Datasets:

---

Francesco-A

/

github-issues_huggingface-datasets

like 0

[]

[ "Thanks for reporting.\r\nThat's because the download url has changed. The old url now redirects to the new one but we don't support redirection for downloads.\r\n\r\nI'm opening a PR to update the url" ]

[]

[ "Hi ! I don't fully understand this change @albertvillanova \r\nThe `_extract` method used to return the compound URL that points to the root of the inside of the archive.\r\nThis way users can use the usual os.path.join or other functions to point to the relevant files. I don't see why you're using a glob pattern ?", "This change is to allow this:\r\n```python\r\ndata_files = f\"https://huggingface.co/datasets/albertvillanova/datasets-tests-compression/resolve/main/sample.zip\"\r\nds = load_dataset(\"json\", split=\"train\", data_files=data_files, streaming=True)\r\nassert isinstance(ds, IterableDataset)\r\n```\r\nNote that in this case the user will not call os.path.join.\r\n\r\nBefore this PR it gave error because pointing to the root, without any subsequent join, gives error:\r\n```python\r\nfsspec.open(\"zip://::https://huggingface.co/datasets/albertvillanova/datasets-tests-compression/resolve/main/sample.zip\")\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" } ]

[ "Hi @Dhanachandra, \r\n\r\nWe have tests for all our metrics and they work as expected: under the hood, we use scikit-learn implementations.\r\n\r\nMaybe the cause is somewhere else. For example:\r\n- Is it a binary or a multiclass or a multilabel classification? Default computation of these metrics is for binary classification; if you would like multiclass or multilabel, you should pass the corresponding parameters; see their documentation (e.g.: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.precision_score.html) or code below:\r\n\r\nhttps://huggingface.co/docs/datasets/using_metrics.html#computing-the-metric-scores\r\n\r\n```python\r\nIn [1]: from datasets import load_metric\r\n\r\nIn [2]: precision = load_metric(\"precision\")\r\n\r\nIn [3]: print(precision.inputs_description)\r\n\r\nArgs:\r\n predictions: Predicted labels, as returned by a model.\r\n references: Ground truth labels.\r\n labels: The set of labels to include when average != 'binary', and\r\n their order if average is None. Labels present in the data can\r\n be excluded, for example to calculate a multiclass average ignoring\r\n a majority negative class, while labels not present in the data will\r\n result in 0 components in a macro average. For multilabel targets,\r\n labels are column indices. By default, all labels in y_true and\r\n y_pred are used in sorted order.\r\n average: This parameter is required for multiclass/multilabel targets.\r\n If None, the scores for each class are returned. Otherwise, this\r\n determines the type of averaging performed on the data:\r\n binary: Only report results for the class specified by pos_label.\r\n This is applicable only if targets (y_{true,pred}) are binary.\r\n micro: Calculate metrics globally by counting the total true positives,\r\n false negatives and false positives.\r\n macro: Calculate metrics for each label, and find their unweighted mean.\r\n This does not take label imbalance into account.\r\n weighted: Calculate metrics for each label, and find their average\r\n weighted by support (the number of true instances for each label).\r\n This alters ‘macro’ to account for label imbalance; it can result\r\n in an F-score that is not between precision and recall.\r\n samples: Calculate metrics for each instance, and find their average\r\n (only meaningful for multilabel classification).\r\n sample_weight: Sample weights.\r\n\r\nReturns:\r\n precision: Precision score.\r\n\r\nExamples:\r\n\r\n >>> precision_metric = datasets.load_metric(\"precision\")\r\n >>> results = precision_metric.compute(references=[0, 1], predictions=[0, 1])\r\n >>> print(results)\r\n {'precision': 1.0}\r\n\r\n >>> predictions = [0, 2, 1, 0, 0, 1]\r\n >>> references = [0, 1, 2, 0, 1, 2]\r\n >>> results = precision_metric.compute(predictions=predictions, references=references, average='macro')\r\n >>> print(results)\r\n {'precision': 0.2222222222222222}\r\n >>> results = precision_metric.compute(predictions=predictions, references=references, average='micro')\r\n >>> print(results)\r\n {'precision': 0.3333333333333333}\r\n >>> results = precision_metric.compute(predictions=predictions, references=references, average='weighted')\r\n >>> print(results)\r\n {'precision': 0.2222222222222222}\r\n >>> results = precision_metric.compute(predictions=predictions, references=references, average=None)\r\n >>> print(results)\r\n {'precision': array([0.66666667, 0. , 0. ])}\r\n```\r\n" ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Duplicate of:\r\n- #4059" ]

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

[]

[]

[ "Thanks for reporting !\r\nI was able to reproduce thanks to your code and find the origin of the bug.\r\nThe cache was not reusing the same file because one object was not deterministic. It comes from a conversion from `set` to `list` in the `datasets.arrrow_dataset.transmit_format` function, where the resulting list would not always be in the same order and therefore the function that computes the hash used by the cache would not always return the same result.\r\nI'm opening a PR to fix this.\r\n\r\nAlso we plan to do a new release in the coming days so you can expect the fix to be available soon.\r\nNote that you can still specify `cache_file_name=` in the second `map()` call to name the cache file yourself if you want to.", "Thanks for the fast reply, waiting for the fix :)\r\n\r\nI tried to use `cache_file_names` and wasn't sure how, I tried to give it the following:\r\n```\r\ntokenized_datasets = tokenized_datasets.map(\r\n group_texts,\r\n batched=True,\r\n num_proc=60,\r\n load_from_cache_file=True,\r\n cache_file_names={k: f'.cache/{str(k)}' for k in tokenized_datasets}\r\n)\r\n```\r\n\r\nand got an error:\r\n```\r\nmultiprocess.pool.RemoteTraceback:\r\n\"\"\"\r\nTraceback (most recent call last):\r\n File \"/venv/lib/python3.6/site-packages/multiprocess/pool.py\", line 119, in worker\r\n result = (True, func(*args, **kwds))\r\n File \"/venv/lib/python3.6/site-packages/datasets/arrow_dataset.py\", line 157, in wrapper\r\n out: Union[\"Dataset\", \"DatasetDict\"] = func(self, *args, **kwargs)\r\n File \"/venv/lib/python3.6/site-packages/datasets/fingerprint.py\", line 163, in wrapper\r\n out = func(self, *args, **kwargs)\r\n File \"/venv/lib/python3.6/site-packages/datasets/arrow_dataset.py\", line 1491, in _map_single\r\n tmp_file = tempfile.NamedTemporaryFile(\"wb\", dir=os.path.dirname(cache_file_name), delete=False)\r\n File \"/usr/lib/python3.6/tempfile.py\", line 690, in NamedTemporaryFile\r\n (fd, name) = _mkstemp_inner(dir, prefix, suffix, flags, output_type)\r\n File \"/usr/lib/python3.6/tempfile.py\", line 401, in _mkstemp_inner\r\n fd = _os.open(file, flags, 0o600)\r\nFileNotFoundError: [Errno 2] No such file or directory: '_00000_of_00060.cache/tmpsvszxtop'\r\n\"\"\"\r\n\r\nThe above exception was the direct cause of the following exception:\r\n\r\nTraceback (most recent call last):\r\n File \"test.py\", line 48, in <module>\r\n cache_file_names={k: f'.cache/{str(k)}' for k in tokenized_datasets}\r\n File \"/venv/lib/python3.6/site-packages/datasets/dataset_dict.py\", line 303, in map\r\n for k, dataset in self.items()\r\n File \"/venv/lib/python3.6/site-packages/datasets/dataset_dict.py\", line 303, in <dictcomp>\r\n for k, dataset in self.items()\r\n File \"/venv/lib/python3.6/site-packages/datasets/arrow_dataset.py\", line 1317, in map\r\n transformed_shards = [r.get() for r in results]\r\n File \"/venv/lib/python3.6/site-packages/datasets/arrow_dataset.py\", line 1317, in <listcomp>\r\n transformed_shards = [r.get() for r in results]\r\n File \"/venv/lib/python3.6/site-packages/multiprocess/pool.py\", line 644, in get\r\n raise self._value\r\nFileNotFoundError: [Errno 2] No such file or directory: '_00000_of_00060.cache/tmpsvszxtop'\r\n```\r\n", "The documentation says\r\n```\r\ncache_file_names (`Optional[Dict[str, str]]`, defaults to `None`): Provide the name of a cache file to use to store the\r\n results of the computation instead of the automatically generated cache file name.\r\n You have to provide one :obj:`cache_file_name` per dataset in the dataset dictionary.\r\n```\r\nWhat is expected is simply the name of a file, not a path. The file will be located in the cache directory of the `wikitext` dataset. You can try again with something like\r\n```python\r\ncache_file_names = {k: f'tokenized_and_grouped_{str(k)}' for k in tokenized_datasets}\r\n```", "Managed to get `cache_file_names` working and caching works well with it\r\nHad to make a small modification for it to work:\r\n```\r\ncache_file_names = {k: f'tokenized_and_grouped_{str(k)}.arrow' for k in tokenized_datasets}\r\n```", "Another comment on `cache_file_names`, it doesn't save the produced cached files in the dataset's cache folder, it requires to give a path to an existing directory for it to work.\r\nI can confirm that this is how it works in `datasets==1.1.3`", "Oh yes indeed ! Maybe we need to update the docstring to mention that it is a path", "I fixed the docstring. Hopefully this is less confusing now: https://github.com/huggingface/datasets/commit/42ccc0012ba8864e6db1392430100f350236183a", "I upgraded to the latest version and I encountered some strange behaviour, the script I posted in the OP doesn't trigger recalculation, however, if I add the following change it does trigger partial recalculation, I am not sure if its something wrong on my machine or a bug:\r\n```\r\nfrom datasets import load_dataset\r\nfrom transformers import AutoTokenizer\r\n\r\ndatasets = load_dataset('wikitext', 'wikitext-103-raw-v1')\r\ntokenizer = AutoTokenizer.from_pretrained('bert-base-uncased', use_fast=True)\r\n\r\ncolumn_names = datasets[\"train\"].column_names\r\ntext_column_name = \"text\" if \"text\" in column_names else column_names[0]\r\ndef tokenize_function(examples):\r\n return tokenizer(examples[text_column_name], return_special_tokens_mask=True)\r\n# CHANGE\r\nprint('hello')\r\n# CHANGE\r\n\r\ntokenized_datasets = datasets.map(\r\n tokenize_function,\r\n batched=True,\r\n...\r\n```\r\nI am using datasets in the `run_mlm.py` script in the transformers examples and I found that if I change the script without touching any of the preprocessing. it still triggers recalculation which is very weird\r\n\r\nEdit: accidently clicked the close issue button ", "This is because the `group_texts` line definition changes (it is defined 3 lines later than in the previous call). Currently if a function is moved elsewhere in a script we consider it to be different.\r\n\r\nNot sure this is actually a good idea to keep this behavior though. We had this as a security in the early development of the lib but now the recursive hashing of objects is robust so we can probably remove that.\r\nMoreover we're already ignoring the line definition for lambda functions.", "I opened a PR to change this, let me know what you think.", "Sounds great, thank you for your quick responses and help! Looking forward for the next release.", "I am having a similar issue where only the grouped files are loaded from cache while the tokenized ones aren't. I can confirm both datasets are being stored to file, but only the grouped version is loaded from cache. Not sure what might be going on. But I've tried to remove all kinds of non deterministic behaviour, but still no luck. Thanks for the help!\r\n\r\n\r\n```python\r\n # Datasets\r\n train = sorted(glob(args.data_dir + '*.{}'.format(args.ext)))\r\n if args.dev_split >= len(train):\r\n raise ValueError(\"Not enough dev files\")\r\n dev = []\r\n state = random.Random(1001)\r\n for _ in range(args.dev_split):\r\n dev.append(train.pop(state.randint(0, len(train) - 1)))\r\n\r\n max_seq_length = min(args.max_seq_length, tokenizer.model_max_length)\r\n\r\n def tokenize_function(examples):\r\n return tokenizer(examples['text'], return_special_tokens_mask=True)\r\n\r\n def group_texts(examples):\r\n # Concatenate all texts from our dataset and generate chunks of max_seq_length\r\n concatenated_examples = {k: sum(examples[k], []) for k in examples.keys()}\r\n total_length = len(concatenated_examples[list(examples.keys())[0]])\r\n # Truncate (not implementing padding)\r\n total_length = (total_length // max_seq_length) * max_seq_length\r\n # Split by chunks of max_seq_length\r\n result = {\r\n k: [t[i : i + max_seq_length] for i in range(0, total_length, max_seq_length)]\r\n for k, t in concatenated_examples.items()\r\n }\r\n return result\r\n\r\n datasets = load_dataset(\r\n 'text', name='DBNL', data_files={'train': train[:10], 'dev': dev[:5]}, \r\n cache_dir=args.data_cache_dir)\r\n datasets = datasets.map(tokenize_function, \r\n batched=True, remove_columns=['text'], \r\n cache_file_names={k: os.path.join(args.data_cache_dir, f'{k}-tokenized') for k in datasets},\r\n load_from_cache_file=not args.overwrite_cache)\r\n datasets = datasets.map(group_texts, \r\n batched=True,\r\n cache_file_names={k: os.path.join(args.data_cache_dir, f'{k}-grouped') for k in datasets},\r\n load_from_cache_file=not args.overwrite_cache)\r\n```\r\n\r\nAnd this is the log\r\n\r\n```\r\n04/26/2021 10:26:59 - WARNING - datasets.builder - Using custom data configuration DBNL-f8d988ad33ccf2c1\r\n04/26/2021 10:26:59 - WARNING - datasets.builder - Reusing dataset text (/home/manjavacasema/data/.cache/text/DBNL-f8d988ad33ccf2c1/0.0.0/e16f44aa1b321ece1f87b07977cc5d70be93d69b20486d6dacd62e12cf25c9a5)\r\n100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 13/13 [00:00<00:00, 21.07ba/s]\r\n100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 40/40 [00:01<00:00, 24.28ba/s]\r\n04/26/2021 10:27:01 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at /home/manjavacasema/data/.cache/train-grouped\r\n04/26/2021 10:27:01 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at /home/manjavacasema/data/.cache/dev-grouped\r\n```\r\n", "Hi ! What tokenizer are you using ?", "It's the ByteLevelBPETokenizer", "This error happened to me too, when I tried to supply my own fingerprint to `map()` via the `new_fingerprint` arg.\r\n\r\nEdit: realized it was because my path was weird and had colons and brackets and slashes in it, since one of the variable values I included in the fingerprint was a dataset split like \"train[:10%]\". I fixed it with [this solution](https://stackoverflow.com/a/13593932/2287177) from StackOverflow to just remove those invalid characters from the fingerprint.", "Good catch @jxmorris12, maybe we should do additional checks on the valid characters for fingerprints ! Would you like to contribute this ?\r\n\r\nI think this can be added here, when we set the fingerprint(s) that are passed `map`:\r\n\r\nhttps://github.com/huggingface/datasets/blob/25bb7c9cbf519fbbf9abf3898083b529e7762705/src/datasets/fingerprint.py#L449-L454\r\n\r\nmaybe something like\r\n```python\r\nif kwargs.get(fingerprint_name) is None:\r\n ...\r\nelse:\r\n # In this case, it's the user who specified the fingerprint manually:\r\n # we need to make sure it's a valid hash\r\n validate_fingerprint(kwargs[fingerprint_name])\r\n```\r\n\r\nOtherwise I can open a PR later", "I opened a PR here to add the fingerprint validation: https://github.com/huggingface/datasets/pull/4587\r\n\r\nEDIT: merged :)", "thank you!" ]

