pyriemann.transfer.TLStretch

class pyriemann.transfer.TLStretch(target_domain, final_dispersion=1.0, centered_data=False, metric='riemann')

Stretch data for transfer learning.

Change the dispersion of the datapoints around their geometric mean for each dataset so that they all have the same desired value.

Note

Using .fit() and then .transform() will give different results than .fit_transform(). In fact, .fit_transform() should be applied on the training dataset (target and source) and .transform() on the test partition of the target dataset.

Parameters
target_domainstr

Domain to consider as target.

dispersionfloat, default=1.0

Target value for the dispersion of the data points.

centered_databool, default=False

Whether the data has been re-centered to the Identity beforehand.

metricstr, default=’riemann’

The metric for calculating the dispersion can be: ‘ale’, ‘alm’, ‘euclid’, ‘harmonic’, ‘identity’, ‘kullback_sym’, ‘logdet’, ‘logeuclid’, ‘riemann’, ‘wasserstein’, or a callable function.

Notes

New in version 0.4.

References

1

Riemannian Procrustes analysis: transfer learning for brain-computer interfaces PLC Rodrigues et al, IEEE Transactions on Biomedical Engineering, vol. 66, no. 8, pp. 2390-2401, December, 2018

Attributes
dispersions_dict

Dictionary with key=domain_name and value=domain_dispersion.

__init__(target_domain, final_dispersion=1.0, centered_data=False, metric='riemann')

Init

fit(X, y_enc)

Fit TLStretch.

Calculate the dispersion around the mean for each domain.

Parameters
Xndarray, shape (n_matrices, n_channels, n_channels)

Set of SPD matrices.

y_encndarray, shape (n_matrices,)

Extended labels for each matrix.

Returns
selfTLStretch instance

The TLStretch instance.

fit_transform(X, y_enc)

Fit TLStretch and then transform data points.

Calculate the dispersion around the mean for each domain and then stretch the data points to the desired final dispersion.

Note

This method is designed for using at training time. The output for .fit_transform() will be different than using .fit() and .transform() separately.

Parameters
Xndarray, shape (n_matrices, n_channels, n_channels)

Set of SPD matrices.

y_encndarray, shape (n_matrices,)

Extended labels for each matrix.

Returns
Xndarray, shape (n_matrices, n_classes)

Set of SPD matrices with desired final dispersion.

get_params(deep=True)

Get parameters for this estimator.

Parameters
deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns
paramsdict

Parameter names mapped to their values.

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters
**paramsdict

Estimator parameters.

Returns
selfestimator instance

Estimator instance.

transform(X, y_enc=None)

Stretch the data points in the target domain.

Note

The stretching operation is properly defined only for the riemann metric.

Parameters
Xndarray, shape (n_matrices, n_channels, n_channels)

Set of SPD matrices.

y_encNone

Not used, here for compatibility with sklearn API.

Returns
Xndarray, shape (n_matrices, n_classes)

Set of SPD matrices with desired final dispersion.