issue on mixed_norm_solver() MNE

cv_issue

from os.path import join as pjoin
import numpy as np
from numpy.linalg import norm
import time
import mne
from mne.datasets import sample
from mne.inverse_sparse.mxne_inverse import (_to_fixed_ori, _prepare_gain)
from mne.inverse_sparse.mxne_optim import mixed_norm_solver, norm_l2inf, norm_l21
# from mne.inverse_sparse.mxne_optim import _mixed_norm_solver_cd

data_path = sample.data_path()
fwd_fname = pjoin(data_path, "MEG/sample/sample_audvis-meg-eeg-oct-6-fwd.fif")
ave_fname = data_path + '/MEG/sample/sample_audvis-ave.fif'
cov_fname = data_path + '/MEG/sample/sample_audvis-shrunk-cov.fif'

condition = 'Left Auditory'
fwd = mne.read_forward_solution(fwd_fname, surf_ori=True)
fwd = _to_fixed_ori(fwd)
noise_cov = mne.read_cov(cov_fname)
evoked = mne.read_evokeds(ave_fname, condition=condition, baseline=(None, 0))
evoked.crop(tmin=0.04, tmax=0.18)
evoked = evoked.pick_types(eeg=True, meg=True)

loose, depth = 0, 0.8

# apply solver on whitened data
gain, gain_info, whitener, source_weighting, mask = _prepare_gain(
    fwd, evoked.info, noise_cov, pca=False, depth=depth,
    loose=loose, weights=None, weights_min=None)

sel = [evoked.ch_names.index(name) for name in gain_info['ch_names']]
M = evoked.data[sel]
M = np.dot(whitener, M)


alpha_max = norm_l2inf(np.dot(gain.T, M), n_orient=1)
alpha = alpha_max / 2.
n_alphas = 5
times = np.zeros(n_alphas)
tol = 1e-6
for i, alpha_div in enumerate(np.logspace(0.69, 1, n_alphas)):
    alpha = alpha_max / alpha_div
    t0 = time.time()
    X, active_set, E = mixed_norm_solver(M, gain, alpha,
                                         solver='cd',
                                         debias=False, tol=tol)
    times[i] = time.time() - t0