[]

[ "Thanks for reporting, @sauravtii.\r\n\r\nUnfortunately, I'm not able to reproduce the issue:\r\n```python\r\nIn [1]: from datasets import load_dataset\r\n\r\nIn [2]: ds = load_dataset(\"josianem/imdb\")\r\n\r\nIn [2]: ds\r\nOut[2]: \r\nDatasetDict({\r\n train: Dataset({\r\n features: ['text', 'label'],\r\n num_rows: 25799\r\n })\r\n test: Dataset({\r\n features: ['text', 'label'],\r\n num_rows: 25000\r\n })\r\n unsupervised: Dataset({\r\n features: ['text', 'label'],\r\n num_rows: 50000\r\n })\r\n})\r\n```\r\n\r\nCould you please retry to load the dataset? Maybe there was a temporary connection issue to Dropbox.", "Thanks @albertvillanova. I am facing another issue now\r\n\r\n```\r\nTraceback (most recent call last):\r\n File \"sample.py\", line 4, in <module>\r\n dataset = load_dataset(\"josianem/imdb\")\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/load.py\", line 1112, in load_dataset\r\n builder_instance.download_and_prepare(\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/builder.py\", line 636, in download_and_prepare\r\n self._download_and_prepare(\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/builder.py\", line 738, in _download_and_prepare\r\n verify_splits(self.info.splits, split_dict)\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/utils/info_utils.py\", line 74, in verify_splits\r\n raise NonMatchingSplitsSizesError(str(bad_splits))\r\ndatasets.utils.info_utils.NonMatchingSplitsSizesError: [{'expected': SplitInfo(name='train', num_bytes=34501348, num_examples=25799, dataset_name='imdb'), 'recorded': SplitInfo(name='train', num_bytes=0, num_examples=0, dataset_name='imdb')}, {'expected': SplitInfo(name='test', num_bytes=32650697, num_examples=25000, dataset_name='imdb'), 'recorded': SplitInfo(name='test', num_bytes=0, num_examples=0, dataset_name='imdb')}, {'expected': SplitInfo(name='unsupervised', num_bytes=67106814, num_examples=50000, dataset_name='imdb'), 'recorded': SplitInfo(name='unsupervised', num_bytes=0, num_examples=0, dataset_name='imdb')}]\r\n```\r\n\r\nThis is my code\r\n\r\n```\r\nfrom datasets import load_dataset, load_metric\r\n\r\ndataset = load_dataset(\"josianem/imdb\")\r\n```", "Your connection didn't work and you got an empty dataset (`num_bytes=0, num_examples=0`):\r\n```\r\ndatasets.utils.info_utils.NonMatchingSplitsSizesError: \r\n[\r\n {\r\n 'expected': SplitInfo(name='train', num_bytes=34501348, num_examples=25799, dataset_name='imdb'), \r\n 'recorded': SplitInfo(name='train', num_bytes=0, num_examples=0, dataset_name='imdb')\r\n }, \r\n {\r\n 'expected': SplitInfo(name='test', num_bytes=32650697, num_examples=25000, dataset_name='imdb'), \r\n 'recorded': SplitInfo(name='test', num_bytes=0, num_examples=0, dataset_name='imdb')\r\n }, \r\n {\r\n 'expected': SplitInfo(name='unsupervised', num_bytes=67106814, num_examples=50000, dataset_name='imdb'), \r\n 'recorded': SplitInfo(name='unsupervised', num_bytes=0, num_examples=0, dataset_name='imdb')\r\n }\r\n]\r\n```\r\n\r\nCould you please try the link in your browser and see if it works? https://www.dropbox.com/s/zts98j4vkqtsns6/aclImdb_v2.tar?dl=1\r\n- If it does not work, you should contact the author of the dataset in their Community tab (https://huggingface.co/datasets/josianem/imdb/discussions) and inform them, so that they can host their data elsewhere, for example on the Hugging Face Hub itself\r\n\r\nIf the link works, you should try to load the dataset but forcing the re-download of the data files (so that the cache is refreshed with the actual data file), by passing `download_mode=\"force_redownload\"`:\r\n```python\r\ndataset = load_dataset(\"josianem/imdb\", download_mode=\"force_redownload\")\r\n```", "After pasting the link in the browser, it did start the download so it seems that the link is working. But even after including the `download_mode` in my code I am facing the same issue:\r\n\r\nError:\r\n```\r\nTraceback (most recent call last):\r\n File \"sample.py\", line 4, in <module>\r\n dataset = load_dataset(\"josianem/imdb\", download_mode=\"force_redownload\")\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/load.py\", line 1112, in load_dataset\r\n builder_instance.download_and_prepare(\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/builder.py\", line 636, in download_and_prepare\r\n self._download_and_prepare(\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/builder.py\", line 704, in _download_and_prepare\r\n split_generators = self._split_generators(dl_manager, **split_generators_kwargs)\r\n File \"/home/saurav/.cache/huggingface/modules/datasets_modules/datasets/imdb/cc6ab4acab2799be15d5d217c24548b856156dafdc850165fdc4f2031f27ff2f/imdb.py\", line 79, in _split_generators\r\n archive = dl_manager.download(_DOWNLOAD_URL)\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/utils/download_manager.py\", line 196, in download\r\n downloaded_path_or_paths = map_nested(\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/utils/py_utils.py\", line 197, in map_nested\r\n return function(data_struct)\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/utils/download_manager.py\", line 217, in _download\r\n return cached_path(url_or_filename, download_config=download_config)\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/utils/file_utils.py\", line 289, in cached_path\r\n output_path = get_from_cache(\r\n File \"/home/saurav/.local/lib/python3.8/site-packages/datasets/utils/file_utils.py\", line 606, in get_from_cache\r\n raise ConnectionError(\"Couldn't reach {}\".format(url))\r\nConnectionError: Couldn't reach https://www.dropbox.com/s/zts98j4vkqtsns6/aclImdb_v2.tar?dl=1\r\n```\r\n\r\nMy code:\r\n```\r\nfrom datasets import load_dataset, load_metric\r\n\r\ndataset = load_dataset(\"josianem/imdb\", download_mode=\"force_redownload\")\r\n```", "I have tried again to reproduce your issue without success: the dataset loads perfectly, both in my local machine and in a Colab notebook.\r\n- See: https://colab.research.google.com/drive/1dky3T0XGFuldggy22NNQQN-UqOFqvnuY?usp=sharing\r\n\r\nI think the cause maight be that you are using a very old version of `datasets`. Please, could you update it and retry?\r\n```\r\npip install -U datasets\r\n```", "That worked!! Thanks @albertvillanova : )\r\n\r\n```\r\nDownloading builder script: 100%|███████| 4.20k/4.20k [00:00<00:00, 6.69MB/s]\r\nDownloading metadata: 100%|█████████████| 2.60k/2.60k [00:00<00:00, 3.41MB/s]\r\nDownloading readme: 100%|███████████████| 7.52k/7.52k [00:00<00:00, 12.6MB/s]\r\nDownloading and preparing dataset imdb/plain_text to /home/saurav/.cache/huggingface/datasets/josianem___imdb/plain_text/1.0.0/cc6ab4acab2799be15d5d217c24548b856156dafdc850165fdc4f2031f27ff2f...\r\nDownloading data: 100%|███████████████████| 301M/301M [01:32<00:00, 3.25MB/s]\r\nDataset imdb downloaded and prepared to /home/saurav/.cache/huggingface/datasets/josianem___imdb/plain_text/1.0.0/cc6ab4acab2799be15d5d217c24548b856156dafdc850165fdc4f2031f27ff2f. Subsequent calls will reuse this data.\r\n100%|█████████████████████████████████████████| 3/3 [00:00<00:00, 794.83it/s]\r\n```\r\n\r\nThe code I used:\r\n```\r\nfrom datasets import load_dataset, load_metric\r\n\r\ndataset = load_dataset(\"josianem/imdb\", download_mode=\"force_redownload\")\r\n\r\n```\r\n\r\nBut when I remove `download_mode=\"force_redownload\"` I get the same error. Any guess on that?", "That is because the cache got the \"empty\" download file the first time you tried and got the connection error.\r\n\r\nThen, once you no longer get the connection error, you need to refresh the cache by passing `download_mode=\"force_redownload\"`." ]

[]

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

[]

[]

[]

[]

[]

[]

[]

[]

[]

