Home

Here is an example code:

metric = { 'mean': "convex", 'distance': "convex" }

distance_methods["convex"] = lambda A, B: np.linalg.norm(A - B, ord='fro')

clf = make_pipeline(XdawnCovariances(), MDM(metric=metric)) skf = StratifiedKFold(n_splits=5) n_matrices, n_channels, n_classes = 100, 3, 2 covset = get_covmats(n_matrices, n_channels) labels = get_labels(n_matrices, n_classes)

score = cross_val_score(clf, covset, labels, cv=skf, scoring='roc_auc') assert score.mean() > 0

We provide a convex model for the first situation based on frobenius distance (method fro_mean_convex), as well as a wrapper around QAOA optimizer (class NaiveQAOAOptimizer) that rounds covariance matrices to a certain precision and convert each resulting integer to binary. The implementation is based on Qiskit's IntegerToBinary, a bounded-coefficient encoding method.

Therefore, complexity of the optimizer raises as a function of the matrix size and the coefficients, and hence it best adapt to covariance matrices having a limited number of channels, with naturally bounded and "differentiable" values. "