275 (-0.367722) | 0.007231 / 0.007607 (-0.000376) | 0.493158 / 0.226044 (0.267113) | 4.934370 / 2.268929 (2.665441) | 2.699593 / 55.444624 (-52.745031) | 2.396371 / 6.876477 (-4.480105) | 2.438052 / 2.142072 (0.295979) | 0.589578 / 4.805227 (-4.215649) | 0.147234 / 6.500664 (-6.353430) | 0.062049 / 0.075469 (-0.013420) |\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) | 1.318246 / 1.841788 (-0.523542) | 19.829025 / 8.074308 (11.754717) | 14.314825 / 10.191392 (4.123433) | 0.168309 / 0.680424 (-0.512115) | 0.018596 / 0.534201 (-0.515605) | 0.397540 / 0.579283 (-0.181743) | 0.421280 / 0.434364 (-0.013084) | 0.479917 / 0.540337 (-0.060421) | 0.643494 / 1.386936 (-0.743442) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#5be59becaa65f1fa08129091b8c778823e4a50ac \"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) | 0.008349 / 0.011353 (-0.003004) | 0.005362 / 0.011008 (-0.005646) | 0.100777 / 0.038508 (0.062269) | 0.078719 / 0.023109 (0.055609) | 0.398105 / 0.275898 (0.122207) | 0.444189 / 0.323480 (0.120709) | 0.006834 / 0.007986 (-0.001152) | 0.004642 / 0.004328 (0.000314) | 0.076284 / 0.004250 (0.072034) | 0.062738 / 0.037052 (0.025685) | 0.409532 / 0.258489 (0.151043) | 0.447218 / 0.293841 (0.153377) | 0.052996 / 0.128546 (-0.075550) | 0.012977 / 0.075646 (-0.062669) | 0.347687 / 0.419271 (-0.071585) | 0.068076 / 0.043533 (0.024543) | 0.394526 / 0.255139 (0.139387) | 0.434110 / 0.283200 (0.150910) | 0.041719 / 0.141683 (-0.099963) | 1.759109 / 1.452155 (0.306955) | 1.866049 / 1.492716 (0.373333) |\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) | 0.287633 / 0.018006 (0.269627) | 0.611540 / 0.000490 (0.611051) | 0.005388 / 0.000200 (0.005188) | 0.000096 / 0.000054 (0.000042) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.027394 / 0.037411 (-0.010017) | 0.089796 / 0.014526 (0.075270) | 0.106931 / 0.176557 (-0.069625) | 0.173560 / 0.737135 (-0.563575) | 0.106948 / 0.296338 (-0.189391) |\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) | 0.575156 / 0.215209 (0.359947) | 5.674170 / 2.077655 (3.596516) | 2.463090 / 1.504120 (0.958971) | 2.128245 / 1.541195 (0.587050) | 2.118982 / 1.468490 (0.650492) | 0.876976 / 4.584777 (-3.707801) | 5.238229 / 3.745712 (1.492517) | 4.548788 / 5.269862 (-0.721074) | 2.905243 / 4.565676 (-1.660433) | 0.090750 / 0.424275 (-0.333525) | 0.008266 / 0.007607 (0.000659) | 0.693305 / 0.226044 (0.467260) | 7.126970 / 2.268929 (4.858041) | 3.152131 / 55.444624 (-52.292494) | 2.532118 / 6.876477 (-4.344359) | 2.678442 / 2.142072 (0.536369) | 0.932745 / 4.805227 (-3.872483) | 0.196290 / 6.500664 (-6.304374) | 0.074082 / 0.075469 (-0.001387) |\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) | 1.599636 / 1.841788 (-0.242152) | 23.271435 / 8.074308 (15.197127) | 19.696709 / 10.191392 (9.505317) | 0.222668 / 0.680424 (-0.457756) | 0.029088 / 0.534201 (-0.505113) | 0.492477 / 0.579283 (-0.086806) | 0.580578 / 0.434364 (0.146214) | 0.558852 / 0.540337 (0.018514) | 0.762083 / 1.386936 (-0.624853) |\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) | 0.009021 / 0.011353 (-0.002332) | 0.005011 / 0.011008 (-0.005997) | 0.076504 / 0.038508 (0.037996) | 0.077303 / 0.023109 (0.054193) | 0.480660 / 0.275898 (0.204762) | 0.493944 / 0.323480 (0.170464) | 0.006339 / 0.007986 (-0.001646) | 0.004302 / 0.004328 (-0.000026) | 0.076228 / 0.004250 (0.071978) | 0.060805 / 0.037052 (0.023753) | 0.477539 / 0.258489 (0.219050) | 0.496799 / 0.293841 (0.202958) | 0.049495 / 0.128546 (-0.079052) | 0.013333 / 0.075646 (-0.062313) | 0.087217 / 0.419271 (-0.332055) | 0.061451 / 0.043533 (0.017918) | 0.485169 / 0.255139 (0.230030) | 0.487348 / 0.283200 (0.204149) | 0.035874 / 0.141683 (-0.105809) | 1.829137 / 1.452155 (0.376982) | 1.906151 / 1.492716 (0.413435) |\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) | 0.304526 / 0.018006 (0.286520) | 0.627499 / 0.000490 (0.627009) | 0.003786 / 0.000200 (0.003586) | 0.000098 / 0.000054 (0.000043) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.035512 / 0.037411 (-0.001899) | 0.096684 / 0.014526 (0.082158) | 0.111879 / 0.176557 (-0.064678) | 0.171489 / 0.737135 (-0.565647) | 0.112175 / 0.296338 (-0.184164) |\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) | 0.604791 / 0.215209 (0.389582) | 6.089137 / 2.077655 (4.011482) | 2.883237 / 1.504120 (1.379117) | 2.561109 / 1.541195 (1.019914) | 2.542400 / 1.468490 (1.073910) | 0.852828 / 4.584777 (-3.731949) | 5.236812 / 3.745712 (1.491100) | 4.756429 / 5.269862 (-0.513432) | 2.885660 / 4.565676 (-1.680016) | 0.095643 / 0.424275 (-0.328632) | 0.008403 / 0.007607 (0.000796) | 0.727707 / 0.226044 (0.501663) | 7.428002 / 2.268929 (5.159074) | 3.816051 / 55.444624 (-51.628573) | 2.971057 / 6.876477 (-3.905420) | 2.915965 / 2.142072 (0.773893) | 1.006553 / 4.805227 (-3.798674) | 0.201840 / 6.500664 (-6.298824) | 0.080795 / 0.075469 (0.005326) |\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) | 1.794951 / 1.841788 (-0.046837) | 23.624556 / 8.074308 (15.550248) | 21.856195 / 10.191392 (11.664802) | 0.253043 / 0.680424 (-0.427381) | 0.031201 / 0.534201 (-0.503000) | 0.461641 / 0.579283 (-0.117642) | 0.577789 / 0.434364 (0.143425) | 0.569197 / 0.540337 (0.028860) | 0.780111 / 1.386936 (-0.606825) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#4904f14459c862f0ab525ec034a636177be5dee4 \"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) | 0.007646 / 0.011353 (-0.003707) | 0.004750 / 0.011008 (-0.006258) | 0.097981 / 0.038508 (0.059473) | 0.088989 / 0.023109 (0.065880) | 0.377732 / 0.275898 (0.101834) | 0.406805 / 0.323480 (0.083325) | 0.006389 / 0.007986 (-0.001597) | 0.003854 / 0.004328 (-0.000474) | 0.073977 / 0.004250 (0.069727) | 0.066497 / 0.037052 (0.029444) | 0.371498 / 0.258489 (0.113009) | 0.417352 / 0.293841 (0.123511) | 0.036326 / 0.128546 (-0.092220) | 0.009876 / 0.075646 (-0.065770) | 0.330142 / 0.419271 (-0.089130) | 0.062423 / 0.043533 (0.018890) | 0.369375 / 0.255139 (0.114236) | 0.406048 / 0.283200 (0.122848) | 0.026564 / 0.141683 (-0.115119) | 1.713295 / 1.452155 (0.261140) | 1.797493 / 1.492716 (0.304777) |\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) | 0.231889 / 0.018006 (0.213882) | 0.512497 / 0.000490 (0.512007) | 0.000390 / 0.000200 (0.000190) | 0.000069 / 0.000054 (0.000015) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.033978 / 0.037411 (-0.003433) | 0.100117 / 0.014526 (0.085592) | 0.112460 / 0.176557 (-0.064097) | 0.179936 / 0.737135 (-0.557200) | 0.114277 / 0.296338 (-0.182061) |\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) | 0.461320 / 0.215209 (0.246111) | 4.563180 / 2.077655 (2.485526) | 2.249474 / 1.504120 (0.745354) | 2.100450 / 1.541195 (0.559255) | 2.231080 / 1.468490 (0.762590) | 0.567907 / 4.584777 (-4.016870) | 4.117233 / 3.745712 (0.371521) | 4.943159 / 5.269862 (-0.326703) | 3.112299 / 4.565676 (-1.453377) | 0.065500 / 0.424275 (-0.358775) | 0.008407 / 0.007607 (0.000800) | 0.545928 / 0.226044 (0.319883) | 5.508058 / 2.268929 (3.239129) | 2.834645 / 55.444624 (-52.609980) | 2.440328 / 6.876477 (-4.436148) | 2.680483 / 2.142072 (0.538410) | 0.697191 / 4.805227 (-4.108036) | 0.176646 / 6.500664 (-6.324018) | 0.073608 / 0.075469 (-0.001861) |\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) | 1.451865 / 1.841788 (-0.389922) | 22.752595 / 8.074308 (14.678287) | 15.543338 / 10.191392 (5.351946) | 0.214644 / 0.680424 (-0.465780) | 0.022050 / 0.534201 (-0.512151) | 0.463898 / 0.579283 (-0.115385) | 0.481691 / 0.434364 (0.047327) | 0.549715 / 0.540337 (0.009378) | 0.773595 / 1.386936 (-0.613341) |\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) | 0.007541 / 0.011353 (-0.003812) | 0.004715 / 0.011008 (-0.006293) | 0.076782 / 0.038508 (0.038274) | 0.086242 / 0.023109 (0.063133) | 0.458053 / 0.275898 (0.182155) | 0.503097 / 0.323480 (0.179617) | 0.006262 / 0.007986 (-0.001724) | 0.003882 / 0.004328 (-0.000447) | 0.075669 / 0.004250 (0.071419) | 0.066004 / 0.037052 (0.028952) | 0.469439 / 0.258489 (0.210950) | 0.529744 / 0.293841 (0.235903) | 0.037228 / 0.128546 (-0.091319) | 0.009794 / 0.075646 (-0.065852) | 0.082464 / 0.419271 (-0.336808) | 0.058797 / 0.043533 (0.015264) | 0.452069 / 0.255139 (0.196930) | 0.488246 / 0.283200 (0.205046) | 0.029324 / 0.141683 (-0.112359) | 1.742237 / 1.452155 (0.290082) | 1.839676 / 1.492716 (0.346959) |\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) | 0.228106 / 0.018006 (0.210100) | 0.491632 / 0.000490 (0.491142) | 0.004993 / 0.000200 (0.004793) | 0.000114 / 0.000054 (0.000060) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.035413 / 0.037411 (-0.001999) | 0.104617 / 0.014526 (0.090091) | 0.121948 / 0.176557 (-0.054609) | 0.186233 / 0.737135 (-0.550902) | 0.121574 / 0.296338 (-0.174764) |\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) | 0.473849 / 0.215209 (0.258640) | 4.788312 / 2.077655 (2.710657) | 2.470535 / 1.504120 (0.966415) | 2.270393 / 1.541195 (0.729198) | 2.361096 / 1.468490 (0.892606) | 0.556184 / 4.584777 (-4.028593) | 4.216852 / 3.745712 (0.471140) | 3.901718 / 5.269862 (-1.368143) | 2.355209 / 4.565676 (-2.210467) | 0.066708 / 0.424275 (-0.357567) | 0.008709 / 0.007607 (0.001102) | 0.571714 / 0.226044 (0.345669) | 5.663150 / 2.268929 (3.394221) | 3.025769 / 55.444624 (-52.418855) | 2.652554 / 6.876477 (-4.223923) | 2.750555 / 2.142072 (0.608483) | 0.681536 / 4.805227 (-4.123691) | 0.157187 / 6.500664 (-6.343477) | 0.073533 / 0.075469 (-0.001936) |\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) | 1.604630 / 1.841788 (-0.237158) | 22.735629 / 8.074308 (14.661321) | 16.762347 / 10.191392 (6.570955) | 0.175514 / 0.680424 (-0.504910) | 0.021497 / 0.534201 (-0.512704) | 0.461438 / 0.579283 (-0.117845) | 0.476184 / 0.434364 (0.041820) | 0.571048 / 0.540337 (0.030710) | 0.747086 / 1.386936 (-0.639850) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#6ea38fc40ee2b10d3b5c6df09b09ad05e02a2cff \"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) | 0.006889 / 0.011353 (-0.004464) | 0.004241 / 0.011008 (-0.006767) | 0.084542 / 0.038508 (0.046034) | 0.080484 / 0.023109 (0.057374) | 0.309356 / 0.275898 (0.033458) | 0.338548 / 0.323480 (0.015068) | 0.004904 / 0.007986 (-0.003082) | 0.005220 / 0.004328 (0.000892) | 0.065501 / 0.004250 (0.061251) | 0.062095 / 0.037052 (0.025043) | 0.317332 / 0.258489 (0.058843) | 0.364797 / 0.293841 (0.070956) | 0.030492 / 0.128546 (-0.098054) | 0.008991 / 0.075646 (-0.066656) | 0.288274 / 0.419271 (-0.130998) | 0.052582 / 0.043533 (0.009049) | 0.310838 / 0.255139 (0.055699) | 0.346304 / 0.283200 (0.063104) | 0.027968 / 0.141683 (-0.113715) | 1.509727 / 1.452155 (0.057573) | 1.577410 / 1.492716 (0.084694) |\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) | 0.269725 / 0.018006 (0.251719) | 0.627685 / 0.000490 (0.627195) | 0.000419 / 0.000200 (0.000219) | 0.000060 / 0.000054 (0.000006) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.031022 / 0.037411 (-0.006389) | 0.081858 / 0.014526 (0.067332) | 0.099477 / 0.176557 (-0.077080) | 0.162981 / 0.737135 (-0.574154) | 0.101987 / 0.296338 (-0.194351) |\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) | 0.386297 / 0.215209 (0.171088) | 3.845321 / 2.077655 (1.767666) | 1.834446 / 1.504120 (0.330326) | 1.699730 / 1.541195 (0.158536) | 1.764342 / 1.468490 (0.295852) | 0.486423 / 4.584777 (-4.098354) | 3.527595 / 3.745712 (-0.218117) | 4.137034 / 5.269862 (-1.132827) | 2.590457 / 4.565676 (-1.975219) | 0.057598 / 0.424275 (-0.366677) | 0.007318 / 0.007607 (-0.000289) | 0.460775 / 0.226044 (0.234730) | 4.627576 / 2.268929 (2.358647) | 2.402566 / 55.444624 (-53.042059) | 2.011392 / 6.876477 (-4.865085) | 2.223915 / 2.142072 (0.081842) | 0.623217 / 4.805227 (-4.182011) | 0.148875 / 6.500664 (-6.351789) | 0.059799 / 0.075469 (-0.015671) |\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) | 1.290768 / 1.841788 (-0.551020) | 20.455083 / 8.074308 (12.380775) | 13.469846 / 10.191392 (3.278454) | 0.170329 / 0.680424 (-0.510095) | 0.018409 / 0.534201 (-0.515792) | 0.394356 / 0.579283 (-0.184927) | 0.422685 / 0.434364 (-0.011679) | 0.476241 / 0.540337 (-0.064096) | 0.662682 / 1.386936 (-0.724254) |\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) | 0.006724 / 0.011353 (-0.004629) | 0.004508 / 0.011008 (-0.006500) | 0.065304 / 0.038508 (0.026796) | 0.080243 / 0.023109 (0.057133) | 0.384545 / 0.275898 (0.108647) | 0.415234 / 0.323480 (0.091754) | 0.006361 / 0.007986 (-0.001624) | 0.004193 / 0.004328 (-0.000135) | 0.065940 / 0.004250 (0.061689) | 0.063633 / 0.037052 (0.026581) | 0.392799 / 0.258489 (0.134310) | 0.443618 / 0.293841 (0.149777) | 0.031134 / 0.128546 (-0.097412) | 0.009058 / 0.075646 (-0.066588) | 0.071051 / 0.419271 (-0.348221) | 0.049096 / 0.043533 (0.005563) | 0.379526 / 0.255139 (0.124387) | 0.403370 / 0.283200 (0.120171) | 0.026378 / 0.141683 (-0.115305) | 1.457879 / 1.452155 (0.005724) | 1.562890 / 1.492716 (0.070174) |\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) | 0.304416 / 0.018006 (0.286410) | 0.626046 / 0.000490 (0.625557) | 0.000469 / 0.000200 (0.000269) | 0.000057 / 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) | 0.032979 / 0.037411 (-0.004433) | 0.086769 / 0.014526 (0.072243) | 0.108188 / 0.176557 (-0.068369) | 0.163077 / 0.737135 (-0.574058) | 0.106276 / 0.296338 (-0.190062) |\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) | 0.406922 / 0.215209 (0.191713) | 4.052828 / 2.077655 (1.975174) | 2.084802 / 1.504120 (0.580682) | 1.927263 / 1.541195 (0.386069) | 1.956078 / 1.468490 (0.487587) | 0.480110 / 4.584777 (-4.104667) | 3.553022 / 3.745712 (-0.192691) | 3.554450 / 5.269862 (-1.715411) | 2.082681 / 4.565676 (-2.482995) | 0.056711 / 0.424275 (-0.367564) | 0.007374 / 0.007607 (-0.000234) | 0.480555 / 0.226044 (0.254510) | 4.795851 / 2.268929 (2.526923) | 2.606675 / 55.444624 (-52.837949) | 2.249964 / 6.876477 (-4.626512) | 2.274234 / 2.142072 (0.132162) | 0.571767 / 4.805227 (-4.233461) | 0.133312 / 6.500664 (-6.367352) | 0.061703 / 0.075469 (-0.013766) |\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) | 1.354308 / 1.841788 (-0.487479) | 20.959352 / 8.074308 (12.885044) | 14.158420 / 10.191392 (3.967028) | 0.197959 / 0.680424 (-0.482465) | 0.018412 / 0.534201 (-0.515789) | 0.394307 / 0.579283 (-0.184976) | 0.402455 / 0.434364 (-0.031909) | 0.463314 / 0.540337 (-0.077024) | 0.621050 / 1.386936 (-0.765886) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#d7298d4d1b169442a8d0bc8c1667298bb89ca501 \"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) | 0.007179 / 0.011353 (-0.004174) | 0.004318 / 0.011008 (-0.006690) | 0.085209 / 0.038508 (0.046701) | 0.089989 / 0.023109 (0.066880) | 0.328188 / 0.275898 (0.052290) | 0.346027 / 0.323480 (0.022547) | 0.005711 / 0.007986 (-0.002275) | 0.003703 / 0.004328 (-0.000625) | 0.065419 / 0.004250 (0.061169) | 0.065354 / 0.037052 (0.028301) | 0.314531 / 0.258489 (0.056042) | 0.354357 / 0.293841 (0.060516) | 0.030918 / 0.128546 (-0.097628) | 0.008632 / 0.075646 (-0.067015) | 0.286817 / 0.419271 (-0.132455) | 0.065267 / 0.043533 (0.021735) | 0.310918 / 0.255139 (0.055779) | 0.330497 / 0.283200 (0.047298) | 0.035695 / 0.141683 (-0.105988) | 1.471101 / 1.452155 (0.018947) | 1.538658 / 1.492716 (0.045942) |\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) | 0.254314 / 0.018006 (0.236308) | 0.591413 / 0.000490 (0.590923) | 0.006082 / 0.000200 (0.005882) | 0.000091 / 0.000054 (0.000037) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.031843 / 0.037411 (-0.005568) | 0.089968 / 0.014526 (0.075442) | 0.101838 / 0.176557 (-0.074718) | 0.164401 / 0.737135 (-0.572734) | 0.103785 / 0.296338 (-0.192554) |\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) | 0.380486 / 0.215209 (0.165277) | 3.798868 / 2.077655 (1.721213) | 1.824645 / 1.504120 (0.320525) | 1.660804 / 1.541195 (0.119610) | 1.784793 / 1.468490 (0.316303) | 0.487222 / 4.584777 (-4.097555) | 3.560580 / 3.745712 (-0.185132) | 5.392662 / 5.269862 (0.122800) | 3.295327 / 4.565676 (-1.270350) | 0.057699 / 0.424275 (-0.366576) | 0.007559 / 0.007607 (-0.000048) | 0.459655 / 0.226044 (0.233611) | 4.587583 / 2.268929 (2.318654) | 2.304845 / 55.444624 (-53.139779) | 1.966433 / 6.876477 (-4.910044) | 2.254591 / 2.142072 (0.112519) | 0.582978 / 4.805227 (-4.222250) | 0.133455 / 6.500664 (-6.367210) | 0.061924 / 0.075469 (-0.013546) |\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) | 1.275685 / 1.841788 (-0.566103) | 20.814545 / 8.074308 (12.740237) | 13.753567 / 10.191392 (3.562175) | 0.164076 / 0.680424 (-0.516348) | 0.018768 / 0.534201 (-0.515433) | 0.390991 / 0.579283 (-0.188293) | 0.404417 / 0.434364 (-0.029947) | 0.457522 / 0.540337 (-0.082815) | 0.624654 / 1.386936 (-0.762282) |\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) | 0.007435 / 0.011353 (-0.003918) | 0.004255 / 0.011008 (-0.006754) | 0.066134 / 0.038508 (0.027626) | 0.086035 / 0.023109 (0.062925) | 0.364688 / 0.275898 (0.088790) | 0.403895 / 0.323480 (0.080415) | 0.005868 / 0.007986 (-0.002117) | 0.003634 / 0.004328 (-0.000694) | 0.065803 / 0.004250 (0.061553) | 0.065113 / 0.037052 (0.028061) | 0.370057 / 0.258489 (0.111568) | 0.412634 / 0.293841 (0.118793) | 0.031660 / 0.128546 (-0.096886) | 0.008699 / 0.075646 (-0.066947) | 0.070618 / 0.419271 (-0.348654) | 0.050814 / 0.043533 (0.007281) | 0.362320 / 0.255139 (0.107181) | 0.383863 / 0.283200 (0.100663) | 0.027980 / 0.141683 (-0.113703) | 1.486389 / 1.452155 (0.034234) | 1.595534 / 1.492716 (0.102817) |\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) | 0.300991 / 0.018006 (0.282985) | 0.565265 / 0.000490 (0.564775) | 0.000400 / 0.000200 (0.000200) | 0.000053 / 0.000054 (-0.000001) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.034942 / 0.037411 (-0.002470) | 0.092498 / 0.014526 (0.077972) | 0.106737 / 0.176557 (-0.069819) | 0.165400 / 0.737135 (-0.571735) | 0.107809 / 0.296338 (-0.188529) |\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) | 0.412156 / 0.215209 (0.196947) | 4.116747 / 2.077655 (2.039092) | 2.199612 / 1.504120 (0.695492) | 2.049310 / 1.541195 (0.508115) | 2.174342 / 1.468490 (0.705852) | 0.482794 / 4.584777 (-4.101983) | 3.561344 / 3.745712 (-0.184368) | 3.465935 / 5.269862 (-1.803926) | 2.076595 / 4.565676 (-2.489081) | 0.056242 / 0.424275 (-0.368033) | 0.007371 / 0.007607 (-0.000236) | 0.489135 / 0.226044 (0.263091) | 4.895691 / 2.268929 (2.626763) | 2.626936 / 55.444624 (-52.817688) | 2.306658 / 6.876477 (-4.569818) | 2.421705 / 2.142072 (0.279633) | 0.599547 / 4.805227 (-4.205680) | 0.133627 / 6.500664 (-6.367037) | 0.063830 / 0.075469 (-0.011639) |\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) | 1.383039 / 