[ "Note that the tests seem to fail because of a bug in an Exception at the moment, see: https://github.com/huggingface/datasets/pull/2936 for the fix", "@lhoestq implemented your changes, I think this might be ready for another look.", "Thanks @lhoestq, implemented the changes, let me know if anything else pops up." ]

[ { "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 ! AFAIK a dataset must be present on a local disk to be able to efficiently memory map the datasets Arrow files. What makes you think that it is possible to load from a cloud storage and have good performance ?\r\n\r\nAnyway it's already possible to download_and_prepare a dataset as Arrow files in a cloud storage with:\r\n```python\r\nbuilder = load_dataset_builder(..., cache_dir=\"/temp/dir\")\r\nbuilder.download_and_prepare(\"/cloud_dir\")\r\n```\r\n\r\nbut then \r\n```python\r\nds = builder.as_dataset()\r\n```\r\nwould fail if \"/cloud_dir\" is not a local directory.", "In my use case, I am trying to mount the S3 bucket as local system with S3FS-FUSE / [goofys](https://github.com/kahing/goofys). I want to use S3 to save the download data and save checkpoint for training for persistent. Setting the s3 location as cache directory is not fast enough. That is why I want to set a work directory for temp data for memory map and only save the final result to s3 cache. ", "You can try setting `HF_DATASETS_DOWNLOADED_DATASETS_PATH` and `HF_DATASETS_EXTRACTED_DATASETS_PATH` to S3, and `HF_DATASETS_CACHE` to your local disk.\r\n\r\nThis way all your downloaded and extracted data are on your mounted S3, but the datasets Arrow files are on your local disk", "If we hope to also persist the Arrow files on the mounted S3 but work with the efficiency of local disk, is there any recommended way to do this, other than copying the Arrow files from local disk to S3?" ]

[]

[ "**NOTE:**\r\n\r\n- Train and Dev: Manually downloaded (auto download is repeatedly giving `ConnectionError` for one of the files), Test: Auto Download\r\n- Dummy test is passing\r\n- The json file has been created with hard-coded paths for the manual downloads _(hardcoding has been removed from the final uploaded script)_\r\n- datasets-cli is still **failing** . It is not picking the right directory for the config. For instance, my folder structure is as below:\r\n ```\r\n ~/Downloads/Data/\r\n |--- English-to-Polish\r\n |--- (corresponding files) \r\n |--- Russian-Polish\r\n |--- (corresponding files) \r\n```\r\n\r\nWhen ru-pl is selected, ideally it has to search in Russian-Polish folder, but it is searching in '/Downloads/Data/' folder and hence getting a FileNotFound error.\r\n\r\nThe command run is \r\n`python datasets-cli test datasets/poleval2019_mt/ --save_infos --all_configs --data_dir ~/Downloads/Data/\r\n`\r\n", "Hi !\r\nThanks for the changes :)\r\n\r\nThe only error left is the dummy data. Since we changed for standard downloads instead of manual downloads its structure changed. Fortunately you can auto-generate the dummy data with this command:\r\n\r\n```\r\ndatasets-cli dummy_data ./datasets/poleval2019_mt --auto_generate --match_text_files \"*\"\r\n```\r\n\r\nCan you regenerate the dummy data using this command please ?", "Thank you for the help @lhoestq !! I was generating the dummy dataset in a wrong way! That _--match_text_files \"*\"_ did the trick! Now all the tests have passed! :-)", "Hi @vrindaprabhu ! Do you still have the Poleval2019 data files somewhere by any chance ? It appears the google drive URLs are not working anymore", "Hi @lhoestq. Just checked. I do not have the backup of the data anywhere. It also appears that PolEval does not repeat its tasks, the data seem to have gone forever. Do you think I should try contacting the organizers for more info?", "We tried already and they don't have the data anymore :(" ]

[]

[ "doing a patch release now :)" ]

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

[]

[]

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

[]

[]

[]

[ "Hi !\r\nLooks like there's a unicode error in the new citation in the miam.py file.\r\nCould you try to fix it ? Not sure from which character it comes from though\r\n\r\nYou can test if it works on your side with\r\n```\r\nRUN_SLOW=1 pytest tests/test_dataset_common.py::LocalDatasetTest::test_load_dataset_all_configs_miam\r\n```", "Unicode error resolved!" ]

[]

[]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "The CI failure is unrelated to your PR - feel free to merge :)", "Haha thanks, you read my mind :)", "<details>\n<summary>Show benchmarks</summary>\n\nPyArrow==6.0.0\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) | 0.008332 / 0.011353 (-0.003021) | 0.004556 / 0.011008 (-0.006452) | 0.102239 / 0.038508 (0.063731) | 0.029332 / 0.023109 (0.006222) | 0.296189 / 0.275898 (0.020291) | 0.355746 / 0.323480 (0.032266) | 0.007705 / 0.007986 (-0.000281) | 0.003488 / 0.004328 (-0.000840) | 0.079142 / 0.004250 (0.074891) | 0.034980 / 0.037052 (-0.002073) | 0.307460 / 0.258489 (0.048971) | 0.345944 / 0.293841 (0.052103) | 0.033815 / 0.128546 (-0.094731) | 0.011603 / 0.075646 (-0.064044) | 0.322097 / 0.419271 (-0.097175) | 0.043753 / 0.043533 (0.000220) | 0.296706 / 0.255139 (0.041567) | 0.323195 / 0.283200 (0.039996) | 0.092295 / 0.141683 (-0.049388) | 1.542556 / 1.452155 (0.090401) | 1.571896 / 1.492716 (0.079180) |\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) | 0.191075 / 0.018006 (0.173069) | 0.407394 / 0.000490 (0.406905) | 0.002033 / 0.000200 (0.001833) | 0.000073 / 0.000054 (0.000018) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.023175 / 0.037411 (-0.014236) | 0.094774 / 0.014526 (0.080248) | 0.105782 / 0.176557 (-0.070775) | 0.146608 / 0.737135 (-0.590528) | 0.107519 / 0.296338 (-0.188819) |\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) | 0.421516 / 0.215209 (0.206306) | 4.201091 / 2.077655 (2.123436) | 1.880285 / 1.504120 (0.376165) | 1.676333 / 1.541195 (0.135139) | 1.734301 / 1.468490 (0.265811) | 0.688504 / 4.584777 (-3.896273) | 3.370289 / 3.745712 (-0.375423) | 3.127661 / 5.269862 (-2.142201) | 1.562570 / 4.565676 (-3.003106) | 0.081687 / 0.424275 (-0.342588) | 0.012334 / 0.007607 (0.004727) | 0.524125 / 0.226044 (0.298080) | 5.245595 / 2.268929 (2.976667) | 2.332622 / 55.444624 (-53.112002) | 1.973212 / 6.876477 (-4.903265) | 2.006507 / 2.142072 (-0.135565) | 0.807126 / 4.805227 (-3.998101) | 0.148254 / 6.500664 (-6.352411) | 0.064240 / 0.075469 (-0.011229) |\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) | 1.206880 / 1.841788 (-0.634907) | 13.854877 / 8.074308 (5.780569) | 13.806772 / 10.191392 (3.615380) | 0.144380 / 0.680424 (-0.536044) | 0.028492 / 0.534201 (-0.505709) | 0.393854 / 0.579283 (-0.185429) | 0.402210 / 0.434364 (-0.032154) | 0.462138 / 0.540337 (-0.078199) | 0.537480 / 1.386936 (-0.849456) |\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) | 0.006692 / 0.011353 (-0.004661) | 0.004529 / 0.011008 (-0.006479) | 0.077925 / 0.038508 (0.039417) | 0.027824 / 0.023109 (0.004715) | 0.342288 / 0.275898 (0.066390) | 0.375071 / 0.323480 (0.051591) | 0.004889 / 0.007986 (-0.003097) | 0.003353 / 0.004328 (-0.000975) | 0.076198 / 0.004250 (0.071947) | 0.037797 / 0.037052 (0.000744) | 0.347834 / 0.258489 (0.089345) | 0.384200 / 0.293841 (0.090359) | 0.032184 / 0.128546 (-0.096362) | 0.011674 / 0.075646 (-0.063972) | 0.086242 / 0.419271 (-0.333029) | 0.044465 / 0.043533 (0.000932) | 0.341712 / 0.255139 (0.086573) | 0.366908 / 0.283200 (0.083709) | 0.091526 / 0.141683 (-0.050156) | 1.495798 / 1.452155 (0.043643) | 1.571700 / 1.492716 (0.078984) |\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) | 0.221962 / 0.018006 (0.203955) | 0.393095 / 0.000490 (0.392605) | 0.000385 / 0.000200 (0.000185) | 0.000074 / 0.000054 (0.000020) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.024365 / 0.037411 (-0.013046) | 0.099278 / 0.014526 (0.084753) | 0.105940 / 0.176557 (-0.070617) | 0.141334 / 0.737135 (-0.595802) | 0.110898 / 0.296338 (-0.185440) |\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) | 0.446150 / 0.215209 (0.230941) | 4.471441 / 2.077655 (2.393786) | 2.124864 / 1.504120 (0.620744) | 1.909950 / 1.541195 (0.368755) | 1.970085 / 1.468490 (0.501595) | 0.706711 / 4.584777 (-3.878066) | 3.380336 / 3.745712 (-0.365376) | 1.866106 / 5.269862 (-3.403756) | 1.160657 / 4.565676 (-3.405019) | 0.082786 / 0.424275 (-0.341489) | 0.012470 / 0.007607 (0.004862) | 0.537620 / 0.226044 (0.311575) | 5.390588 / 2.268929 (3.121659) | 2.539137 / 55.444624 (-52.905488) | 2.191867 / 6.876477 (-4.684610) | 2.236212 / 2.142072 (0.094139) | 0.810756 / 4.805227 (-3.994471) | 0.150933 / 6.500664 (-6.349731) | 0.066141 / 0.075469 (-0.009328) |\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) | 1.271595 / 1.841788 (-0.570193) | 13.840013 / 8.074308 (5.765705) | 13.334443 / 10.191392 (3.143051) | 0.150096 / 0.680424 (-0.530328) | 0.016919 / 0.534201 (-0.517282) | 0.375534 / 0.579283 (-0.203749) | 0.387203 / 0.434364 (-0.047161) | 0.463500 / 0.540337 (-0.076838) | 0.553496 / 1.386936 (-0.833440) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#5f2e47230c13f977bcebdc4380623f59da67a75f \"CML watermark\")\n" ]

[]

[ "Actually `map(batched=True)` can already change the size of the dataset.\r\nIt can accept examples of length `N` and returns a batch of length `M` (can be null or greater than `N`).\r\n\r\nI'll make that explicit in the doc that I'm currently writing.", "You're two steps ahead of me :) In my testing, it also works if `M` < `N`.\r\n\r\nA batched map of different length seems to work if you directly overwrite all of the original keys, but fails if any of the original keys are preserved.\r\n\r\nFor example,\r\n```python\r\n# Create a dummy dataset\r\ndset = load_dataset(\"wikitext\", \"wikitext-2-raw-v1\")[\"test\"]\r\ndset = dset.map(lambda ex: {\"length\": len(ex[\"text\"]), \"foo\": 1})\r\n\r\n# Do an allreduce on each batch, overwriting both keys\r\ndset.map(lambda batch: {\"length\": [sum(batch[\"length\"])], \"foo\": [1]})\r\n# Dataset(schema: {'length': 'int64', 'foo': 'int64'}, num_rows: 5)\r\n\r\n# Now attempt an allreduce without touching the `foo` key\r\ndset.map(lambda batch: {\"length\": [sum(batch[\"length\"])]})\r\n# This fails with the error message below\r\n```\r\n\r\n```bash\r\n File \"/path/to/nlp/src/nlp/arrow_dataset.py\", line 728, in map\r\n arrow_schema = pa.Table.from_pydict(test_output).schema\r\n File \"pyarrow/io.pxi\", line 1532, in pyarrow.lib.Codec.detect\r\n File \"pyarrow/table.pxi\", line 1503, in pyarrow.lib.Table.from_arrays\r\n File \"pyarrow/public-api.pxi\", line 390, in pyarrow.lib.pyarrow_wrap_table\r\n File \"pyarrow/error.pxi\", line 85, in pyarrow.lib.check_status\r\npyarrow.lib.ArrowInvalid: Column 1 named foo expected length 1 but got length 2\r\n```\r\n\r\nAdding the `remove_columns=[\"length\", \"foo\"]` argument to `map()` solves the issue. Leaving the above error for future visitors. Perfect, thank you!" ]

