Verification regarding weighted means of SPD matrices

verif_weighted_mean_spd.py

import pyriemann
import numpy as np
from pyriemann.utils.mean import mean_riemann

def generate_cov(Nt, Ne):
    """Generate a set of cavariances matrices for test purpose"""
    rs = np.random.RandomState(1234)
    diags = 2.0 + 0.1 * rs.randn(Nt, Ne)
    A = 2*rs.rand(Ne, Ne) - 1
    A /= np.atleast_2d(np.sqrt(np.sum(A**2, 1))).T
    covmats = np.empty((Nt, Ne, Ne))
    for i in range(Nt):
        covmats[i] = np.dot(np.dot(A, np.diag(diags[i])), A.T)
    return covmats, diags, A

covs, diags, A = generate_cov(4, 5)
mean1 = mean_riemann(covs, sample_weight=np.array([0.1, 0.2, 0.3, 0.4]))
covs2 = np.empty_like(covs)
for i in range(4):
  covs2[i,:,:] = covs[i,:,:]*(i+1)/10
mean2 = mean_riemann(covs2)

print(mean1)
print(mean2)

mean_equi = mean_riemann(covs, sample_weight=np.array([0.25, 0.25, 0.25, 0.25]))
covs3 = np.empty_like(covs)
for i in range (4):
    covs3[i,:,:] = covs[i,:,:]*0.25
mean_equi2 = mean_riemann(covs3)

print(mean_equi)
print(mean_equi2)