1.841788 (-0.458748) | 21.005346 / 8.074308 (12.931038) | 14.911083 / 10.191392 (4.719691) | 0.190995 / 0.680424 (-0.489429) | 0.018510 / 0.534201 (-0.515691) | 0.396346 / 0.579283 (-0.182937) | 0.411496 / 0.434364 (-0.022868) | 0.470972 / 0.540337 (-0.069366) | 0.615670 / 1.386936 (-0.771266) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#d6d2ba47759d8acbf3d750b1cc4d89b195b1f9c9 \"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) | 0.007249 / 0.011353 (-0.004104) | 0.004261 / 0.011008 (-0.006747) | 0.100645 / 0.038508 (0.062137) | 0.078522 / 0.023109 (0.055413) | 0.423526 / 0.275898 (0.147628) | 0.439541 / 0.323480 (0.116061) | 0.005812 / 0.007986 (-0.002173) | 0.003615 / 0.004328 (-0.000713) | 0.075908 / 0.004250 (0.071658) | 0.062490 / 0.037052 (0.025437) | 0.414941 / 0.258489 (0.156452) | 0.447267 / 0.293841 (0.153426) | 0.035127 / 0.128546 (-0.093419) | 0.009642 / 0.075646 (-0.066004) | 0.354093 / 0.419271 (-0.065179) | 0.060970 / 0.043533 (0.017437) | 0.418579 / 0.255139 (0.163440) | 0.427972 / 0.283200 (0.144772) | 0.025838 / 0.141683 (-0.115845) | 1.778349 / 1.452155 (0.326194) | 1.845965 / 1.492716 (0.353249) |\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) | 0.227304 / 0.018006 (0.209298) | 0.571833 / 0.000490 (0.571343) | 0.001328 / 0.000200 (0.001128) | 0.000071 / 0.000054 (0.000017) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.031343 / 0.037411 (-0.006068) | 0.096400 / 0.014526 (0.081875) | 0.106881 / 0.176557 (-0.069676) | 0.175449 / 0.737135 (-0.561686) | 0.108751 / 0.296338 (-0.187588) |\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) | 0.480204 / 0.215209 (0.264995) | 4.622063 / 2.077655 (2.544408) | 2.211505 / 1.504120 (0.707385) | 2.065154 / 1.541195 (0.523959) | 2.159446 / 1.468490 (0.690956) | 0.584571 / 4.584777 (-4.000206) | 4.392449 / 3.745712 (0.646737) | 4.790166 / 5.269862 (-0.479695) | 2.840615 / 4.565676 (-1.725062) | 0.070845 / 0.424275 (-0.353430) | 0.009112 / 0.007607 (0.001505) | 0.580251 / 0.226044 (0.354207) | 5.660311 / 2.268929 (3.391382) | 2.836136 / 55.444624 (-52.608489) | 2.412859 / 6.876477 (-4.463618) | 2.556710 / 2.142072 (0.414637) | 0.691946 / 4.805227 (-4.113282) | 0.160123 / 6.500664 (-6.340541) | 0.072593 / 0.075469 (-0.002876) |\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) | 1.547339 / 1.841788 (-0.294448) | 21.724793 / 8.074308 (13.650485) | 16.315304 / 10.191392 (6.123912) | 0.188733 / 0.680424 (-0.491690) | 0.022109 / 0.534201 (-0.512092) | 0.481623 / 0.579283 (-0.097660) | 0.464316 / 0.434364 (0.029952) | 0.557953 / 0.540337 (0.017615) | 0.756023 / 1.386936 (-0.630913) |\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) | 0.008637 / 0.011353 (-0.002716) | 0.005286 / 0.011008 (-0.005723) | 0.091387 / 0.038508 (0.052879) | 0.114092 / 0.023109 (0.090983) | 0.457547 / 0.275898 (0.181649) | 0.506878 / 0.323480 (0.183398) | 0.006849 / 0.007986 (-0.001137) | 0.004255 / 0.004328 (-0.000073) | 0.079556 / 0.004250 (0.075306) | 0.077729 / 0.037052 (0.040677) | 0.454094 / 0.258489 (0.195605) | 0.515812 / 0.293841 (0.221971) | 0.038271 / 0.128546 (-0.090275) | 0.010110 / 0.075646 (-0.065536) | 0.094254 / 0.419271 (-0.325017) | 0.065392 / 0.043533 (0.021860) | 0.459749 / 0.255139 (0.204610) | 0.489829 / 0.283200 (0.206629) | 0.040393 / 0.141683 (-0.101290) | 1.810414 / 1.452155 (0.358259) | 1.913212 / 1.492716 (0.420496) |\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) | 0.236898 / 0.018006 (0.218891) | 0.513118 / 0.000490 (0.512628) | 0.004432 / 0.000200 (0.004232) | 0.000115 / 0.000054 (0.000060) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.035074 / 0.037411 (-0.002337) | 0.102384 / 0.014526 (0.087858) | 0.117326 / 0.176557 (-0.059231) | 0.182596 / 0.737135 (-0.554539) | 0.116384 / 0.296338 (-0.179955) |\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) | 0.514544 / 0.215209 (0.299335) | 5.152930 / 2.077655 (3.075275) | 2.624477 / 1.504120 (1.120357) | 2.363209 / 1.541195 (0.822014) | 2.436060 / 1.468490 (0.967570) | 0.592523 / 4.584777 (-3.992254) | 4.209668 / 3.745712 (0.463956) | 6.284372 / 5.269862 (1.014511) | 3.667303 / 4.565676 (-0.898374) | 0.067017 / 0.424275 (-0.357259) | 0.008607 / 0.007607 (0.001000) | 0.600840 / 0.226044 (0.374796) | 5.992630 / 2.268929 (3.723701) | 3.114532 / 55.444624 (-52.330093) | 2.693242 / 6.876477 (-4.183235) | 2.767187 / 2.142072 (0.625115) | 0.687591 / 4.805227 (-4.117636) | 0.158477 / 6.500664 (-6.342187) | 0.075504 / 0.075469 (0.000034) |\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) | 1.605039 / 1.841788 (-0.236749) | 21.524730 / 8.074308 (13.450422) | 17.014643 / 10.191392 (6.823251) | 0.201580 / 0.680424 (-0.478843) | 0.023028 / 0.534201 (-0.511173) | 0.483801 / 0.579283 (-0.095482) | 0.490221 / 0.434364 (0.055857) | 0.589292 / 0.540337 (0.048955) | 0.758532 / 1.386936 (-0.628404) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#8c9c24d1d90f0c2db043ae2bc39f7c292454a58c \"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) | 0.008080 / 0.011353 (-0.003273) | 0.004859 / 0.011008 (-0.006149) | 0.101895 / 0.038508 (0.063387) | 0.091168 / 0.023109 (0.068059) | 0.378914 / 0.275898 (0.103016) | 0.417172 / 0.323480 (0.093692) | 0.006314 / 0.007986 (-0.001672) | 0.004069 / 0.004328 (-0.000259) | 0.076566 / 0.004250 (0.072315) | 0.070986 / 0.037052 (0.033934) | 0.380935 / 0.258489 (0.122446) | 0.417131 / 0.293841 (0.123290) | 0.036343 / 0.128546 (-0.092203) | 0.009996 / 0.075646 (-0.065650) | 0.346386 / 0.419271 (-0.072886) | 0.063162 / 0.043533 (0.019630) | 0.372620 / 0.255139 (0.117481) | 0.404902 / 0.283200 (0.121702) | 0.028217 / 0.141683 (-0.113466) | 1.793875 / 1.452155 (0.341721) | 1.836284 / 1.492716 (0.343568) |\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) | 0.223830 / 0.018006 (0.205823) | 0.503643 / 0.000490 (0.503153) | 0.004957 / 0.000200 (0.004757) | 0.000107 / 0.000054 (0.000053) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.035455 / 0.037411 (-0.001957) | 0.108015 / 0.014526 (0.093489) | 0.116887 / 0.176557 (-0.059669) | 0.188174 / 0.737135 (-0.548961) | 0.117217 / 0.296338 (-0.179121) |\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) | 0.471681 / 0.215209 (0.256472) | 4.694509 / 2.077655 (2.616855) | 2.369539 / 1.504120 (0.865419) | 2.176839 / 1.541195 (0.635644) | 2.300536 / 1.468490 (0.832045) | 0.575689 / 4.584777 (-4.009088) | 4.232765 / 3.745712 (0.487053) | 4.766775 / 5.269862 (-0.503087) | 2.864667 / 4.565676 (-1.701010) | 0.069390 / 0.424275 (-0.354885) | 0.008822 / 0.007607 (0.001214) | 0.559620 / 0.226044 (0.333576) | 5.580401 / 2.268929 (3.311472) | 2.920293 / 55.444624 (-52.524331) | 2.552166 / 6.876477 (-4.324311) | 2.795890 / 2.142072 (0.653818) | 0.687863 / 4.805227 (-4.117364) | 0.159129 / 6.500664 (-6.341535) | 0.073475 / 0.075469 (-0.001994) |\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) | 1.505892 / 1.841788 (-0.335896) | 24.127650 / 8.074308 (16.053342) | 16.758238 / 10.191392 (6.566846) | 0.200555 / 0.680424 (-0.479869) | 0.021596 / 0.534201 (-0.512605) | 0.480668 / 0.579283 (-0.098615) | 0.483528 / 0.434364 (0.049164) | 0.571241 / 0.540337 (0.030903) | 0.790547 / 1.386936 (-0.596390) |\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) | 0.007997 / 0.011353 (-0.003356) | 0.004842 / 0.011008 (-0.006166) | 0.077190 / 0.038508 (0.038681) | 0.092765 / 0.023109 (0.069656) | 0.457475 / 0.275898 (0.181577) | 0.523914 / 0.323480 (0.200434) | 0.006349 / 0.007986 (-0.001637) | 0.003902 / 0.004328 (-0.000427) | 0.075860 / 0.004250 (0.071609) | 0.069708 / 0.037052 (0.032656) | 0.459612 / 0.258489 (0.201123) | 0.555028 / 0.293841 (0.261187) | 0.036854 / 0.128546 (-0.091692) | 0.010078 / 0.075646 (-0.065568) | 0.083871 / 0.419271 (-0.335400) | 0.061221 / 0.043533 (0.017689) | 0.435737 / 0.255139 (0.180598) | 0.509700 / 0.283200 (0.226500) | 0.038091 / 0.141683 (-0.103592) | 1.777161 / 1.452155 (0.325006) | 1.859603 / 1.492716 (0.366886) |\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) | 0.250020 / 0.018006 (0.232014) | 0.486198 / 0.000490 (0.485708) | 0.007080 / 0.000200 (0.006880) | 0.000114 / 0.000054 (0.000060) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.038163 / 0.037411 (0.000751) | 0.110812 / 0.014526 (0.096286) | 0.122489 / 0.176557 (-0.054068) | 0.188215 / 0.737135 (-0.548920) | 0.122375 / 0.296338 (-0.173963) |\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) | 0.484534 / 0.215209 (0.269325) | 4.828654 / 2.077655 (2.751000) | 2.545102 / 1.504120 (1.040982) | 2.368867 / 1.541195 (0.827672) | 2.458042 / 1.468490 (0.989552) | 0.576372 / 4.584777 (-4.008404) | 4.814033 / 3.745712 (1.068321) | 6.175972 / 5.269862 (0.906110) | 4.033422 / 4.565676 (-0.532254) | 0.068544 / 0.424275 (-0.355731) | 0.008906 / 0.007607 (0.001299) | 0.581767 / 0.226044 (0.355723) | 5.808623 / 2.268929 (3.539695) | 3.120312 / 55.444624 (-52.324313) | 2.774834 / 6.876477 (-4.101642) | 2.770413 / 2.142072 (0.628340) | 0.692715 / 4.805227 (-4.112512) | 0.158883 / 6.500664 (-6.341782) | 0.075894 / 0.075469 (0.000425) |\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) | 1.631250 / 1.841788 (-0.210538) | 24.693250 / 8.074308 (16.618942) | 17.434790 / 10.191392 (7.243398) | 0.196456 / 0.680424 (-0.483968) | 0.022505 / 0.534201 (-0.511696) | 0.474788 / 0.579283 (-0.104495) | 0.500947 / 0.434364 (0.066583) | 0.553596 / 0.540337 (0.013259) | 0.737767 / 1.386936 (-0.649169) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#f87d6e6394bf4b390ccc82235eb7667f874e5d43 \"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) | 0.006629 / 0.011353 (-0.004724) | 0.004115 / 0.011008 (-0.006894) | 0.083934 / 0.038508 (0.045426) | 0.074952 / 0.023109 (0.051843) | 0.313069 / 0.275898 (0.037171) | 0.345878 / 0.323480 (0.022398) | 0.006034 / 0.007986 (-0.001952) | 0.003413 / 0.004328 (-0.000916) | 0.065130 / 0.004250 (0.060880) | 0.057363 / 0.037052 (0.020310) | 0.314483 / 0.258489 (0.055994) | 0.352626 / 0.293841 (0.058785) | 0.031325 / 0.128546 (-0.097221) | 0.008577 / 0.075646 (-0.067069) | 0.288137 / 0.419271 (-0.131135) | 0.053651 / 0.043533 (0.010118) | 0.313006 / 0.255139 (0.057867) | 0.338668 / 0.283200 (0.055468) | 0.023709 / 0.141683 (-0.117974) | 1.481209 / 1.452155 (0.029054) | 1.559801 / 1.492716 (0.067085) |\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) | 0.211543 / 0.018006 (0.193537) | 0.452185 / 0.000490 (0.451696) | 0.003177 / 0.000200 (0.002977) | 0.000078 / 0.000054 (0.000024) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.028821 / 0.037411 (-0.008591) | 0.083290 / 0.014526 (0.068765) | 0.097478 / 0.176557 (-0.079079) | 0.153506 / 0.737135 (-0.583629) | 0.097054 / 0.296338 (-0.199284) |\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) | 0.385847 / 0.215209 (0.170638) | 3.835629 / 2.077655 (1.757974) | 1.880938 / 1.504120 (0.376819) | 1.711848 / 1.541195 (0.170653) | 1.785099 / 1.468490 (0.316609) | 0.486256 / 4.584777 (-4.098521) | 3.629026 / 3.745712 (-0.116686) | 3.321578 / 5.269862 (-1.948283) | 2.024314 / 4.565676 (-2.541363) | 0.058097 / 0.424275 (-0.366179) | 0.007724 / 0.007607 (0.000117) | 0.458293 / 0.226044 (0.232249) | 4.581314 / 2.268929 (2.312386) | 2.314379 / 55.444624 (-53.130246) | 1.966089 / 6.876477 (-4.910387) | 2.203824 / 2.142072 (0.061752) | 0.611581 / 4.805227 (-4.193647) | 0.149166 / 6.500664 (-6.351498) | 0.059825 / 0.075469 (-0.015644) |\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) | 1.235546 / 1.841788 (-0.606242) | 19.747439 / 8.074308 (11.673131) | 14.628383 / 10.191392 (4.436991) | 0.193074 / 0.680424 (-0.487350) | 0.020327 / 0.534201 (-0.513874) | 0.397051 / 0.579283 (-0.182232) | 0.418491 / 0.434364 (-0.015873) | 0.462055 / 0.540337 (-0.078282) | 0.637524 / 1.386936 (-0.749412) |\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) | 0.007069 / 0.011353 (-0.004284) | 0.004106 / 0.011008 (-0.006902) | 0.065818 / 0.038508 (0.027310) | 0.077101 / 0.023109 (0.053991) | 0.363323 / 0.275898 (0.087425) | 0.399463 / 0.323480 (0.075983) | 0.005540 / 0.007986 (-0.002446) | 0.003480 / 0.004328 (-0.000849) | 0.065176 / 0.004250 (0.060926) | 0.060867 / 0.037052 (0.023815) | 0.365763 / 0.258489 (0.107273) | 0.407789 / 0.293841 (0.113949) | 0.032018 / 0.128546 (-0.096528) | 0.008550 / 0.075646 (-0.067096) | 0.071750 / 0.419271 (-0.347521) | 0.050625 / 0.043533 (0.007092) | 0.361434 / 0.255139 (0.106295) | 0.384799 / 0.283200 (0.101599) | 0.026104 / 0.141683 (-0.115579) | 1.496093 / 1.452155 (0.043938) | 1.592909 / 1.492716 (0.100193) |\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) | 0.185794 / 0.018006 (0.167787) | 0.453379 / 0.000490 (0.452890) | 0.004365 / 0.000200 (0.004165) | 0.000092 / 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) | 0.031666 / 0.037411 (-0.005746) | 0.088323 / 0.014526 (0.073798) | 0.104602 / 0.176557 (-0.071954) | 0.159827 / 0.737135 (-0.577308) | 0.103725 / 0.296338 (-0.192614) |\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) | 0.413509 / 0.215209 (0.198300) | 4.126071 / 2.077655 (2.048416) | 2.137088 / 1.504120 (0.632968) | 1.981034 / 1.541195 (0.439839) | 2.063660 / 1.468490 (0.595170) | 0.478798 / 4.584777 (-4.105979) | 3.642801 / 3.745712 (-0.102911) | 3.428994 / 5.269862 (-1.840867) | 2.031902 / 4.565676 (-2.533774) | 0.056244 / 0.424275 (-0.368032) | 0.007365 / 0.007607 (-0.000242) | 0.484371 / 0.226044 (0.258327) | 4.838537 / 2.268929 (2.569608) | 2.559497 / 55.444624 (-52.885127) | 2.251863 / 6.876477 (-4.624614) | 2.339227 / 2.142072 (0.197155) | 0.607228 / 4.805227 (-4.198000) | 0.133877 / 6.500664 (-6.366787) | 0.062049 / 0.075469 (-0.013420) |\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) | 1.350389 / 1.841788 (-0.491399) | 20.060359 / 8.074308 (11.986051) | 14.305675 / 10.191392 (4.114283) | 0.165642 / 0.680424 (-0.514782) | 0.018206 / 0.534201 (-0.515994) | 0.396907 / 0.579283 (-0.182376) | 0.431896 / 0.434364 (-0.002468) | 0.475778 / 0.540337 (-0.064559) | 0.644688 / 1.386936 (-0.742248) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#8f6fa96ae5de873a49ef28739e8f64edf8b18cae \"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) | 0.009048 / 0.011353 (-0.002305) | 0.005787 / 0.011008 (-0.005221) | 0.111617 / 0.038508 (0.073109) | 0.087603 / 0.023109 (0.064494) | 0.446481 / 0.275898 (0.170583) | 0.491726 / 0.323480 (0.168247) | 0.007052 / 0.007986 (-0.000934) | 0.004481 / 0.004328 (0.000152) | 0.084331 / 0.004250 (0.080081) | 0.072006 / 0.037052 (0.034953) | 0.454238 / 0.258489 (0.195749) | 0.496749 / 0.293841 (0.202908) | 0.049027 / 0.128546 (-0.079520) | 0.014005 / 0.075646 (-0.061641) | 0.372550 / 0.419271 (-0.046722) | 0.071414 / 0.043533 (0.027881) | 0.459432 / 0.255139 (0.204293) | 0.467332 / 0.283200 (0.184133) | 0.037539 / 0.141683 (-0.104144) | 1.869179 / 1.452155 (0.417024) | 1.983641 / 1.492716 (0.490925) |\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) | 0.265426 / 0.018006 (0.247419) | 0.672527 / 0.000490 (0.672037) | 0.001152 / 0.000200 (0.000953) | 0.000181 / 0.000054 (0.000127) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.032967 / 0.037411 (-0.004445) | 0.103023 / 0.014526 (0.088497) | 0.115978 / 0.176557 (-0.060578) | 0.191698 / 0.737135 (-0.545438) | 0.117867 / 0.296338 (-0.178471) |\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) | 0.602208 / 0.215209 (0.386999) | 6.147784 / 2.077655 (4.070129) | 2.768933 / 1.504120 (1.264813) | 2.415619 / 1.541195 (0.874424) | 2.456159 / 1.468490 (0.987669) | 0.836270 / 4.584777 (-3.748507) | 5.447754 / 3.745712 (1.702042) | 7.751825 / 5.269862 (2.481963) | 4.591892 / 4.565676 (0.026215) | 0.108269 / 0.424275 (-0.316006) | 0.009626 / 0.007607 (0.002019) | 0.719260 / 0.226044 (0.493216) | 7.313442 / 2.268929 (5.044514) | 3.490739 / 55.444624 (-51.953885) | 2.743543 / 6.876477 (-4.132934) | 3.035071 / 2.142072 (0.892999) | 1.042791 / 4.805227 (-3.762436) | 0.217080 / 6.500664 (-6.283584) | 0.084286 / 0.075469 (0.008817) |\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) | 1.655427 / 1.841788 (-0.186361) | 25.386536 / 8.074308 (17.312228) | 21.740666 / 10.191392 (11.549274) | 0.246388 / 0.680424 (-0.434036) | 0.029723 / 0.534201 (-0.504478) | 0.491537 / 0.579283 (-0.087746) | 0.603495 / 0.434364 (0.169131) | 0.573938 / 0.540337 (0.033600) | 0.981875 / 1.386936 (-0.405061) |\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) | 0.009664 / 0.011353 (-0.001689) | 0.006446 / 0.011008 (-0.004562) | 0.085113 / 0.038508 (0.046605) | 0.094533 / 0.023109 (0.071424) | 0.498388 / 0.275898 (0.222490) | 0.540127 / 0.323480 (0.216647) | 0.007316 / 0.007986 (-0.000670) | 0.004252 / 0.004328 (-0.000077) | 0.086292 / 0.004250 (0.082041) | 0.067956 / 0.037052 (0.030903) | 0.507664 / 0.258489 (0.249175) | 0.554324 / 0.293841 (0.260483) | 0.050107 / 0.128546 (-0.078439) | 0.014277 / 0.075646 (-0.061370) | 0.098838 / 0.419271 (-0.320433) | 0.066053 / 0.043533 (0.022521) | 0.491090 / 0.255139 (0.235951) | 0.537432 / 0.283200 (0.254232) | 0.035937 / 0.141683 (-0.105746) | 1.820715 / 1.452155 (0.368561) | 1.996268 / 1.492716 (0.503552) |\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) | 0.300859 / 0.018006 (0.282852) | 0.610958 / 0.000490 (0.610468) | 0.000474 / 0.000200 (0.000274) | 0.000098 / 0.000054 (0.000044) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.036372 / 0.037411 (-0.001039) | 0.109115 / 0.014526 (0.094589) | 0.122802 / 0.176557 (-0.053755) | 0.187092 / 0.737135 (-0.550044) | 0.123432 / 0.296338 (-0.172906) |\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) | 0.646979 / 0.215209 (0.431770) | 6.577713 / 2.077655 (4.500058) | 3.004606 / 1.504120 (1.500486) | 2.661183 / 1.541195 (1.119989) | 2.726717 / 1.468490 (1.258227) | 0.889497 / 4.584777 (-3.695280) | 5.485055 / 3.745712 (1.739343) | 4.852043 / 5.269862 (-0.417819) | 3.177392 / 4.565676 (-1.388285) | 0.099796 / 0.424275 (-0.324479) | 0.009868 / 0.007607 (0.002261) | 0.819919 / 0.226044 (0.593874) | 7.911255 / 2.268929 (5.642326) | 3.839877 / 55.444624 (-51.604747) | 3.088663 / 6.876477 (-3.787813) | 3.371184 / 2.142072 (1.229112) | 1.072762 / 4.805227 (-3.732466) | 0.224536 / 6.500664 (-6.276128) | 0.083415 / 0.075469 (0.007946) |\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) | 1.754426 / 1.841788 (-0.087361) | 25.546690 / 8.074308 (17.472382) | 22.998252 / 10.191392 (12.806860) | 0.258019 / 0.680424 (-0.422405) | 0.030104 / 0.534201 (-0.504097) | 0.518406 / 0.579283 (-0.060877) | 0.605753 / 0.434364 (0.171389) | 0.599630 / 0.540337 (0.059292) | 0.819042 / 1.386936 (-0.567894) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#350f4fd6caabbdfacb5fbf9193ab255c3d0daa4c \"CML watermark\")\n" ]