[]

[ "Trailing whitespace was removed. So more changes in diff than just this fix." ]

[]

[ "Google colab has an old version of Apache Arrow built-in.\r\nBe sure you execute the \"pip install\" cell and restart the notebook environment if the colab asks for it.", "Thanks for reply, worked fine!\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" } ]

[ "Hi ! You can download the `oscar.py` file from this repository at `/datasets/oscar/oscar.py`.\r\n\r\nThen you can load the dataset by passing the local path to `oscar.py` to `load_dataset`:\r\n```python\r\nload_dataset(\"path/to/oscar.py\", \"unshuffled_deduplicated_it\")\r\n```", "it works,but another error occurs.\r\n```\r\nConnectionError: Couldn't reach https://s3.amazonaws.com/datasets.huggingface.co/oscar/1.0/unshuffled/deduplicated/it/it_sha256.txt (SSLError(MaxRetryError(\"HTTPSConnectionPool(host='s3.amazonaws.com', port=443): Max retries exceeded with url: /datasets.huggingface.co/oscar/1.0/unshuffled/deduplicated/it/it_sha256.txt (Caused by SSLError(SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1129)')))\")))\r\n```\r\nI can access `https://s3.amazonaws.com/datasets.huggingface.co/oscar/1.0/unshuffled/deduplicated/it/it_sha256.txt` and `https://aws.amazon.com/cn/s3/` directly, so why it reports a SSLError, should I need tomodify the host file？", "Could it be an issue with your python environment or your version of OpenSSL ?", "you are so wise!\r\nit report [ConnectionError] in python 3.9.7\r\nand works well in python 3.8.12\r\n\r\nI need you help again: how can I specify the path for download files?\r\nthe data is too large and my C hardware is not enough", "Cool ! And you can specify the path for download files with to the `cache_dir` parameter:\r\n```python\r\nfrom datasets import load_dataset\r\ndataset = load_dataset('oscar', 'unshuffled_deduplicated_it', cache_dir='path/to/directory')", "It takes me some days to download data completely, Despise sometimes it occurs again, change py version is feasible way to avoid this ConnectionEror.\r\nparameter `cache_dir` works well, thanks for your kindness again!" ]

[]

[]

[]

[]

[]

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

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Thank you so much @[lhoestq](https://github.com/lhoestq) for the time you take to your review the PR!" ]

[ { "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" } ]

[ "@lhoestq I would like to draw your attention to the proposed API by @lewtun, using `data_dir` to pass the ZIP URL.\r\n\r\nI'm not totally convinced with this... What do you think?\r\n\r\nMaybe we could discuss other approaches...\r\n\r\nOne brainstorming idea: what about using URL chaining with the hop operator in `data_files`?", "`data_dir` is currently exclusively used for manually downloaded data.\r\n\r\nMaybe we can have an API that only uses data_files as you are suggesting, using URL chaining ?\r\n```python\r\nfrom datasets import load_dataset\r\n\r\nurl = \"https://domain.org/filename.zip\"\r\ndata_files = {\"train\": \"zip://train_filename.csv::\" + url, \"test\": \"zip://test_filename.csv::\" + url}\r\ndataset = load_dataset(\"csv\", data_files=data_files)\r\n```\r\n\r\nURL chaining is used by `fsspec` to get access to files in nested filesystems of any kind. Since `fsspec` is being used by `pandas`, `dask` and also extensively by `datasets` I think it would be nice to use it here too", "URL chaining sounds super nice to me! And it's also a nice way to leverage the same concepts we currently have in the docs around `fsspec` :)" ]

[ { "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 think https://github.com/fsspec/universal_pathlib is the project you are looking for.\r\n\r\n`xPath` has the methods often used in dataset scripts, and `mkdir` is not one of them (`dl_manager`'s role is to \"interact\" with the file system, so using `mkdir` is discouraged).", "Right is there a difference between UPath and xPath? Typically is xPath less well implemented compared to Upath, ie missing some implementations of some methods? Or are there methods in xPath that are not implemented with UPath?", "`xPath` is an internal component (it doesn't have a leading underscore in the name, but it should) not meant to be used outside of `datasets`, and it's only tested on HTTP URLs, not S3.\r\n\r\n", "Okay I understand that xPath won't support my usecase. What I was perhaps getting to is why not use UPath in `datasets` instead of `xPath` if UPath seems to have strictly more robust implementations.", "It seems like `universal_pathlib` does not support `fsspec` URL chaining (`::` is the chaining symbol) and \"compression\" filesystems (e.g., `zip`), but this is what we need to access and stream files from within an archive (e.g., we want to stream URLs such as this one: `zip://data.parquet::https://www.dummyurl.com/archive.zip`)" ]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_6041). 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) | 0.007587 / 0.011353 (-0.003766) | 0.004469 / 0.011008 (-0.006540) | 0.098028 / 0.038508 (0.059520) | 0.086378 / 0.023109 (0.063269) | 0.412290 / 0.275898 (0.136392) | 0.449912 / 0.323480 (0.126432) | 0.004769 / 0.007986 (-0.003217) | 0.003708 / 0.004328 (-0.000621) | 0.075541 / 0.004250 (0.071290) | 0.063821 / 0.037052 (0.026768) | 0.417213 / 0.258489 (0.158724) | 0.471954 / 0.293841 (0.178113) | 0.036243 / 0.128546 (-0.092303) | 0.009540 / 0.075646 (-0.066106) | 0.339043 / 0.419271 (-0.080228) | 0.061853 / 0.043533 (0.018320) | 0.418510 / 0.255139 (0.163371) | 0.462372 / 0.283200 (0.179173) | 0.027328 / 0.141683 (-0.114355) | 1.745114 / 1.452155 (0.292959) | 1.879839 / 1.492716 (0.387123) |\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) | 0.211042 / 0.018006 (0.193035) | 0.512865 / 0.000490 (0.512375) | 0.008744 / 0.000200 (0.008544) | 0.000121 / 0.000054 (0.000066) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.032493 / 0.037411 (-0.004918) | 0.096472 / 0.014526 (0.081946) | 0.110340 / 0.176557 (-0.066216) | 0.183195 / 0.737135 (-0.553940) | 0.112829 / 0.296338 (-0.183510) |\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) | 0.478040 / 0.215209 (0.262830) | 4.743776 / 2.077655 (2.666121) | 2.389770 / 1.504120 (0.885650) | 2.168468 / 1.541195 (0.627274) | 2.238154 / 1.468490 (0.769663) | 0.572308 / 4.584777 (-4.012469) | 4.154783 / 3.745712 (0.409071) | 3.771509 / 5.269862 (-1.498353) | 2.384828 / 4.565676 (-2.180848) | 0.068122 / 0.424275 (-0.356153) | 0.008573 / 0.007607 (0.000965) | 0.560300 / 0.226044 (0.334256) | 5.591163 / 2.268929 (3.322235) | 2.929660 / 55.444624 (-52.514965) | 2.517721 / 6.876477 (-4.358756) | 2.762285 / 2.142072 (0.620213) | 0.687193 / 4.805227 (-4.118034) | 0.157839 / 6.500664 (-6.342825) | 0.071862 / 0.075469 (-0.003607) |\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) | 1.484788 / 1.841788 (-0.357000) | 21.696071 / 8.074308 (13.621763) | 15.476166 / 10.191392 (5.284774) | 0.185034 / 0.680424 (-0.495390) | 0.021181 / 0.534201 (-0.513020) | 0.463324 / 0.579283 (-0.115959) | 0.502455 / 0.434364 (0.068091) | 0.559880 / 0.540337 (0.019543) | 0.767281 / 1.386936 (-0.619655) |\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) | 0.007572 / 0.011353 (-0.003781) | 0.004331 / 0.011008 (-0.006677) | 0.075023 / 0.038508 (0.036515) | 0.085474 / 0.023109 (0.062365) | 0.464900 / 0.275898 (0.189002) | 0.503348 / 0.323480 (0.179868) | 0.006885 / 0.007986 (-0.001101) | 0.003647 / 0.004328 (-0.000681) | 0.074874 / 0.004250 (0.070623) | 0.071076 / 0.037052 (0.034024) | 0.465495 / 0.258489 (0.207006) | 0.506418 / 0.293841 (0.212577) | 0.038900 / 0.128546 (-0.089647) | 0.009467 / 0.075646 (-0.066180) | 0.082547 / 0.419271 (-0.336724) | 0.058457 / 0.043533 (0.014924) | 0.459114 / 0.255139 (0.203975) | 0.484872 / 0.283200 (0.201673) | 0.027443 / 0.141683 (-0.114240) | 1.713996 / 1.452155 (0.261841) | 1.893639 / 1.492716 (0.400922) |\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) | 0.248693 / 0.018006 (0.230687) | 0.488805 / 0.000490 (0.488315) | 0.000421 / 0.000200 (0.000221) | 0.000067 / 0.000054 (0.000012) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.034886 / 0.037411 (-0.002525) | 0.103215 / 0.014526 (0.088689) | 0.116422 / 0.176557 (-0.060134) | 0.182789 / 0.737135 (-0.554346) | 0.117788 / 0.296338 (-0.178550) |\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) | 0.482782 / 0.215209 (0.267573) | 4.802895 / 2.077655 (2.725241) | 2.489823 / 1.504120 (0.985703) | 2.324005 / 1.541195 (0.782810) | 2.457674 / 1.468490 (0.989184) | 0.566980 / 4.584777 (-4.017797) | 4.117359 / 3.745712 (0.371647) | 3.841180 / 5.269862 (-1.428681) | 2.322410 / 4.565676 (-2.243266) | 0.066367 / 0.424275 (-0.357908) | 0.008501 / 0.007607 (0.000894) | 0.561453 / 0.226044 (0.335408) | 5.694861 / 2.268929 (3.425932) | 3.129829 / 55.444624 (-52.314796) | 2.647375 / 6.876477 (-4.229102) | 2.673071 / 2.142072 (0.530998) | 0.676120 / 4.805227 (-4.129108) | 0.153483 / 6.500664 (-6.347181) | 0.070797 / 0.075469 (-0.004672) |\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) | 1.575697 / 1.841788 (-0.266091) | 22.447462 / 8.074308 (14.373154) | 15.964906 / 10.191392 (5.773514) | 0.218343 / 0.680424 (-0.462081) | 0.021051 / 0.534201 (-0.513150) | 0.466079 / 0.579283 (-0.113204) | 0.493190 / 0.434364 (0.058826) | 0.565929 / 0.540337 (0.025592) | 0.768638 / 1.386936 (-0.618298) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#104bafffef7ddc775ec2d0b10b2b262466041eb7 \"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) | 0.006268 / 0.011353 (-0.005085) | 0.003715 / 0.011008 (-0.007293) | 0.080628 / 0.038508 (0.042120) | 0.070294 / 0.023109 (0.047185) | 0.404749 / 0.275898 (0.128851) | 0.434130 / 0.323480 (0.110650) | 0.005533 / 0.007986 (-0.002452) | 0.002980 / 0.004328 (-0.001349) | 0.063016 / 0.004250 (0.058766) | 0.051667 / 0.037052 (0.014615) | 0.403859 / 0.258489 (0.145370) | 0.437913 / 0.293841 (0.144073) | 0.027518 / 0.128546 (-0.101029) | 0.007991 / 0.075646 (-0.067655) | 0.260723 / 0.419271 (-0.158548) | 0.046580 / 0.043533 (0.003047) | 0.405453 / 0.255139 (0.150314) | 0.428390 / 0.283200 (0.145190) | 0.022774 / 0.141683 (-0.118909) | 1.488204 / 1.452155 (0.036049) | 1.536557 / 1.492716 (0.043841) |\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) | 0.185864 / 0.018006 (0.167858) | 0.431388 / 0.000490 (0.430898) | 0.003743 / 0.000200 (0.003543) | 0.000065 / 0.000054 (0.000011) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.024062 / 0.037411 (-0.013350) | 0.075749 / 0.014526 (0.061224) | 0.083519 / 0.176557 (-0.093037) | 0.147965 / 0.737135 (-0.589170) | 0.085635 / 0.296338 (-0.210703) |\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) | 0.400455 / 0.215209 (0.185246) | 4.084294 / 2.077655 (2.006640) | 1.928795 / 1.504120 (0.424675) | 1.743205 / 1.541195 (0.202010) | 1.811233 / 1.468490 (0.342743) | 0.504976 / 4.584777 (-4.079801) | 3.073134 / 3.745712 (-0.672578) | 2.816357 / 5.269862 (-2.453505) | 1.857462 / 4.565676 (-2.708214) | 0.058329 / 0.424275 (-0.365946) | 0.006850 / 0.007607 (-0.000757) | 0.466017 / 0.226044 (0.239973) | 4.660158 / 2.268929 (2.391230) | 2.396614 / 55.444624 (-53.048010) | 2.007491 / 6.876477 (-4.868986) | 2.206997 / 2.142072 (0.064925) | 0.592233 / 4.805227 (-4.212994) | 0.125364 / 6.500664 (-6.375300) | 0.061166 / 0.075469 (-0.014303) |\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) | 1.290148 / 1.841788 (-0.551640) | 18.317462 / 8.074308 (10.243154) | 13.465142 / 10.191392 (3.273750) | 0.149696 / 0.680424 (-0.530728) | 0.017120 / 0.534201 (-0.517081) | 0.334818 / 0.579283 (-0.244465) | 0.363976 / 0.434364 (-0.070388) | 0.388271 / 0.540337 (-0.152066) | 0.542383 / 1.386936 (-0.844553) |\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) | 0.006029 / 0.011353 (-0.005324) | 0.003656 / 0.011008 (-0.007352) | 0.063518 / 0.038508 (0.025010) | 0.058214 / 0.023109 (0.035105) | 0.435987 / 0.275898 (0.160089) | 0.442769 / 0.323480 (0.119289) | 0.004675 / 0.007986 (-0.003310) | 0.002911 / 0.004328 (-0.001418) | 0.063020 / 0.004250 (0.058769) | 0.049422 / 0.037052 (0.012369) | 0.435521 / 0.258489 (0.177032) | 0.478251 / 0.293841 (0.184411) | 0.027294 / 0.128546 (-0.101252) | 0.008073 / 0.075646 (-0.067574) | 0.068397 / 0.419271 (-0.350875) | 0.044796 / 0.043533 (0.001263) | 0.416646 / 0.255139 (0.161507) | 0.435021 / 0.283200 (0.151821) | 0.024686 / 0.141683 (-0.116997) | 1.495650 / 1.452155 (0.043496) | 1.495846 / 1.492716 (0.003130) |\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) | 0.211205 / 0.018006 (0.193199) | 0.414497 / 0.000490 (0.414007) | 0.001704 / 0.000200 (0.001504) | 0.000073 / 0.000054 (0.000019) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.025237 / 0.037411 (-0.012174) | 0.077291 / 0.014526 (0.062765) | 0.085736 / 0.176557 (-0.090821) | 0.141059 / 0.737135 (-0.596076) | 0.087620 / 0.296338 (-0.208719) |\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) | 0.421995 / 0.215209 (0.206786) | 4.158503 / 2.077655 (2.080849) | 2.313598 / 1.504120 (0.809479) | 2.183553 / 1.541195 (0.642359) | 2.279656 / 1.468490 (0.811166) | 0.500146 / 4.584777 (-4.084631) | 3.092654 / 3.745712 (-0.653059) | 4.371616 / 5.269862 (-0.898245) | 2.605096 / 4.565676 (-1.960581) | 0.057658 / 0.424275 (-0.366617) | 0.006574 / 0.007607 (-0.001033) | 0.491455 / 0.226044 (0.265411) | 4.926730 / 2.268929 (2.657801) | 2.635749 / 55.444624 (-52.808875) | 2.255780 / 6.876477 (-4.620697) | 2.305547 / 2.142072 (0.163474) | 0.589027 / 4.805227 (-4.216200) | 0.126229 / 6.500664 (-6.374435) | 0.063268 / 0.075469 (-0.012201) |\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) | 1.299102 / 1.841788 (-0.542686) | 18.547417 / 8.074308 (10.473109) | 13.860030 / 10.191392 (3.668638) | 0.145482 / 0.680424 (-0.534942) | 0.016543 / 0.534201 (-0.517658) | 0.330788 / 0.579283 (-0.248496) | 0.362020 / 0.434364 (-0.072344) | 0.380635 / 0.540337 (-0.159703) | 0.517375 / 1.386936 (-0.869561) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#bf602e0193baca21e283babbac9622ae36d1e6b6 \"CML watermark\")\n" ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "This is ready for review cc @mariosasko :) please let me know what you think !" ]

