Compute the split info for the splits in the given data dir.
tfds.folder_dataset.compute_split_info_from_directory(
*,
out_dir: Optional[epath.PathLike] = None,
data_dir: tfds.typing.PathLike
,
filename_template: Union[None, str, tfds.core.ShardedFileTemplate
] = None
) -> List[tfds.core.SplitInfo
]
Arguments |
out_dir
|
directory where to save the metadata. It should be available from
the apache beam workers. If not set, apache beam won't be used (only
available with some file formats).
|
data_dir
|
directory where the data is.
|
filename_template
|
the template to which the data files correspond. If None,
then the default template is used.
|
Returns |
list of split infos for the splits in the given data dir.
|