[]

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

[]

[]

[]

[]

[]

[ "Hmm the doctest failed again - maybe the full result needs to be on one single line", "cc @lhoestq for the final review", "Cool ! Feel free to merge if it's all good for you" ]

[ { "color": "0e8a16", "default": false, "description": "Contribution to a dataset script", "id": 4564477500, "name": "dataset contribution", "node_id": "LA_kwDODunzps8AAAABEBBmPA", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20contribution" } ]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Some CI fails are unrelated to your PR and fixed on master, feel free to merge master into your branch :)", "Thanks @nateraw for the addition of this dataset.\r\n\r\nI would suggest to transfer it to the Hugging Face Hub, under a \"pascal\" organization namespace: \"pascal/voc\".\r\n\r\nWhat do you think?", "FYI I think this dataset is also available at (internal) https://huggingface.co/datasets/HuggingFaceM4/pascal_voc", "@lhoestq @albertvillanova what do you think best path forward is? No idea when I'll get to looking at this again, but would be nice to know plan so when I find time I can just get it done in one sitting. ", "My (not strong) opinion on this:\r\n- as we are removing dataset scripts from GitHub, this dataset should be created directly on the Hub\r\n- I proposed doing it under some kind of \"official\" org namespace, like pascal or pascal2; other suggestions are welcome\r\n- the link given by @lhoestq might serve as inspiration for your implementation (I think yours misses data about action classification): their implementation comprises tasks: classification/detection, segmentation, action classification, person layout; it misses other tasks though\r\n\r\nWhat do you think?" ]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_5992). All of your documentation changes will be reflected on that endpoint." ]

[]

[ "Fix #2552." ]

[]

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

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_4296). All of your documentation changes will be reflected on that endpoint." ]

[]

[ "@albertvillanova Is there a specific reason you’re adding the Pile under “the” instead of under “pile”? That does not appear to be consistent with other datasets.", "Hi @StellaAthena,\r\n\r\nI asked myself the same question, but at the end I decided to be consistent with previously added Pile subsets:\r\n- #2817\r\n\r\nI guess the reason is to stress that the definite article is always used before the name of the dataset (your site says: \"The Pile. An 800GB Dataset of Diverse Text for Language Modeling\"). Other datasets are not usually preceded by the definite article, like \"the SQuAD\" or \"the GLUE\" or \"the Common Voice\"...\r\n\r\nCC: @lhoestq ", "> I guess the reason is to stress that the definite article is always used before the name of the dataset (your site says: \"The Pile. An 800GB Dataset of Diverse Text for Language Modeling\").\r\n\r\nYes that's because of this that it starts with \"the\"" ]

[]

[]

[]

[ "Looks great to me! ", "The new download manager is ready. I removed the old folder and I fixed a few remaining dependencies.\r\nI tested it on squad and a few others from the dataset folder and it works fine.\r\n\r\nThe only impact of these changes is that it breaks the `download_and_prepare` script that was used to register the checksums when we create a dataset, as the checksum logic is not implemented.\r\n\r\nLet me know if you have remarks", "Ok merged it (a bit fast for you to update the copyright, now I see that. but it's ok, we'll do a pass on these doc/copyright before releasing anyway)", "Actually two additional things here @lhoestq (I merged too fast sorry, let's make a new PR for additional developments):\r\n- I think we can remove some dependencies now (e.g. `promises`) in setup.py, can you have a look?\r\n- also, I think we can remove the boto3 dependency like here: https://github.com/huggingface/transformers/pull/3968" ]

[ { "color": "0e8a16", "default": false, "description": "Contribution to a dataset script", "id": 4564477500, "name": "dataset contribution", "node_id": "LA_kwDODunzps8AAAABEBBmPA", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20contribution" } ]

[ "Thanks for your contribution, @ssss1029. Are you still interested in adding this dataset?\r\n\r\nWe are removing the dataset scripts from this GitHub repo and moving them to the Hugging Face Hub: https://huggingface.co/datasets\r\n\r\nWe would suggest you create this dataset there. Please, feel free to tell us if you need some help." ]

[]

[ "This can also mean one of the files was not downloaded correctly.\r\n\r\nWe log an erroneous file's name before raising the reader's error, so this is how you can find the problematic file. Then, you should delete it and call `load_dataset` again.\r\n\r\n(I checked all the uploaded files, and they seem to be valid Parquet files, so I don't think this is a bug on their side)\r\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" } ]

[ "@albertvillanova ", "Hi @Aktsvigun, thanks for reporting.\r\n\r\nI'm investigating this.", "Hi @albertvillanova ,\r\nany update on this? Can I probably help in some way?", "Hi @Aktsvigun! We are planning to address this issue before our next release, in a couple of weeks at most. 😅 \r\n\r\nIn the meantime, if you would like to contribute, feel free to open a Pull Request. You are welcome. Here you can find more information: [How to contribute to Datasets?](CONTRIBUTING.md)", "I can't reproduce the bug on master. I believe this issue was fixed by https://github.com/huggingface/datasets/pull/3551." ]

[]

[]

[ { "color": "0e8a16", "default": false, "description": "Contribution to a dataset script", "id": 4564477500, "name": "dataset contribution", "node_id": "LA_kwDODunzps8AAAABEBBmPA", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20contribution" } ]

[ "Thanks for your contribution, @shayne-longpre.\r\n\r\nAre you still interested in adding this dataset? As we are transferring the dataset scripts from this GitHub repo, we would recommend you to add this to the Hugging Face Hub: https://huggingface.co/datasets" ]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_3946). All of your documentation changes will be reflected on that endpoint.", "The failing CI test has nothing to do with this PR." ]

[]

