das.io

Load data for training/testing. See doc/data.md for a description of the data schema.

class das.io.MemoryMappedDirectoryStore(path, normalize_keys=False, dimension_separator: Optional[Literal['.', '/']] = None)

das.io.load(location, x_suffix='', y_suffix='')

Load data for training/testing from zarr store, npy directory, or hdf5 file.

Parameters

• location ([type]) – [description]

• x_suffix (str, optional) – alternative key for the training source and target (allows for different x/y’s for the same y/x in one data file)

• y_suffix (str, optional) – alternative key for the training source and target (allows for different x/y’s for the same y/x in one data file)

Returns

dict-like complying with data schema defined above