pyriemann.transfer.TLCenter

class pyriemann.transfer.TLCenter(target_domain, metric='riemann')

Recenter data for transfer learning.

Recenter the data points from each domain to the Identity on manifold, ie make the mean of the datasets become the identity. This operation corresponds to a whitening step if the SPD matrices represent the spatial covariance matrices of multivariate signals.

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.

metricstr, default=’riemann’

The metric for mean, can be: ‘ale’, ‘alm’, ‘euclid’, ‘harmonic’, ‘identity’, ‘kullback_sym’, ‘logdet’, ‘logeuclid’, ‘riemann’, ‘wasserstein’, or a callable function. Note, however, that only when using the ‘riemann’ metric that we are ensured to re-center the data points precisely to the Identity.

Notes

New in version 0.4.

References

1

Transfer Learning: A Riemannian Geometry Framework With Applications to Brain–Computer Interfaces P Zanini et al, IEEE Transactions on Biomedical Engineering, vol. 65, no. 5, pp. 1107-1116, August, 2017

Attributes
recenter_dict

Dictionary with key=domain_name and value=domain_mean

__init__(target_domain, metric='riemann')

Init

fit(X, y_enc)

Fit TLCenter.

Calculate the mean of all matrices in 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
selfTLCenter instance

The TLCenter instance.

fit_transform(X, y_enc)

Fit TLCenter and then transform data points.

Calculate the mean of all matrices in each domain and then recenter them to Identity.

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 mean in the Identity.

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)

Re-center the data points in the target domain to Identity.

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 mean in the Identity.