[ "I changed the way I implemented fingerprint updates to use decorator functions.\r\n\r\nI also added a new attribute called `_inplace_history` that stores the in-place history of transforms (like cast_, rename_columns, etc.). This history is useful to replay the changes that were done in-place when unpickling a dataset that is memory mapped from a file.\r\n\r\nLet me know what you think @thomwolf " ]

[ { "color": "0e8a16", "default": false, "description": "Contribution to a dataset script", "id": 4564477500, "name": "dataset contribution", "node_id": "LA_kwDODunzps8AAAABEBBmPA", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20contribution" } ]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_4557). All of your documentation changes will be reflected on that endpoint.", "> Just to confirm: we should add this metadata via GitHub and not Hub PRs for canonical datasets right?\r\n\r\nyes :)", "As discussed with @lewtun, we are closing this PR, because it requires first the task names to be aligned between AutoTrain and datasets." ]

[]

[]

[]

[]

[]

[ "@lhoestq I saw some common changes you made on the other PR's (Similar Opus Datasets). I fixed those changes here. Can you please review it once ? \r\nThanks.", "merging since the CI is fixed on master" ]

[]

[]

[]

[]

[]

[]

[]

[ "merging since the CI is fixed on master" ]

[]

[]

[ { "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 @sanderland ! Thanks for reporting :) This is a bug, I opened a PR to fix it. We'll do a new release soon\r\n\r\nIn the meantime you can fix it by specifying in advance that the \"label\" are integers:\r\n```python\r\nimport numpy as np\r\n\r\nds = Dataset.from_dict(\r\n {\r\n \"text\": [\"the lazy dog jumps over the quick fox\", \"another sentence\"],\r\n \"label\": [[], []],\r\n }\r\n)\r\n# explicitly say that the \"label\" type is int64, even though it contains only null values\r\nds = ds.cast_column(\"label\", Sequence(Value(\"int64\")))\r\n\r\ndef mapper(features):\r\n features['label'] = [\r\n [0,0,0] for l in features['label']\r\n ]\r\n return features\r\n\r\nds_mapped = ds.map(mapper,batched=True)\r\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" } ]

[ "I did some investigation and, as it seems, the bug stems from [this line](https://github.com/huggingface/datasets/blob/8004d7c3e1d74b29c3e5b0d1660331cd26758363/src/datasets/arrow_dataset.py#L325). The lifecycle of the dataset from the linked line is bound to one of the returned `tf.data.Dataset`. So my (hacky) solution involves wrapping the linked dataset with `weakref.proxy` and adding a custom `__del__` to `tf.python.data.ops.dataset_ops.TensorSliceDataset` (this is the type of a dataset that is returned by `tf.data.Dataset.from_tensor_slices`; this works for TF 2.x, but I'm not sure `tf.python.data.ops.dataset_ops` is a valid path for TF 1.x) that deletes the linked dataset, which is assigned to the dataset object as a property. Will open a draft PR soon!", "Thanks a lot for investigating !" ]

[]

[]

[]

[ "Thanks @brainshawn for this update" ]

[]

[ "Hi ! Thanks this will be very useful :)\r\n\r\nIt looks like there are some changes in the github diff that are not related to your contribution, can you try fixing this by merging `master` into your PR, or create a new PR from an updated version of `master` ?", "I have already solved the conflict on this latest version. This is my first time sending PR, if there's anything I need to adjust just let me know~", "Thanks, most changes are gone :)\r\nIt still seems to include changes though - do you mind try creating a new branch from upstream/master and create a new PR please ?", "Yeah sure, I'll try to send a new PR today!", "Please forward to [#3850](https://github.com/huggingface/datasets/pull/3850)", "Thanks ! Closing this one in favor of https://github.com/huggingface/datasets/pull/3850/files" ]

[]

[ "I see, master is already fixed in https://github.com/huggingface/datasets/commit/9e058f098a0919efd03a136b9b9c3dec5076f626" ]

[ { "color": "d876e3", "default": true, "description": "Further information is requested", "id": 1935892912, "name": "question", "node_id": "MDU6TGFiZWwxOTM1ODkyOTEy", "url": "https://api.github.com/repos/huggingface/datasets/labels/question" } ]

