Attend the Women in ML Symposium on December 7 Register now


Stay organized with collections Save and categorize content based on your preferences.

Add metadata required to load with TFDS.

See documentation for usage:

data_dir Dataset path on which to save the metadata
features dict of tfds.features.FeatureConnector matching the proto specs.
split_infos Can be either: * A path to the pre-computed split info values ( the out_dir kwarg of tfds.folder_dataset.compute_split_info) * A list of tfds.core.SplitInfo (returned value of tfds.folder_dataset.compute_split_info) * None to auto-compute the split info.
version Optional dataset version (auto-infer by default, or fallback to 1.0.0)
filename_template the template for the filenames of the data. If None, then the default template '{DATASET}-{SPLIT}.{FILEFORMAT}-{SHARD_X_OF_Y}' is used. A string or a ShardedFileTemplate can be given for custom templates.
check_data If True, perform additional check to validate the data in data_dir is valid
**ds_info_kwargs Additional metadata forwarded to tfds.core.DatasetInfo ( description, homepage,...). Will appear in the doc.