ValueError: too many values to unpack (expected 2)

Support & Discussions
,
1
  • MNE version: 1.2.1
  • operating system: Ubuntu 20.04.5 LTS

Dear MNE community,

I’m trying to replicate the “Permutation t-test on source data with spatio-temporal clustering” tutorial with my own data:

https://mne.tools/stable/auto_tutorials/stats-source-space/20_cluster_1samp_spatiotemporal.html#sphx-glr-auto-tutorials-stats-source-space-20-cluster-1samp-spatiotemporal-py

In my case, instead of making a paired contrast, like in the tutorial:

X = np.abs(X)  # only magnitude
X = X[:, :, :, 0] - X[:, :, :, 1]  # make paired contrast

I’m comparing two independent groups, so I use “spatio_temporal_cluster_test” instead of “spatio_temporal_cluster_1samp_test”

So I perform a cluster perm test as follows:

# Here we set a cluster forming threshold based on a p-value for
# the cluster based permutation test.
# We use a two-tailed threshold, the "1 - p_threshold" is needed
# because for two-tailed tests we must specify a positive threshold.
p_threshold = 0.05
df = X.shape[0] - 1  # degrees of freedom for the test

t_threshold = stats.distributions.f.ppf(1. - p_threshold / 2.,
                                        X.shape[0] - 1, X2.shape[0] - 1)

n_permutations = 1000  # Should be above 1000.

# Now let's actually do the clustering. This can take a long time...
print('Clustering.')
T_obs, clusters, cluster_p_values, H0 = clu = \
    spatio_temporal_cluster_test([X, X2],
                                 n_permutations=n_permutations,
                                 threshold=t_threshold,
                                 max_step=2,
                                 n_jobs=3,
                                 seed=49,
                                 buffer_size=None,
                                 adjacency=adjacency,
                                 out_type='mask',
                                 verbose=True)

However, the error message:

In [27]: runcell(27, '/home/.../3.1._Source_Estimation_ERP_Stat.py')
Traceback (most recent call last):

  File "/home/.../3.1._Source_Estimation_ERP_Stat.py", line 388, in <module>
    stc_all_cluster_vis = summarize_clusters_stc(clu,

  File "/home/.../anaconda3/envs/mne/lib/python3.10/site-packages/mne/stats/cluster_level.py", line 1515, in summarize_clusters_stc
    t_inds, v_inds = clusters[cluster_ind]

ValueError: too many values to unpack (expected 2)

is displayed to me when I run:

stc_all_cluster_vis = summarize_clusters_stc(clu,
                                             tstep=tstep,
                                             vertices=fsave_vertices,
                                             subject='fsaverage')

The spatio_temporal_cluster_test found 5 clusters, and two of them are “statistically significant”.

The variable ‘clusters’ is a list of length 5 and the entries are numpy arrays of boleans. The result of:

In [31]: clusters[0].shape

is:

Out[31]: (221, 20484)

Could anyone help me to understand what I’m doing wrong? I can provide you more information if needed. Any hint would be quite helpful

Thanks in advance,

Bruno

2

without looking too closely beyond your first paragraph or so, you should know that there’s a separate tutorial that already does 2-sample tests: 2 samples permutation test on source data with spatio-temporal clustering — MNE 1.3.1 documentation So I suggest starting from there rather than trying to adapt the 1-sample tutorial yourself. If after that your’e still stuck, let us know.

3

Thanks, for the hint!

The problem was that I was passing 'out_type=‘mask’ to the ‘spatial_cluster_test’.

I should have used 'out_type=‘indices’. Now the results are being correctly plotted.

1 Like