[ "@albertvillanova @lhoestq , could you please help me on this issue. ", "Hi ! As mentioned on the [forum](https://discuss.huggingface.co/t/how-to-wrap-a-generator-with-hf-dataset/18464), the simplest for now would be to define a [dataset script](https://huggingface.co/docs/datasets/dataset_script) which can contain your generator. But we can also explore adding something like `ds = Dataset.from_iterable(seqio_dataset)`", "@lhoestq , hey i did as you instructed, but sadly i cannot get pass through the download_manager, as i dont have anything to download. i was skipping the ` def _split_generators(self, dl_manager):` function. but i cannot get around it. I get a `NotImplementedError: `\r\n\r\nthe following is my code for the same: \r\n\r\n\r\n\r\n```\r\nimport datasets \r\nimport functools\r\nimport glob \r\nfrom datasets import load_from_disk\r\nimport seqio\r\nimport tensorflow as tf\r\nimport t5.data\r\nfrom datasets import load_dataset\r\nfrom t5.data import postprocessors\r\nfrom t5.data import preprocessors\r\nfrom t5.evaluation import metrics\r\nfrom seqio import FunctionDataSource, utils\r\n\r\nTaskRegistry = seqio.TaskRegistry\r\n\r\ndata_path = glob.glob(\"/home/stephen/Desktop/MEGA_CORPUS/COMBINED_CORPUS/*\", recursive=False)\r\n\r\n\r\ndef gen_dataset(split, shuffle=False, seed=None, column=\"text\", dataset_path=None):\r\n dataset = load_from_disk(dataset_path)\r\n if shuffle:\r\n if seed:\r\n dataset = dataset.shuffle(seed=seed)\r\n else:\r\n dataset = dataset.shuffle()\r\n while True:\r\n for item in dataset[str(split)]:\r\n yield item[column]\r\n\r\n\r\ndef dataset_fn(split, shuffle_files, seed=None, dataset_path=None):\r\n return tf.data.Dataset.from_generator(\r\n functools.partial(gen_dataset, split, shuffle_files, seed, dataset_path=dataset_path),\r\n output_signature=tf.TensorSpec(shape=(), dtype=tf.string, name=dataset_path)\r\n )\r\n\r\[email protected]_over_dataset\r\ndef target_to_key(x, key_map, target_key):\r\n \"\"\"Assign the value from the dataset to target_key in key_map\"\"\"\r\n return {**key_map, target_key: x}\r\n\r\n\r\n_CITATION = \"Not ready yet\"\r\n_DESCRIPTION = \"a custom seqio based mixed samples on a given temperature value, that again returns a dataset in HF dataset format well samples on the Mixture temperature\"\r\n_HOMEPAGE = \"ldcil.org\"\r\n\r\nclass CustomSeqio(datasets.GeneratorBasedBuilder):\r\n\r\n def _info(self):\r\n return datasets.DatasetInfo(\r\n description=_DESCRIPTION,\r\n features=datasets.Features(\r\n {\r\n \"text\": datasets.Value(\"string\"),\r\n }\r\n ),\r\n homepage=\"https://ldcil.org\",\r\n citation=_CITATION,)\r\n\r\ndef generate_examples(self):\r\n seqio_train_list = []\r\n for lang in data_path:\r\n dataset_name = lang.split(\"/\")[-1]\r\n dataset_shapes = None \r\n\r\n TaskRegistry.add(\r\n str(dataset_name),\r\n source=seqio.FunctionDataSource(\r\n dataset_fn=functools.partial(dataset_fn, dataset_path=lang),\r\n splits=(\"train\", \"test\"),\r\n caching_permitted=False,\r\n num_input_examples=dataset_shapes,\r\n ),\r\n preprocessors=[\r\n functools.partial(\r\n target_to_key, key_map={\r\n \"targets\": None,\r\n }, target_key=\"targets\")],\r\n output_features={\"targets\": seqio.Feature(vocabulary=seqio.PassThroughVocabulary, add_eos=False, dtype=tf.string, rank=0)},\r\n metric_fns=[]\r\n )\r\n\r\n seqio_train_dataset = seqio.get_mixture_or_task(dataset_name).get_dataset(\r\n sequence_length=None,\r\n split=\"train\",\r\n shuffle=True,\r\n num_epochs=1,\r\n shard_info=seqio.ShardInfo(index=0, num_shards=10),\r\n use_cached=False,\r\n seed=42)\r\n seqio_train_list.append(seqio_train_dataset)\r\n \r\n lang_name_list = []\r\n for lang in data_path:\r\n lang_name = lang.split(\"/\")[-1]\r\n lang_name_list.append(lang_name)\r\n\r\n seqio_mixture = seqio.MixtureRegistry.add(\r\n \"seqio_mixture\",\r\n lang_name_list,\r\n default_rate=0.7)\r\n \r\n seqio_mixture_dataset = seqio.get_mixture_or_task(\"seqio_mixture\").get_dataset(\r\n sequence_length=None,\r\n split=\"train\",\r\n shuffle=True,\r\n num_epochs=1,\r\n shard_info=seqio.ShardInfo(index=0, num_shards=10),\r\n use_cached=False,\r\n seed=42)\r\n\r\n for id, ex in enumerate(seqio_mixture_dataset):\r\n yield id, {\"text\": ex[\"targets\"].numpy().decode()}\r\n```\r\n\r\nand i load it by:\r\n\r\n`seqio_mixture = load_dataset(\"seqio_loader\")`", "@lhoestq , just to make things clear ... \r\n\r\nthe following is my original code, thats not in the HF dataset loading script: \r\n\r\n```\r\nimport functools\r\nimport seqio\r\nimport tensorflow as tf\r\nimport t5.data\r\nfrom datasets import load_from_disk\r\nfrom t5.data import postprocessors\r\nfrom t5.data import preprocessors\r\nfrom t5.evaluation import metrics\r\nfrom seqio import FunctionDataSource, utils\r\nimport glob \r\n\r\nTaskRegistry = seqio.TaskRegistry\r\n\r\n\r\n\r\ndef gen_dataset(split, shuffle=False, seed=None, column=\"text\", dataset_path=None):\r\n dataset = load_from_disk(dataset_path)\r\n if shuffle:\r\n if seed:\r\n dataset = dataset.shuffle(seed=seed)\r\n else:\r\n dataset = dataset.shuffle()\r\n while True:\r\n for item in dataset[str(split)]:\r\n yield item[column]\r\n\r\n\r\ndef dataset_fn(split, shuffle_files, seed=None, dataset_path=None):\r\n return tf.data.Dataset.from_generator(\r\n functools.partial(gen_dataset, split, shuffle_files, seed, dataset_path=dataset_path),\r\n output_signature=tf.TensorSpec(shape=(), dtype=tf.string, name=dataset_path)\r\n )\r\n\r\n\r\[email protected]_over_dataset\r\ndef target_to_key(x, key_map, target_key):\r\n \"\"\"Assign the value from the dataset to target_key in key_map\"\"\"\r\n return {**key_map, target_key: x}\r\n\r\ndata_path = glob.glob(\"/home/stephen/Desktop/MEGA_CORPUS/COMBINED_CORPUS/*\", recursive=False)\r\n\r\nseqio_train_list = []\r\n\r\nfor lang in data_path:\r\n dataset_name = lang.split(\"/\")[-1]\r\n dataset_shapes = None \r\n\r\n TaskRegistry.add(\r\n str(dataset_name),\r\n source=seqio.FunctionDataSource(\r\n dataset_fn=functools.partial(dataset_fn, dataset_path=lang),\r\n splits=(\"train\", \"test\"),\r\n caching_permitted=False,\r\n num_input_examples=dataset_shapes,\r\n ),\r\n preprocessors=[\r\n functools.partial(\r\n target_to_key, key_map={\r\n \"targets\": None,\r\n }, target_key=\"targets\")],\r\n output_features={\"targets\": seqio.Feature(vocabulary=seqio.PassThroughVocabulary, add_eos=False, dtype=tf.string, rank=0)},\r\n metric_fns=[]\r\n )\r\n\r\n seqio_train_dataset = seqio.get_mixture_or_task(dataset_name).get_dataset(\r\n sequence_length=None,\r\n split=\"train\",\r\n shuffle=True,\r\n num_epochs=1,\r\n shard_info=seqio.ShardInfo(index=0, num_shards=10),\r\n use_cached=False,\r\n seed=42)\r\n seqio_train_list.append(seqio_train_dataset)\r\n\r\nlang_name_list = []\r\nfor lang in data_path:\r\n lang_name = lang.split(\"/\")[-1]\r\n lang_name_list.append(lang_name)\r\n\r\nseqio_mixture = seqio.MixtureRegistry.add(\r\n \"seqio_mixture\",\r\n lang_name_list,\r\n default_rate=0.7\r\n)\r\n\r\nseqio_mixture_dataset = seqio.get_mixture_or_task(\"seqio_mixture\").get_dataset(\r\n sequence_length=None,\r\n split=\"train\",\r\n shuffle=True,\r\n num_epochs=1,\r\n shard_info=seqio.ShardInfo(index=0, num_shards=10),\r\n use_cached=False,\r\n seed=42)\r\n\r\nfor _, ex in zip(range(15), seqio_mixture_dataset):\r\n print(ex[\"targets\"].numpy().decode())\r\n```\r\n\r\nwhere the seqio_mixture_dataset is the generator that i wanted to be wrapped in HF dataset. \r\n\r\nalso additionally, could you please tell me how do i set the `default_rate=0.7` args where `seqio_mixture` is defined to be made as a custom option in the HF load_dataset() method,\r\n\r\nmaybe like this: \r\n`seqio_mixture_dataset = datasets.load_dataset(\"seqio_loader\",temperature=0.5)`", "I like the idea of having `Dataset.from_iterable(iterable)` in the API. The only problem is that we also want to make this part cachable, which is tricky if `iterable` is a generator. \r\n\r\nSome resources on this issue:\r\n* https://github.com/uqfoundation/dill/issues/311\r\n* https://stackoverflow.com/questions/7180212/why-cant-generators-be-pickled\r\n* https://github.com/tonyroberts/generator_tools - python package for pickling generators; pickles bytecode, so it creates version-specific dumps", "For the caching maybe we can have `Dataset.from_generator` as TF and pickle+hash the generator function (not the generator object itself) ?\r\n\r\nAnd then keep `Dataset.from_iterable` fo pickable objects like lists", "@lhoestq, @mariosasko do you too have any examples where the dataset is a generator and needs to be wrapped into hf dataset ? ", "@lhoestq, following to my previous question ... what possibly could be done in this [link1](https://github.com/huggingface/datasets/issues/4417#issuecomment-1146627404) [link2](https://github.com/huggingface/datasets/issues/4417#issuecomment-1146627593) case? do you have any ideas? ", "@lhoestq +1 for the `Dataset.from_generator` idea.\r\n\r\nHaving thought about it, let's avoid adding `Dataset.from_iterable` to the API since dictionaries are technically iteralbles (\"iterable\" is a broad term in Python), and we already provide `Dataset.from_dict`. And for lists maybe we can add `Dataset.from_list` similar to `pa.Table.from_pylist`. WDYT?\r\n", "Hi @StephennFernandes!\r\n\r\nTo fix the issues in the copied code, rename `generate_examples` to` _generate_examples` and add one level of indentation as this is a method of `GeneratorBasedBuilder` and define `_split_generators` as follows (again as a method of `GeneratorBasedBuilder):\r\n```python\r\n def _split_generators(self, dl_manager):\r\n return [\r\n datasets.SplitGenerator(\r\n name=datasets.Split.TRAIN,\r\n gen_kwargs={},\r\n ),\r\n ]\r\n```\r\n\r\nAnd if you are feeling extra adventurous, you can try to use ArrowWriter to directly create a cache file:\r\n```python\r\nfrom datasets import Dataset\r\nfrom datasets.arrow_writer import ArrowWriter\r\n\r\nwriter = ArrowWriter(path=\"path/to/cache_file.arrow\", writer_batch_size=1000)\r\n\r\nwith writer:\r\n for ex in generator:\r\n writer.write(ex) \r\n writer.finalize()\r\n\r\ndset = Dataset.from_file(\"path/to/cache_file.arrow\")\r\n```\r\n\r\n", "I have a problem which I think is very similar: I would like to \"stream\" data to a HF Array (memory-mapped) Dataset, where the final size of the dataset is unknown, but could be much larger than what fits into memory.\r\nWhat I want to end up with is an Array Dataset which I can open using `Dataset.load_from_disk(dataset_path=\"somename\")` and use e.g. as the training set. \r\n\r\nFor this I would have thought there should be an API which allows me to open/create the dataset (and define the features etc), then write examples to the dataset, but I could not find a way to do this. \r\n\r\nI tried doing this and it looks like it works, but it feels very hacky and I am not sure if this might fail to update some of the fields in the json files which may turn out to be important:\r\n```\r\nfrom datasets import Dataset, Features, ClassLabel, Sequence, Value\r\nfrom datasets.arrow_writer import ArrowWriter \r\n# 1) define the features\r\nfeatures = Features(dict(\r\n id=Value(dtype=\"string\"),\r\n tokens=Sequence(feature=Value(dtype=\"string\")),\r\n ner_tags=Sequence(feature=ClassLabel(names=['O', 'B-corporation', 'I-corporation', 'B-creative-work', 'I-creative-work', 'B-group', 'I-group', 'B-location', 'I-location', 'B-person', 'I-person', 'B-product', 'I-product'])),\r\n))\r\n# 2) create empty dataset for examples with these features and store to disk\r\nempty = dict(\r\n id = [],\r\n tokens = [],\r\n ner_tags = [],\r\n)\r\nds = Dataset.from_dict(empty, features=features)\r\nds.save_to_disk(dataset_path=\"debug_ds1\")\r\n\r\n# 3) directly write all the examples to the arrow dataset \r\nwith ArrowWriter(path=\"debug_ds1/dataset.arrow\") as writer: \r\n writer.write(dict(id=0, tokens=[\"a\", \"b\"], ner_tags=[0, 0])) \r\n writer.write(dict(id=1, tokens=[\"x\", \"y\"], ner_tags=[1, 0])) \r\n writer.finalize() \r\n \r\nds2 = Dataset.load_from_disk(dataset_path=\"debug_ds1\")\r\nlen(ds2)\r\n```\r\nIs there a cleaner/proper way to do this?\r\n\r\nI like the sound of `Dataset.from_iterable` or `Dataset.from_generator` (should not from iterable be able to handle from generator too as all generators are iterables?) but how would I define the features for me examples there? ", "Hi @johann-petrak! You can pass the features directly to ArrowWriter's initializer like so `ArrowWriter(..., features=features)`.\r\n\r\nAnd the reason why I prefer `Dataset.from_generator` over `Dataset.from_iterable` is mentioned in one of my previous comments.", "@mariosasko so at the moment we still have to create a fake `Dataset` first and then use `ArrowWriter` to write an actual dataset? I'm using the latest version of `datasets` on pypi but my final file is always empty. Is there anything wrong with the code below?\r\n\r\n```python\r\n total = 0\r\n with ArrowWriter(path=str(final_data_path), features=features) as writer:\r\n for batch in loader:\r\n for traj in batch:\r\n for generator in question_generators:\r\n for xi in generator(traj):\r\n # print(f\"Question: {xi.question}, answer: {xi.answer}\")\r\n total += 1\r\n writer.write(\r\n {\r\n \"id\": f\"qa_{total}\",\r\n \"question\": xi.question,\r\n \"answer\": xi.answer,\r\n }\r\n )\r\n writer.finalize()\r\n print(f\"Total #questions = {total}\") # this prints 402\r\n```", "This works for me if I then (actually I also close the writer: `writer.close()`) open the Arrow file as a dataset using `ds=Dataset.from_file(final_data_path)` then `ds.save_to_disk(somedir)`. The Dataset created that way contains the expected examples.", "Oh thanks. That did the trick I believe. Shouldn't ArrowWriter have a context manager that does these operations?", "You can just use `Dataset.from_file` to get your dataset, no need to do an extra `save_to_disk` somewhere else ;)", "I was thinking that `save_to_disk` is necessary when one wants to re-use that dataset as a proper HF dataset later, no?\r\nAt least what I wanted to achieve is create a dataset that can be opened like any other local or remote dataset. ", "`save_to_disk`/`load_from_disk` is indeed more general, e.g. it supports datasets that consist in several files, and saves some extra info in a dataset_info.json file (description, citation, split sizes, etc.)\r\n\r\nIf you have one single file it's fine to simply do `.from_file()`" ]

[]

[]

[]

[ "This issue seems to have been resolved, so I'm closing it." ]

[]

[]

[]