[ { "color": "a2eeef", "default": true, "description": "New feature or request", "id": 1935892871, "name": "enhancement", "node_id": "MDU6TGFiZWwxOTM1ODkyODcx", "url": "https://api.github.com/repos/huggingface/datasets/labels/enhancement" }, { "color": "fef2c0", "default": false, "description": "", "id": 3287858981, "name": "streaming", "node_id": "MDU6TGFiZWwzMjg3ODU4OTgx", "url": "https://api.github.com/repos/huggingface/datasets/labels/streaming" } ]

[ "I think `stream` could be misleading since the data is not being streamed from remote endpoints (one could think that's the case when they see `load_dataset` followed by `stream`). Hence, I prefer the second option.\r\n\r\nPS: When we resolve https://github.com/huggingface/datasets/issues/4542, we could add `as_tf_dataset` to the API for consistency and deprecate `to_tf_dataset`." ]

[]

[ "Thank you! Sorry I forgot this one @albertvillanova" ]

[]

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

[ { "color": "2edb81", "default": false, "description": "A bug in a dataset script provided in the library", "id": 2067388877, "name": "dataset bug", "node_id": "MDU6TGFiZWwyMDY3Mzg4ODc3", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20bug" } ]

[ "I just ran it on colab and got this\r\n```\r\n[{'expected': SplitInfo(name='train', num_bytes=610252351, num_examples=532812,\r\ndataset_name='blog_authorship_corpus'), 'recorded': SplitInfo(name='train',\r\nnum_bytes=611607465, num_examples=533285, dataset_name='blog_authorship_corpus')},\r\n{'expected': SplitInfo(name='validation', num_bytes=37500394, num_examples=31277,\r\ndataset_name='blog_authorship_corpus'), 'recorded': SplitInfo(name='validation',\r\nnum_bytes=35652716, num_examples=30804, dataset_name='blog_authorship_corpus')}]\r\n```\r\nwhich is different from the `dataset_infos.json` and also different from yours.\r\n\r\nIt looks like the script for generating examples is not consistent", "The files provided by the authors are corrupted and the script seems to ignore the xml files that can't be decoded (it does `try:... except UnicodeDecodeError`). Maybe depending of the environment some files can be opened and some others don't but not sure why", "Feel free to do `ignore_verifications=True` for now... The verifications only include a check on the checksums of the downloaded files, and a check on the number of examples in each splits.", "I'm getting this same issue when loading the `imdb` corpus via `dataset = load_dataset(\"imdb\")`. When I try `ignore_verifications=True`, no examples are read into the `train` portion of the dataset. ", "> I'm getting this same issue when loading the `imdb` corpus via `dataset = load_dataset(\"imdb\")`. When I try `ignore_verifications=True`, no examples are read into the `train` portion of the dataset.\r\n\r\nWhen the checksums don't match, it may mean that the file you downloaded is corrupted. In this case you can try to load the dataset again `load_dataset(\"imdb\", download_mode=\"force_redownload\")`\r\n\r\nAlso I just checked on my side and it worked fine:\r\n\r\n```python\r\nfrom datasets import load_dataset\r\n\r\ndataset = load_dataset(\"imdb\")\r\nprint(len(dataset[\"train\"]))\r\n# 25000\r\n```\r\n\r\nLet me know if redownloading fixes your issue @EmilyAlsentzer .\r\nIf not, feel free to open a separate issue.", "It doesn't seem to fix the problem. I'll open a separate issue. Thanks. ", "I wasn't aware of the \"force_redownload\" option and manually removed the '/home/me/.cache/huggingface/datasets/' dir, this worked for me (dataset 'cnn_dailymail')", "Yes I think this might not be documented well enough. Let’s add it to the doc @lhoestq @SBrandeis.\r\nAnd everything on how to control the cache behavior better (removing, overriding, changing the path, etc)", "Already fixed:\r\n```python\r\nIn [1]: from datasets import load_dataset\r\n\r\nIn [2]: ds = load_dataset(\"blog_authorship_corpus\")\r\n\r\nIn [3]: ds\r\nOut[3]: \r\nDatasetDict({\r\n train: Dataset({\r\n features: ['text', 'date', 'gender', 'age', 'horoscope', 'job'],\r\n num_rows: 689793\r\n })\r\n validation: Dataset({\r\n features: ['text', 'date', 'gender', 'age', 'horoscope', 'job'],\r\n num_rows: 37919\r\n })\r\n})\r\n", "In my case, I had to remove the cache datasets directory completely as @putssander suggested, the download_mode='forced_redownload' was insufficient.\r\n\r\nI had a private repository with data files that I loaded with a loading script. It was working fine until I pushed a new version of the data files and then the NonMatchingSplitsSizesError was raised.\r\n" ]

[ { "color": "25b21e", "default": false, "description": "A bug in a metric script", "id": 2067393914, "name": "metric bug", "node_id": "MDU6TGFiZWwyMDY3MzkzOTE0", "url": "https://api.github.com/repos/huggingface/datasets/labels/metric%20bug" } ]

