osl_dynamics.data.rw

Functions for reading and writing data.

Module Contents

Functions

validate_inputs(inputs)	Validates inputs.
file_ext(filename)	Returns the extension of a file.
list_dir(path[, keep_ext])	Lists a directory.
load_data(data[, data_field, picks, ...])	Loads time series data.
load_fif(filename[, picks, reject_by_annotation])	Load a fif file.
load_matlab(filename, field)	Loads a MATLAB file.
loadmat(filename[, return_dict])	Wrapper for scipy.io.loadmat or mat73.loadmat.

Attributes

_logger
_allowed_ext

osl_dynamics.data.rw._logger

osl_dynamics.data.rw._allowed_ext = ['.npy', '.mat', '.txt', '.fif']

osl_dynamics.data.rw.validate_inputs(inputs)

Validates inputs.

Parameters:

inputs (list of str or str or np.ndarray) – Inputs files or data.

Returns:

validated_inputs – Validated inputs.

Return type:

list of str or str

osl_dynamics.data.rw.file_ext(filename)

Returns the extension of a file.

Parameters:

filename (str) – Path to file.

osl_dynamics.data.rw.list_dir(path, keep_ext=None)

Lists a directory.

Parameters:

• path (str) – Directory to list.

• keep_ext (str or list, optional) – Extensions of files to include in the returned list. Default is to include all files.

Returns:

files – Full path to files with the correct extension.

Return type:

list

osl_dynamics.data.rw.load_data(data, data_field='X', picks=None, reject_by_annotation=None, mmap_location=None, mmap_mode='c')

Loads time series data.

Checks the data shape is time by channels and that the data is float32.

Parameters:

• data (numpy.ndarray or str or list) – An array or path to a .npy, .mat, .txt or .fif file containing the data.

• data_field (str, optional) – If a MATLAB filename is passed, this is the field that corresponds to the data.

• picks (str or list of str, optional) – Argument passed to mne.io.Raw.get_data or mne.Epochs.get_data. Only used if a fif file is passed.

• reject_by_annotation (str, optional) –

  Argument passed to mne.io.Raw.get_data. Only used if a fif file is passed.

• mmap_location (str, optional) – Filename to save the data as a numpy memory map.

• mmap_mode (str, optional) – Mode to load memory maps in. Default is 'c'.

Returns:

data – Data.

Return type:

np.memmap or np.ndarray

osl_dynamics.data.rw.load_fif(filename, picks=None, reject_by_annotation=None)

Load a fif file.

Parameters:

• filename (str) – Path to fif file. Must end with 'raw.fif' or 'epo.fif'.

• picks (str or list of str, optional) –

  Argument passed to mne.io.Raw.get_data or mne.Epochs.get_data.

• reject_by_annotation (str, optional) –

  Argument passed to mne.io.Raw.get_data.

Returns:

data – Time series data in format (n_samples, n_channels). If an mne.Epochs fif file is pass ('epo.fif') the we concatenate the epochs in the first axis.

Return type:

np.ndarray

osl_dynamics.data.rw.load_matlab(filename, field)

Loads a MATLAB file.

Parameters:

• filename (str) – Filename of MATLAB file to read.

• field (str) – Field that corresponds to the data.

Returns:

data – Data in the MATLAB file.

Return type:

np.ndarray

osl_dynamics.data.rw.loadmat(filename, return_dict=False)

Wrapper for scipy.io.loadmat or mat73.loadmat.

Parameters:

• filename (str) – Filename of MATLAB file to read.

• return_dict (bool, optional) – If there’s only one field should we still return a dict? Default is to return a numpy array if there is only one field. If there are multiple fields, a dict is always returned.

Returns:

mat – Data in the MATLAB file.

Return type:

dict or np.ndarray