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

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_6047). All of your documentation changes will be reflected on that endpoint.", "<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) | 0.006384 / 0.011353 (-0.004969) | 0.003872 / 0.011008 (-0.007136) | 0.083454 / 0.038508 (0.044946) | 0.069120 / 0.023109 (0.046011) | 0.312573 / 0.275898 (0.036675) | 0.345814 / 0.323480 (0.022334) | 0.005729 / 0.007986 (-0.002257) | 0.003225 / 0.004328 (-0.001103) | 0.063950 / 0.004250 (0.059700) | 0.053998 / 0.037052 (0.016946) | 0.316492 / 0.258489 (0.058003) | 0.350738 / 0.293841 (0.056897) | 0.030770 / 0.128546 (-0.097776) | 0.008474 / 0.075646 (-0.067173) | 0.286989 / 0.419271 (-0.132282) | 0.052473 / 0.043533 (0.008940) | 0.314361 / 0.255139 (0.059222) | 0.335170 / 0.283200 (0.051970) | 0.022885 / 0.141683 (-0.118798) | 1.465430 / 1.452155 (0.013275) | 1.527799 / 1.492716 (0.035083) |\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) | 0.209377 / 0.018006 (0.191371) | 0.455583 / 0.000490 (0.455094) | 0.003352 / 0.000200 (0.003152) | 0.000080 / 0.000054 (0.000025) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.026284 / 0.037411 (-0.011127) | 0.080710 / 0.014526 (0.066185) | 0.091741 / 0.176557 (-0.084816) | 0.147602 / 0.737135 (-0.589534) | 0.091173 / 0.296338 (-0.205166) |\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) | 0.386592 / 0.215209 (0.171383) | 3.856665 / 2.077655 (1.779011) | 1.835745 / 1.504120 (0.331625) | 1.671814 / 1.541195 (0.130619) | 1.711224 / 1.468490 (0.242734) | 0.484704 / 4.584777 (-4.100073) | 3.649239 / 3.745712 (-0.096473) | 3.784051 / 5.269862 (-1.485810) | 2.241195 / 4.565676 (-2.324482) | 0.056613 / 0.424275 (-0.367662) | 0.007140 / 0.007607 (-0.000467) | 0.464585 / 0.226044 (0.238540) | 4.616537 / 2.268929 (2.347609) | 2.371969 / 55.444624 (-53.072656) | 1.977754 / 6.876477 (-4.898723) | 2.083385 / 2.142072 (-0.058687) | 0.582330 / 4.805227 (-4.222897) | 0.132744 / 6.500664 (-6.367920) | 0.059822 / 0.075469 (-0.015647) |\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) | 1.259566 / 1.841788 (-0.582221) | 18.990166 / 8.074308 (10.915858) | 13.992069 / 10.191392 (3.800677) | 0.160001 / 0.680424 (-0.520423) | 0.018622 / 0.534201 (-0.515579) | 0.392921 / 0.579283 (-0.186362) | 0.418225 / 0.434364 (-0.016139) | 0.471252 / 0.540337 (-0.069086) | 0.653227 / 1.386936 (-0.733709) |\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) | 0.006641 / 0.011353 (-0.004712) | 0.003738 / 0.011008 (-0.007271) | 0.064053 / 0.038508 (0.025545) | 0.069467 / 0.023109 (0.046357) | 0.360625 / 0.275898 (0.084727) | 0.394291 / 0.323480 (0.070811) | 0.005236 / 0.007986 (-0.002750) | 0.003304 / 0.004328 (-0.001024) | 0.064078 / 0.004250 (0.059827) | 0.054605 / 0.037052 (0.017552) | 0.374567 / 0.258489 (0.116078) | 0.411227 / 0.293841 (0.117386) | 0.031614 / 0.128546 (-0.096933) | 0.008323 / 0.075646 (-0.067324) | 0.070616 / 0.419271 (-0.348656) | 0.050077 / 0.043533 (0.006544) | 0.362229 / 0.255139 (0.107090) | 0.388310 / 0.283200 (0.105110) | 0.024053 / 0.141683 (-0.117630) | 1.508913 / 1.452155 (0.056759) | 1.562140 / 1.492716 (0.069423) |\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) | 0.230172 / 0.018006 (0.212165) | 0.449363 / 0.000490 (0.448873) | 0.002374 / 0.000200 (0.002174) | 0.000097 / 0.000054 (0.000043) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.029813 / 0.037411 (-0.007598) | 0.087298 / 0.014526 (0.072772) | 0.096712 / 0.176557 (-0.079845) | 0.152864 / 0.737135 (-0.584271) | 0.098204 / 0.296338 (-0.198135) |\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) | 0.408664 / 0.215209 (0.193455) | 4.075068 / 2.077655 (1.997414) | 2.096365 / 1.504120 (0.592245) | 1.936096 / 1.541195 (0.394901) | 1.961872 / 1.468490 (0.493382) | 0.483383 / 4.584777 (-4.101394) | 3.686926 / 3.745712 (-0.058787) | 4.798824 / 5.269862 (-0.471037) | 2.652279 / 4.565676 (-1.913398) | 0.056695 / 0.424275 (-0.367580) | 0.007592 / 0.007607 (-0.000016) | 0.484710 / 0.226044 (0.258665) | 4.842153 / 2.268929 (2.573225) | 2.636828 / 55.444624 (-52.807796) | 2.243666 / 6.876477 (-4.632811) | 2.375972 / 2.142072 (0.233899) | 0.578544 / 4.805227 (-4.226683) | 0.132579 / 6.500664 (-6.368085) | 0.061287 / 0.075469 (-0.014182) |\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) | 1.360287 / 1.841788 (-0.481501) | 19.464110 / 8.074308 (11.389802) | 14.530875 / 10.191392 (4.339483) | 0.149479 / 0.680424 (-0.530944) | 0.018471 / 0.534201 (-0.515730) | 0.395399 / 0.579283 (-0.183884) | 0.412897 / 0.434364 (-0.021467) | 0.465194 / 0.540337 (-0.075144) | 0.611752 / 1.386936 (-0.775184) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#79a535de98b590da7bc223a6498c59790882f14a \"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) | 0.008986 / 0.011353 (-0.002367) | 0.005104 / 0.011008 (-0.005905) | 0.108371 / 0.038508 (0.069863) | 0.091655 / 0.023109 (0.068546) | 0.430183 / 0.275898 (0.154285) | 0.481387 / 0.323480 (0.157907) | 0.006662 / 0.007986 (-0.001324) | 0.004681 / 0.004328 (0.000353) | 0.089325 / 0.004250 (0.085075) | 0.065096 / 0.037052 (0.028044) | 0.435021 / 0.258489 (0.176532) | 0.478635 / 0.293841 (0.184794) | 0.047628 / 0.128546 (-0.080918) | 0.013496 / 0.075646 (-0.062150) | 0.389661 / 0.419271 (-0.029611) | 0.082260 / 0.043533 (0.038727) | 0.474165 / 0.255139 (0.219026) | 0.464877 / 0.283200 (0.181677) | 0.039784 / 0.141683 (-0.101899) | 1.874694 / 1.452155 (0.422539) | 1.980183 / 1.492716 (0.487467) |\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) | 0.254044 / 0.018006 (0.236038) | 0.631495 / 0.000490 (0.631005) | 0.000628 / 0.000200 (0.000428) | 0.000086 / 0.000054 (0.000032) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.038773 / 0.037411 (0.001362) | 0.103681 / 0.014526 (0.089156) | 0.125081 / 0.176557 (-0.051476) | 0.198345 / 0.737135 (-0.538790) | 0.122217 / 0.296338 (-0.174121) |\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) | 0.611677 / 0.215209 (0.396468) | 6.220790 / 2.077655 (4.143135) | 2.729858 / 1.504120 (1.225739) | 2.351944 / 1.541195 (0.810749) | 2.449137 / 1.468490 (0.980647) | 0.896842 / 4.584777 (-3.687935) | 5.537491 / 3.745712 (1.791778) | 8.480182 / 5.269862 (3.210320) | 5.251404 / 4.565676 (0.685728) | 0.100449 / 0.424275 (-0.323826) | 0.009008 / 0.007607 (0.001401) | 0.750060 / 0.226044 (0.524016) | 7.390940 / 2.268929 (5.122011) | 3.478256 / 55.444624 (-51.966369) | 2.883597 / 6.876477 (-3.992880) | 3.082256 / 2.142072 (0.940183) | 1.114339 / 4.805227 (-3.690889) | 0.225389 / 6.500664 (-6.275275) | 0.083972 / 0.075469 (0.008503) |\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) | 1.741522 / 1.841788 (-0.100266) | 25.674700 / 8.074308 (17.600392) | 24.324412 / 10.191392 (14.133020) | 0.257878 / 0.680424 (-0.422546) | 0.038384 / 0.534201 (-0.495817) | 0.508302 / 0.579283 (-0.070981) | 0.612979 / 0.434364 (0.178615) | 0.584366 / 0.540337 (0.044029) | 0.881115 / 1.386936 (-0.505821) |\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) | 0.009114 / 0.011353 (-0.002239) | 0.005333 / 0.011008 (-0.005675) | 0.094944 / 0.038508 (0.056436) | 0.099178 / 0.023109 (0.076068) | 0.529813 / 0.275898 (0.253915) | 0.551282 / 0.323480 (0.227802) | 0.006442 / 0.007986 (-0.001543) | 0.004283 / 0.004328 (-0.000045) | 0.084257 / 0.004250 (0.080007) | 0.067557 / 0.037052 (0.030504) | 0.514733 / 0.258489 (0.256244) | 0.568200 / 0.293841 (0.274359) | 0.050969 / 0.128546 (-0.077577) | 0.014495 / 0.075646 (-0.061151) | 0.097089 / 0.419271 (-0.322182) | 0.063142 / 0.043533 (0.019609) | 0.513327 / 0.255139 (0.258188) | 0.520593 / 0.283200 (0.237394) | 0.036824 / 0.141683 (-0.104859) | 1.954875 / 1.452155 (0.502720) | 1.976307 / 1.492716 (0.483591) |\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) | 0.304070 / 0.018006 (0.286063) | 0.611073 / 0.000490 (0.610583) | 0.005027 / 0.000200 (0.004827) | 0.000113 / 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) | 0.037993 / 0.037411 (0.000582) | 0.115876 / 0.014526 (0.101350) | 0.118087 / 0.176557 (-0.058469) | 0.186437 / 0.737135 (-0.550699) | 0.129883 / 0.296338 (-0.166456) |\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) | 0.658292 / 0.215209 (0.443083) | 6.618257 / 2.077655 (4.540602) | 3.203786 / 1.504120 (1.699667) | 2.858714 / 1.541195 (1.317519) | 2.940974 / 1.468490 (1.472484) | 0.856238 / 4.584777 (-3.728538) | 5.427708 / 3.745712 (1.681996) | 4.810048 / 5.269862 (-0.459813) | 3.120006 / 4.565676 (-1.445671) | 0.098098 / 0.424275 (-0.326177) | 0.010077 / 0.007607 (0.002470) | 0.790890 / 0.226044 (0.564845) | 7.956679 / 2.268929 (5.687750) | 3.955710 / 55.444624 (-51.488914) | 3.446419 / 6.876477 (-3.430057) | 3.541228 / 2.142072 (1.399156) | 1.013420 / 4.805227 (-3.791808) | 0.213741 / 6.500664 (-6.286923) | 0.080857 / 0.075469 (0.005388) |\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) | 1.813265 / 1.841788 (-0.028522) | 25.965199 / 8.074308 (17.890891) | 21.892761 / 10.191392 (11.701369) | 0.257843 / 0.680424 (-0.422580) | 0.029388 / 0.534201 (-0.504813) | 0.510609 / 0.579283 (-0.068674) | 0.626579 / 0.434364 (0.192215) | 0.576865 / 0.540337 (0.036528) | 0.826610 / 1.386936 (-0.560326) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#a1a9c00249b330f97f66ceb86c2939261091f4fe \"CML watermark\")\n" ]

[]