[ "Can you share your data files or a minimally reproducible example?", "Sure, [here is a Colab notebook](https://colab.research.google.com/drive/1uiS89fnHMG7HV_cYxp3r-_LqJQvNNKs9?usp=sharing) reproducing the error.\r\n\r\n> ArrowInvalid: Column 1 named references expected length 36 but got length 56", "This is because `add` takes as input a batch of elements and you provided only one. I think we should have `add` for one prediction/reference and `add_batch` for a batch of predictions/references. This would make it more coherent with the way we use Arrow.\r\n\r\nLet me do this change", "Thanks for noticing though. I was mainly used to do `.compute` directly ^^", "Thanks @lhoestq it works :)" ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "I think you can revert the last commit - it should fail if data_files={} IMO", "The validation of non-empty data_files is addressed in this PR:\r\n- #5802", "<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) | 0.008622 / 0.011353 (-0.002730) | 0.005970 / 0.011008 (-0.005038) | 0.117797 / 0.038508 (0.079289) | 0.040955 / 0.023109 (0.017846) | 0.419538 / 0.275898 (0.143640) | 0.455816 / 0.323480 (0.132336) | 0.006481 / 0.007986 (-0.001505) | 0.004507 / 0.004328 (0.000178) | 0.089073 / 0.004250 (0.084822) | 0.052389 / 0.037052 (0.015337) | 0.420053 / 0.258489 (0.161564) | 0.466886 / 0.293841 (0.173045) | 0.042660 / 0.128546 (-0.085886) | 0.014673 / 0.075646 (-0.060973) | 0.411229 / 0.419271 (-0.008042) | 0.076993 / 0.043533 (0.033460) | 0.431693 / 0.255139 (0.176554) | 0.446283 / 0.283200 (0.163084) | 0.131408 / 0.141683 (-0.010275) | 1.820339 / 1.452155 (0.368184) | 1.952946 / 1.492716 (0.460230) |\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) | 0.246543 / 0.018006 (0.228537) | 0.489806 / 0.000490 (0.489317) | 0.013999 / 0.000200 (0.013800) | 0.000323 / 0.000054 (0.000269) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.032541 / 0.037411 (-0.004870) | 0.130569 / 0.014526 (0.116043) | 0.139630 / 0.176557 (-0.036926) | 0.217018 / 0.737135 (-0.520118) | 0.147914 / 0.296338 (-0.148425) |\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) | 0.494767 / 0.215209 (0.279558) | 4.949313 / 2.077655 (2.871658) | 2.277023 / 1.504120 (0.772903) | 2.036677 / 1.541195 (0.495482) | 2.064461 / 1.468490 (0.595970) | 0.842484 / 4.584777 (-3.742293) | 4.720646 / 3.745712 (0.974934) | 4.025673 / 5.269862 (-1.244189) | 2.198606 / 4.565676 (-2.367070) | 0.103042 / 0.424275 (-0.321233) | 0.014794 / 0.007607 (0.007187) | 0.617867 / 0.226044 (0.391822) | 6.197146 / 2.268929 (3.928218) | 2.804927 / 55.444624 (-52.639697) | 2.426420 / 6.876477 (-4.450057) | 2.515182 / 2.142072 (0.373109) | 1.008098 / 4.805227 (-3.797129) | 0.204982 / 6.500664 (-6.295682) | 0.078643 / 0.075469 (0.003174) |\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) | 1.490790 / 1.841788 (-0.350997) | 17.268042 / 8.074308 (9.193734) | 17.129647 / 10.191392 (6.938255) | 0.170351 / 0.680424 (-0.510073) | 0.021317 / 0.534201 (-0.512884) | 0.517068 / 0.579283 (-0.062215) | 0.500200 / 0.434364 (0.065836) | 0.641974 / 0.540337 (0.101637) | 0.763984 / 1.386936 (-0.622952) |\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) | 0.008358 / 0.011353 (-0.002995) | 0.005710 / 0.011008 (-0.005298) | 0.091077 / 0.038508 (0.052569) | 0.040413 / 0.023109 (0.017303) | 0.416634 / 0.275898 (0.140736) | 0.451122 / 0.323480 (0.127642) | 0.006417 / 0.007986 (-0.001569) | 0.004360 / 0.004328 (0.000032) | 0.089543 / 0.004250 (0.085292) | 0.051137 / 0.037052 (0.014085) | 0.420228 / 0.258489 (0.161739) | 0.458649 / 0.293841 (0.164808) | 0.041828 / 0.128546 (-0.086718) | 0.014268 / 0.075646 (-0.061379) | 0.105301 / 0.419271 (-0.313970) | 0.058931 / 0.043533 (0.015398) | 0.413445 / 0.255139 (0.158306) | 0.443882 / 0.283200 (0.160682) | 0.124946 / 0.141683 (-0.016737) | 1.842259 / 1.452155 (0.390104) | 1.948162 / 1.492716 (0.455445) |\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) | 0.235799 / 0.018006 (0.217792) | 0.487667 / 0.000490 (0.487177) | 0.001112 / 0.000200 (0.000912) | 0.000094 / 0.000054 (0.000039) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.034233 / 0.037411 (-0.003178) | 0.136593 / 0.014526 (0.122068) | 0.145598 / 0.176557 (-0.030959) | 0.206545 / 0.737135 (-0.530590) | 0.150781 / 0.296338 (-0.145558) |\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) | 0.522345 / 0.215209 (0.307136) | 5.192092 / 2.077655 (3.114438) | 2.543182 / 1.504120 (1.039062) | 2.285212 / 1.541195 (0.744018) | 2.312803 / 1.468490 (0.844313) | 0.859334 / 4.584777 (-3.725443) | 4.620235 / 3.745712 (0.874523) | 3.964060 / 5.269862 (-1.305802) | 2.046347 / 4.565676 (-2.519330) | 0.105284 / 0.424275 (-0.318991) | 0.015051 / 0.007607 (0.007444) | 0.646530 / 0.226044 (0.420485) | 6.386396 / 2.268929 (4.117467) | 3.131833 / 55.444624 (-52.312791) | 2.761898 / 6.876477 (-4.114579) | 2.833216 / 2.142072 (0.691143) | 1.026024 / 4.805227 (-3.779204) | 0.206776 / 6.500664 (-6.293888) | 0.078845 / 0.075469 (0.003376) |\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) | 1.580851 / 1.841788 (-0.260937) | 17.826213 / 8.074308 (9.751905) | 16.929460 / 10.191392 (6.738068) | 0.232483 / 0.680424 (-0.447941) | 0.021123 / 0.534201 (-0.513078) | 0.522196 / 0.579283 (-0.057087) | 0.503495 / 0.434364 (0.069131) | 0.622777 / 0.540337 (0.082440) | 0.753272 / 1.386936 (-0.633664) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#3f9dfbd93707665132abc862b14bb9b50597b739 \"CML watermark\")\n" ]

[]

[]

[]

[]

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

[ "cc @albertvillanova the author would like to add this dataset to the hub: https://github.com/microsoft/table-transformer/issues/68#issuecomment-1319114621. Could you help him out?" ]

[]

[ "Also fix #3125." ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "I guess the report link is this instead: https://huggingface.co/datasets/xtreme/discussions/1" ]

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/datasets/pr_3843). All of your documentation changes will be reflected on that endpoint.", "Cool ! Looks like it breaks `test_streaming_gg_drive_gzipped` for some reason..." ]

[ { "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 @BramVanroy!\r\n\r\nGood question. This project relies on PyArrow (tables) to store data too big to fit in RAM. In the case of metrics, this means that the number of predictions and references has to match to form a table.\r\n\r\nThat's why your example throws an error even though it matches the schema:\r\n```python\r\nrefs = [\r\n ['The dog bit the man.', 'It was not unexpected.', 'The man bit him first.'],\r\n ['The dog had bit the man.', 'No one was surprised.', 'The man had bitten the dog.'],\r\n] # len(refs) = 2\r\n\r\nhyps = ['The dog bit the man.', \"It wasn't surprising.\", 'The man had just bitten him.'] # len(hyps) = 3\r\n```\r\n\r\nInstead, it should be:\r\n```python\r\nrefs = [\r\n ['The dog bit the man.', 'The dog had bit the man.'],\r\n ['It was not unexpected.', 'No one was surprised.'],\r\n ['The man bit him first.', 'The man had bitten the dog.'], \r\n] # len(refs) = 3\r\n\r\nhyps = ['The dog bit the man.', \"It wasn't surprising.\", 'The man had just bitten him.'] # len(hyps) = 3\r\n```\r\n\r\nHowever, `sacreblue` works with the format that's described in your example, hence this part:\r\nhttps://github.com/huggingface/datasets/blob/87c71b9c29a40958973004910f97e4892559dfed/metrics/sacrebleu/sacrebleu.py#L94-L99\r\n\r\nHope you get an idea!", "Thanks, that makes sense. It is a bit unfortunate because it may be confusing to users since the input format is suddenly different than what they may expect from the underlying library/metric. But it is understandable due to how `datasets` works!" ]

[ { "color": "ffffff", "default": true, "description": "This will not be worked on", "id": 1935892913, "name": "wontfix", "node_id": "MDU6TGFiZWwxOTM1ODkyOTEz", "url": "https://api.github.com/repos/huggingface/datasets/labels/wontfix" }, { "color": "2edb81", "default": false, "description": "A bug in a dataset script provided in the library", "id": 2067388877, "name": "dataset bug", "node_id": "MDU6TGFiZWwyMDY3Mzg4ODc3", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20bug" } ]

[ "I have encountered the same error with `v1.0.1` and `v1.0.2` on both Windows and Linux environments. However, cloning the repo and using the path to the dataset's root directory worked for me. Even after having the dataset cached - passing the path is the only way (for now) to load the dataset.\r\n\r\n```python\r\nfrom datasets import load_dataset\r\n\r\ndataset = load_dataset(\"squad\") # Works\r\ndataset = load_dataset(\"code_search_net\", \"python\") # Error\r\ndataset = load_dataset(\"covid_qa_deepset\") # Error\r\n\r\npath = \"/huggingface/datasets/datasets/{}/\"\r\ndataset = load_dataset(path.format(\"code_search_net\"), \"python\") # Works\r\ndataset = load_dataset(path.format(\"covid_qa_deepset\")) # Works\r\n```\r\n\r\n", "Hi @mrm8488 and @amoux!\r\n The datasets you are trying to load have been added to the library during the community sprint for v2 last month. They will be available with the v2 release!\r\nFor now, there are still a couple of solutions to load the datasets:\r\n1. As suggested by @amoux, you can clone the git repo and pass the local path to the script\r\n2. You can also install the latest (master) version of `datasets` using pip: `pip install git+https://github.com/huggingface/datasets.git@master`", "If you don't want to clone entire `datasets` repo, just download the `muchocine` directory and pass the local path to the directory. Cheers!", "Muchocine was added recently, that's why it wasn't available yet.\r\n\r\nTo load it you can just update `datasets`\r\n```\r\npip install --upgrade datasets\r\n```\r\n\r\nand then you can load `muchocine` with\r\n\r\n```python\r\nfrom datasets import load_dataset\r\n\r\ndataset = load_dataset(\"muchocine\", split=\"train\")\r\n```", "Thanks @lhoestq " ]

[ { "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 @nelson-liu,\r\n\r\nYou can pass the parameter `features` to `load_dataset`: https://huggingface.co/docs/datasets/_modules/datasets/load.html#load_dataset\r\n\r\nIf you look at the code of the MNLI script you referred in your question (https://github.com/huggingface/datasets/blob/master/datasets/multi_nli/multi_nli.py#L62-L77), you can see how the Features were originally specified. \r\n\r\nFeel free to use it as a template, customize it and pass it to `load_dataset` using the parameter `features`.", "ah got it, thanks!" ]

[]

[ "IMO this PR is ready for review. I do not know why tests fail...", "The CI fail is unrelated to this PR, and it has been fixed on master, merging :)", "> I just have one comment: we use rouge, not rogue :p\r\n\r\nOops!", "rebased on master" ]

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Actually the CI fails because of this\r\n\r\n\r\nIt looks like we can't have a proper way to test google drive in the CI right now. Though it seems to work locally if you're not banned. I think I'll just disable those tests for now", "this fix will not be included?", "No we can't do anything except stop using google drive when possible" ]

[]

[ "Nice catch @donhuvy, that's because some models don't need the `token_type_ids`, as in this case, as the example is using `distilbert-base-cased`, and according to the DistilBert documentation at https://huggingface.co/transformers/v3.0.2/model_doc/distilbert.html, `DistilBert doesn’t have token_type_ids, you don’t need to indicate which token belongs to which segment. Just separate your segments with the separation token tokenizer.sep_token (or [SEP])`. `token_type_ids` are neither required in some other well known models such as RoBERTa. \r\n\r\nHere the issue comes due to a mismatch between the tokenizer and the model, as the Colab is using a BERT tokenizer (`bert-base-cased`), while the model is a DistilBERT (`distilbert-base-cased`), so aligning the tokenizer and the model solves it!", "#self-assign", "@donhuvy I've created https://github.com/huggingface/datasets/pull/5902 to solve it! 🤗", "This has been addressed in #5902.\r\n\r\nThe Quicktour notebook is deprecated now - please use the notebook version of the [Quickstart doc page](https://huggingface.co/docs/datasets/main/en/quickstart) instead (\"Open in Colab\" button)." ]

[ { "color": "ffffff", "default": true, "description": "This will not be worked on", "id": 1935892913, "name": "wontfix", "node_id": "MDU6TGFiZWwxOTM1ODkyOTEz", "url": "https://api.github.com/repos/huggingface/datasets/labels/wontfix" } ]

