mne.inverse_sparse.
tf_mixed_norm
(evoked, forward, noise_cov, alpha_space, alpha_time, loose=0.2, depth=0.8, maxit=3000, tol=0.0001, weights=None, weights_min=None, pca=True, debias=True, wsize=64, tstep=4, window=0.02, return_residual=False, verbose=None)¶TimeFrequency Mixednorm estimate (TFMxNE)
Compute L1/L2 + L1 mixednorm solution on time frequency dictionary. Works with evoked data.
References:
A. Gramfort, D. Strohmeier, J. Haueisen, M. Hamalainen, M. Kowalski TimeFrequency MixedNorm Estimates: Sparse M/EEG imaging with nonstationary source activations Neuroimage, Volume 70, 15 April 2013, Pages 410422, ISSN 10538119, DOI: 10.1016/j.neuroimage.2012.12.051.
A. Gramfort, D. Strohmeier, J. Haueisen, M. Hamalainen, M. Kowalski Functional Brain Imaging with M/EEG Using Structured Sparsity in TimeFrequency Dictionaries Proceedings Information Processing in Medical Imaging Lecture Notes in Computer Science, 2011, Volume 6801/2011, 600611, DOI: 10.1007/9783642220920_49 https://doi.org/10.1007/9783642220920_49
Parameters:  evoked : instance of Evoked
forward : dict
noise_cov : instance of Covariance
alpha_space : float in [0, 100]
alpha_time : float in [0, 100]
loose : float in [0, 1]
depth: None  float in [0, 1] :
maxit : int
tol : float
weights: None  array  SourceEstimate :
weights_min: float :
pca: bool :
debias: bool :
wsize: int :
tstep: int :
window : float or (float, float)
return_residual : bool
verbose : bool, str, int, or None


Returns:  stc : instance of SourceEstimate
residual : instance of Evoked

See also