[ "How is it possible to get an infinite loop ? Can you add more details ?", "Yes, in Windows, if the filename is too long, a `FileNotFoundError` is raised. The exception should be raised in this case. Otherwise, we get into an infinite loop.\r\n\r\nIf other process has the file locked, then `PermissionError` is raised. In this case, `pass` is OK.", "Note that the filelock module comes from this project that hasn't changed in years - while still being used by ten of thousands of projects:\r\nhttps://github.com/benediktschmitt/py-filelock\r\n\r\nUnless we have proper tests for this, I wouldn't recommend to change it", "I'm pretty sure many things from the library could break for windows users that haven't disabled the max path length limit.\r\nMaybe it would be simpler to simply raise an error on startup. For exampe, for windows users the error could ask them to disable the limit if it's not been disabled yet ?" ]

[]

[]

[ { "color": "2edb81", "default": false, "description": "A bug in a dataset script provided in the library", "id": 2067388877, "name": "dataset bug", "node_id": "MDU6TGFiZWwyMDY3Mzg4ODc3", "url": "https://api.github.com/repos/huggingface/datasets/labels/dataset%20bug" } ]

[ "If you need an example of a small audio datasets, I just created few hours ago a speech dataset with only 300MB of compressed audio files https://huggingface.co/datasets/indonesian-nlp/librivox-indonesia. It works also with streaming (@albertvillanova helped me adding this functionality) :-)", "@cahya-wirawan omg this is awesome!! thank you! ", "We have contacted the authors to ask them." ]

[]

[ "Hi! I cannot reproduce this error on my machine.\r\n\r\nThe raised error could mean that one of the downloaded files is corrupted. To verify this is not the case, you can run `load_dataset` as follows:\r\n```python\r\ntrain_dataset = load_dataset('xtreme', 'udpos.English', split=\"train\", cache_dir=args.cache_dir, download_mode=\"force_redownload\", verification_mode=\"all_checks\")\r\n```", "Hi! Apologies for the delayed response! I tried the above and it doesn't solve the issue. Actually, the dataset gets downloaded most times, but sometimes this error occurs (at random afaik). Is it possible that there is a server issue for this particular dataset? I am able to download other datasets using the same code on the same machine with no issues :( I get this error now : \r\n```\r\nDownloading data: 16%|███████████████▌ | 55.9M/355M [04:45<25:25, 196kB/s]\r\nTraceback (most recent call last):\r\n File \"/home/skhanuja/Optimal-Resource-Allocation-for-Multilingual-Finetuning/src/train_al.py\", line 1107, in <module>\r\n main()\r\n File \"/home/skhanuja/Optimal-Resource-Allocation-for-Multilingual-Finetuning/src/train_al.py\", line 439, in main\r\n en_dataset = load_dataset(\"xtreme\", \"udpos.English\", split=\"train\", download_mode=\"force_redownload\", verification_mode=\"all_checks\")\r\n File \"/home/skhanuja/miniconda3/envs/multilingual_ft/lib/python3.10/site-packages/datasets/load.py\", line 1782, in load_dataset\r\n builder_instance.download_and_prepare(\r\n File \"/home/skhanuja/miniconda3/envs/multilingual_ft/lib/python3.10/site-packages/datasets/builder.py\", line 872, in download_and_prepare\r\n self._download_and_prepare(\r\n File \"/home/skhanuja/miniconda3/envs/multilingual_ft/lib/python3.10/site-packages/datasets/builder.py\", line 1649, in _download_and_prepare\r\n super()._download_and_prepare(\r\n File \"/home/skhanuja/miniconda3/envs/multilingual_ft/lib/python3.10/site-packages/datasets/builder.py\", line 949, in _download_and_prepare\r\n verify_checksums(\r\n File \"/home/skhanuja/miniconda3/envs/multilingual_ft/lib/python3.10/site-packages/datasets/utils/info_utils.py\", line 62, in verify_checksums\r\n raise NonMatchingChecksumError(\r\ndatasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:\r\n['https://lindat.mff.cuni.cz/repository/xmlui/bitstream/handle/11234/1-3105/ud-treebanks-v2.5.tgz']\r\nSet `verification_mode='no_checks'` to skip checksums verification and ignore this error\r\n```", "If this happens randomly, then this means the data file from the error message is not always downloaded correctly. \r\n\r\nThe only solution in this scenario is to download the dataset again by passing `download_mode=\"force_redownload\"` to the `load_dataset` call." ]

[ { "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 @GuillemGSubies, thanks for reporting.\r\n\r\nI have made a PR to fix this issue and allow metrics to be computed also for multilabel classification problems.", "Looks nice, thank you very much! 🚀 ", "Sorry for reopening but I just noticed that the `_compute` method for the F1 metric is still not good enough for multilabel problems:\r\n\r\nhttps://github.com/huggingface/datasets/blob/92a3ee549705aa0a107c9fa5caf463b3b3da2616/metrics/f1/f1.py#L115\r\n\r\nSomehow we should be able to change the parameter `average` at least", "@GuillemGSubies, the parameter `average` passed to `_compute` is then passed to `f1_score`. This is right." ]

[]

[]

[ { "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 ! Can you try again with pyarrow 6.0.0 ? I think it includes some changes regarding filesystems compatibility with fsspec.", "Hi @lhoestq! I ended up using `fsspec.implementations.arrow.HadoopFileSystem` which doesn't have the problem I described with pyarrow 5.0.0.\r\n\r\nI'll try again with `PyArrowHDFS` once I update arrow to 6.0.0.\r\n\r\nThanks!" ]

[]

[]

[]

[]

[]

[ "_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) | 0.006498 / 0.011353 (-0.004855) | 0.003970 / 0.011008 (-0.007038) | 0.099242 / 0.038508 (0.060734) | 0.044363 / 0.023109 (0.021254) | 0.313900 / 0.275898 (0.038002) | 0.386562 / 0.323480 (0.063082) | 0.003837 / 0.007986 (-0.004149) | 0.004203 / 0.004328 (-0.000125) | 0.076191 / 0.004250 (0.071940) | 0.058823 / 0.037052 (0.021771) | 0.333838 / 0.258489 (0.075349) | 0.368235 / 0.293841 (0.074394) | 0.030774 / 0.128546 (-0.097772) | 0.008787 / 0.075646 (-0.066860) | 0.326474 / 0.419271 (-0.092798) | 0.050903 / 0.043533 (0.007370) | 0.303928 / 0.255139 (0.048789) | 0.321532 / 0.283200 (0.038333) | 0.024162 / 0.141683 (-0.117520) | 1.479662 / 1.452155 (0.027507) | 1.520300 / 1.492716 (0.027584) |\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) | 0.212403 / 0.018006 (0.194397) | 0.448019 / 0.000490 (0.447529) | 0.005465 / 0.000200 (0.005265) | 0.000388 / 0.000054 (0.000334) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.027533 / 0.037411 (-0.009878) | 0.117477 / 0.014526 (0.102952) | 0.121182 / 0.176557 (-0.055374) | 0.181150 / 0.737135 (-0.555985) | 0.128557 / 0.296338 (-0.167782) |\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) | 0.397763 / 0.215209 (0.182554) | 3.959460 / 2.077655 (1.881805) | 1.822057 / 1.504120 (0.317937) | 1.627020 / 1.541195 (0.085826) | 1.695394 / 1.468490 (0.226904) | 0.536848 / 4.584777 (-4.047929) | 3.765205 / 3.745712 (0.019493) | 3.196300 / 5.269862 (-2.073561) | 1.623583 / 4.565676 (-2.942094) | 0.065823 / 0.424275 (-0.358452) | 0.011062 / 0.007607 (0.003455) | 0.500428 / 0.226044 (0.274384) | 5.008816 / 2.268929 (2.739888) | 2.314660 / 55.444624 (-53.129965) | 2.007429 / 6.876477 (-4.869047) | 2.141438 / 2.142072 (-0.000635) | 0.656697 / 4.805227 (-4.148530) | 0.143555 / 6.500664 (-6.357109) | 0.063928 / 0.075469 (-0.011541) |\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) | 1.169038 / 1.841788 (-0.672750) | 15.027186 / 8.074308 (6.952878) | 13.571484 / 10.191392 (3.380092) | 0.166437 / 0.680424 (-0.513986) | 0.017656 / 0.534201 (-0.516545) | 0.397725 / 0.579283 (-0.181558) | 0.451019 / 0.434364 (0.016655) | 0.469134 / 0.540337 (-0.071203) | 0.575885 / 1.386936 (-0.811051) |\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) | 0.006887 / 0.011353 (-0.004465) | 0.004166 / 0.011008 (-0.006842) | 0.077137 / 0.038508 (0.038629) | 0.055631 / 0.023109 (0.032522) | 0.397658 / 0.275898 (0.121760) | 0.473981 / 0.323480 (0.150502) | 0.005365 / 0.007986 (-0.002621) | 0.003401 / 0.004328 (-0.000928) | 0.076481 / 0.004250 (0.072231) | 0.056014 / 0.037052 (0.018961) | 0.415253 / 0.258489 (0.156764) | 0.457620 / 0.293841 (0.163779) | 0.031850 / 0.128546 (-0.096696) | 0.008869 / 0.075646 (-0.066777) | 0.083475 / 0.419271 (-0.335796) | 0.049232 / 0.043533 (0.005699) | 0.392947 / 0.255139 (0.137808) | 0.417243 / 0.283200 (0.134043) | 0.024554 / 0.141683 (-0.117129) | 1.508081 / 1.452155 (0.055926) | 1.541845 / 1.492716 (0.049129) |\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) | 0.228470 / 0.018006 (0.210464) | 0.450933 / 0.000490 (0.450443) | 0.001508 / 0.000200 (0.001308) | 0.000083 / 0.000054 (0.000029) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.030189 / 0.037411 (-0.007222) | 0.118853 / 0.014526 (0.104327) | 0.124809 / 0.176557 (-0.051747) | 0.175066 / 0.737135 (-0.562069) | 0.129819 / 0.296338 (-0.166519) |\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) | 0.451830 / 0.215209 (0.236621) | 4.505352 / 2.077655 (2.427698) | 2.309303 / 1.504120 (0.805183) | 2.120983 / 1.541195 (0.579789) | 2.198808 / 1.468490 (0.730317) | 0.543836 / 4.584777 (-4.040940) | 3.836650 / 3.745712 (0.090938) | 1.872293 / 5.269862 (-3.397568) | 1.122335 / 4.565676 (-3.443342) | 0.067463 / 0.424275 (-0.356812) | 0.012143 / 0.007607 (0.004536) | 0.553674 / 0.226044 (0.327630) | 5.572101 / 2.268929 (3.303173) | 2.772151 / 55.444624 (-52.672473) | 2.451557 / 6.876477 (-4.424920) | 2.521241 / 2.142072 (0.379169) | 0.665799 / 4.805227 (-4.139428) | 0.143842 / 6.500664 (-6.356822) | 0.065373 / 0.075469 (-0.010096) |\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) | 1.271013 / 1.841788 (-0.570775) | 15.290054 / 8.074308 (7.215746) | 14.807044 / 10.191392 (4.615652) | 0.163767 / 0.680424 (-0.516657) | 0.017383 / 0.534201 (-0.516818) | 0.393046 / 0.579283 (-0.186237) | 0.423056 / 0.434364 (-0.011308) | 0.459193 / 0.540337 (-0.081145) | 0.559964 / 1.386936 (-0.826972) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#011b75f044ef7fa6b8981ef3496615296aeb315b \"CML watermark\")\n", "_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) | 0.006112 / 0.011353 (-0.005241) | 0.003712 / 0.011008 (-0.007297) | 0.099996 / 0.038508 (0.061488) | 0.037526 / 0.023109 (0.014417) | 0.305834 / 0.275898 (0.029936) | 0.361368 / 0.323480 (0.037888) | 0.004849 / 0.007986 (-0.003136) | 0.002912 / 0.004328 (-0.001417) | 0.077729 / 0.004250 (0.073479) | 0.053203 / 0.037052 (0.016151) | 0.318088 / 0.258489 (0.059599) | 0.371745 / 0.293841 (0.077904) | 0.029384 / 0.128546 (-0.099162) | 0.008504 / 0.075646 (-0.067142) | 0.318472 / 0.419271 (-0.100799) | 0.046043 / 0.043533 (0.002510) | 0.310418 / 0.255139 (0.055279) | 0.335044 / 0.283200 (0.051844) | 0.020364 / 0.141683 (-0.121319) | 1.503201 / 1.452155 (0.051047) | 1.556408 / 1.492716 (0.063692) |\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) | 0.210245 / 0.018006 (0.192239) | 0.418918 / 0.000490 (0.418428) | 0.002552 / 0.000200 (0.002352) | 0.000084 / 0.000054 (0.000030) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.022295 / 0.037411 (-0.015116) | 0.099534 / 0.014526 (0.085008) | 0.106432 / 0.176557 (-0.070124) | 0.165110 / 0.737135 (-0.572026) | 0.109851 / 0.296338 (-0.186488) |\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) | 0.423947 / 0.215209 (0.208738) | 4.232978 / 2.077655 (2.155323) | 2.004849 / 1.504120 (0.500729) | 1.814345 / 1.541195 (0.273151) | 1.809192 / 1.468490 (0.340702) | 0.561146 / 4.584777 (-4.023631) | 3.385043 / 3.745712 (-0.360669) | 1.708265 / 5.269862 (-3.561597) | 1.030290 / 4.565676 (-3.535387) | 0.067095 / 0.424275 (-0.357180) | 0.011052 / 0.007607 (0.003445) | 0.522416 / 0.226044 (0.296371) | 5.207003 / 2.268929 (2.938075) | 2.367067 / 55.444624 (-53.077558) | 1.998705 / 6.876477 (-4.877772) | 2.068633 / 2.142072 (-0.073439) | 0.672396 / 4.805227 (-4.132831) | 0.135818 / 6.500664 (-6.364846) | 0.065229 / 0.075469 (-0.010240) |\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) | 1.187079 / 1.841788 (-0.654709) | 13.893153 / 8.074308 (5.818845) | 13.951328 / 10.191392 (3.759936) | 0.142519 / 0.680424 (-0.537905) | 0.016546 / 0.534201 (-0.517655) | 0.364008 / 0.579283 (-0.215275) | 0.385957 / 0.434364 (-0.048407) | 0.425218 / 0.540337 (-0.115120) | 0.519586 / 1.386936 (-0.867350) |\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) | 0.005914 / 0.011353 (-0.005439) | 0.003619 / 0.011008 (-0.007389) | 0.077806 / 0.038508 (0.039298) | 0.037254 / 0.023109 (0.014144) | 0.378976 / 0.275898 (0.103078) | 0.433620 / 0.323480 (0.110140) | 0.003291 / 0.007986 (-0.004694) | 0.004523 / 0.004328 (0.000194) | 0.077604 / 0.004250 (0.073353) | 0.047493 / 0.037052 (0.010441) | 0.396027 / 0.258489 (0.137538) | 0.453345 / 0.293841 (0.159504) | 0.028170 / 0.128546 (-0.100376) | 0.008431 / 0.075646 (-0.067215) | 0.083985 / 0.419271 (-0.335286) | 0.045149 / 0.043533 (0.001617) | 0.369364 / 0.255139 (0.114225) | 0.407191 / 0.283200 (0.123991) | 0.024033 / 0.141683 (-0.117649) | 1.516838 / 1.452155 (0.064683) | 1.564260 / 1.492716 (0.071544) |\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) | 0.200848 / 0.018006 (0.182842) | 0.407818 / 0.000490 (0.407328) | 0.003971 / 0.000200 (0.003771) | 0.000077 / 0.000054 (0.000023) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.025033 / 0.037411 (-0.012378) | 0.103585 / 0.014526 (0.089059) | 0.108741 / 0.176557 (-0.067816) | 0.161061 / 0.737135 (-0.576075) | 0.112763 / 0.296338 (-0.183576) |\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) | 0.479913 / 0.215209 (0.264704) | 4.801904 / 2.077655 (2.724249) | 2.511433 / 1.504120 (1.007313) | 2.307523 / 1.541195 (0.766328) | 2.338343 / 1.468490 (0.869853) | 0.557731 / 4.584777 (-4.027046) | 3.386261 / 3.745712 (-0.359451) | 2.999978 / 5.269862 (-2.269883) | 1.463058 / 4.565676 (-3.102619) | 0.067645 / 0.424275 (-0.356630) | 0.011224 / 0.007607 (0.003617) | 0.596854 / 0.226044 (0.370810) | 5.940946 / 2.268929 (3.672017) | 2.980194 / 55.444624 (-52.464430) | 2.634961 / 6.876477 (-4.241516) | 2.648160 / 2.142072 (0.506088) | 0.669728 / 4.805227 (-4.135499) | 0.135536 / 6.500664 (-6.365128) | 0.066865 / 0.075469 (-0.008604) |\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) | 1.287151 / 1.841788 (-0.554637) | 14.491681 / 8.074308 (6.417373) | 14.185752 / 10.191392 (3.994360) | 0.129391 / 0.680424 (-0.551032) | 0.016650 / 0.534201 (-0.517551) | 0.380111 / 0.579283 (-0.199172) | 0.392877 / 0.434364 (-0.041487) | 0.439402 / 0.540337 (-0.100935) | 0.530865 / 1.386936 (-0.856071) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#9aaee6fd0b2bcbe18e4829602084bcd83d669c5e \"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) | 0.011446 / 0.011353 (0.000093) | 0.006623 / 0.011008 (-0.004386) | 0.131915 / 0.038508 (0.093407) | 0.047364 / 0.023109 (0.024255) | 0.369203 / 0.275898 (0.093305) | 0.451509 / 0.323480 (0.128029) | 0.006265 / 0.007986 (-0.001720) | 0.004072 / 0.004328 (-0.000257) | 0.098626 / 0.004250 (0.094375) | 0.079523 / 0.037052 (0.042470) | 0.406038 / 0.258489 (0.147549) | 0.450564 / 0.293841 (0.156723) | 0.050793 / 0.128546 (-0.077753) | 0.014667 / 0.075646 (-0.060979) | 0.401359 / 0.419271 (-0.017913) | 0.072299 / 0.043533 (0.028767) | 0.404456 / 0.255139 (0.149317) | 0.396223 / 0.283200 (0.113023) | 0.037048 / 0.141683 (-0.104635) | 1.869123 / 1.452155 (0.416968) | 1.953621 / 1.492716 (0.460905) |\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) | 0.237246 / 0.018006 (0.219240) | 0.533207 / 0.000490 (0.532717) | 0.007392 / 0.000200 (0.007192) | 0.000117 / 0.000054 (0.000062) |\n\n### Benchmark: benchmark_indices_mapping.json\n\n| metric | select | shard | shuffle | sort | train_test_split |\n|--------|---|---|---|---|---|\n| new / old (diff) | 0.029458 / 0.037411 (-0.007954) | 0.112438 / 0.014526 (0.097912) | 0.139115 / 0.176557 (-0.037441) | 0.215225 / 0.737135 (-0.521911) | 0.134440 / 0.296338 (-0.161898) |\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) | 0.616783 / 0.215209 (0.401574) | 6.113925 / 2.077655 (4.036270) | 2.403465 / 1.504120 (0.899345) | 1.967523 / 1.541195 (0.426329) | 2.042144 / 1.468490 (0.573654) | 0.927447 / 4.584777 (-3.657330) | 5.280413 / 3.745712 (1.534701) | 2.715335 / 5.269862 (-2.554527) | 1.755640 / 4.565676 (-2.810036) | 0.114370 / 0.424275 (-0.309905) | 0.013583 / 0.007607 (0.005976) | 0.761701 / 0.226044 (0.535657) | 7.466049 / 2.268929 (5.197120) | 3.041943 / 55.444624 (-52.402682) | 2.314477 / 6.876477 (-4.562000) | 2.469285 / 2.142072 (0.327213) | 1.216055 / 4.805227 (-3.589172) | 0.214205 / 6.500664 (-6.286459) | 0.080901 / 0.075469 (0.005432) |\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) | 1.565185 / 1.841788 (-0.276603) | 18.387986 / 8.074308 (10.313678) | 19.665109 / 10.191392 (9.473717) | 0.226670 / 0.680424 (-0.453754) | 0.028430 / 0.534201 (-0.505771) | 0.510526 / 0.579283 (-0.068757) | 0.623178 / 0.434364 (0.188814) | 0.592039 / 0.540337 (0.051702) | 0.728462 / 1.386936 (-0.658474) |\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) | 0.009161 / 0.011353 (-0.002192) | 0.004891 / 0.011008 (-0.006117) | 0.106502 / 0.038508 (0.067994) | 0.048234 / 0.023109 (0.025125) | 0.451173 / 0.275898 (0.175275) | 0.557948 / 0.323480 (0.234468) | 0.005350 / 0.007986 (-0.002635) | 0.004559 / 0.004328 (0.000230) | 0.110393 / 0.004250 (0.106142) | 0.060624 / 0.037052 (0.023572) | 0.459265 / 0.258489 (0.200776) | 0.575302 / 0.293841 (0.281461) | 0.051379 / 0.128546 (-0.077167) | 0.015576 / 0.075646 (-0.060070) | 0.116650 / 0.419271 (-0.302621) | 0.065534 / 0.043533 (0.022001) | 0.461431 / 0.255139 (0.206292) | 0.487677 / 0.283200 (0.204477) | 0.037773 / 0.141683 (-0.103910) | 1.992416 / 1.452155 (0.540261) | 1.991280 / 1.492716 (0.498564) |\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) | 0.233607 / 0.018006 (0.215601) | 0.507539 / 0.000490 (0.507049) | 0.001307 / 0.000200 (0.001107) | 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) | 0.032897 / 0.037411 (-0.004514) | 0.126549 / 0.014526 (0.112023) | 0.137893 / 0.176557 (-0.038663) | 0.192124 / 0.737135 (-0.545012) | 0.147300 / 0.296338 (-0.149038) |\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) | 0.679371 / 0.215209 (0.464162) | 6.673249 / 2.077655 (4.595595) | 2.979141 / 1.504120 (1.475022) | 2.568789 / 1.541195 (1.027594) | 2.537540 / 1.468490 (1.069050) | 0.973555 / 4.584777 (-3.611222) | 5.313536 / 3.745712 (1.567824) | 2.693283 / 5.269862 (-2.576579) | 1.819483 / 4.565676 (-2.746194) | 0.111644 / 0.424275 (-0.312631) | 0.013218 / 0.007607 (0.005611) | 0.776114 / 0.226044 (0.550070) | 7.758907 / 2.268929 (5.489978) | 3.417611 / 55.444624 (-52.027013) | 2.859502 / 6.876477 (-4.016975) | 2.927726 / 2.142072 (0.785653) | 1.163671 / 4.805227 (-3.641556) | 0.228636 / 6.500664 (-6.272028) | 0.082077 / 0.075469 (0.006607) |\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) | 1.746150 / 1.841788 (-0.095637) | 17.961955 / 8.074308 (9.887647) | 21.590545 / 10.191392 (11.399153) | 0.210017 / 0.680424 (-0.470406) | 0.028435 / 0.534201 (-0.505766) | 0.509253 / 0.579283 (-0.070030) | 0.606993 / 0.434364 (0.172629) | 0.587189 / 0.540337 (0.046851) | 0.684023 / 1.386936 (-0.702913) |\n\n</details>\n</details>\n\n![](https://cml.dev/watermark.png#9aaee6fd0b2bcbe18e4829602084bcd83d669c5e \"CML watermark\")\n" ]

[]

[]

[ { "color": "72f99f", "default": false, "description": "Discussions on the datasets", "id": 2067401494, "name": "Dataset discussion", "node_id": "MDU6TGFiZWwyMDY3NDAxNDk0", "url": "https://api.github.com/repos/huggingface/datasets/labels/Dataset%20discussion" } ]