[Mne_analysis] mixed design ANOVA with spatio-temporal clustering

Talitha Ford tcford at swin.edu.au
Mon Mar 6 23:50:04 EST 2017
Search archives:

Hi again,
My apologies, I found the source of the first value error, however now I have this one:

ValueError: bad axis1 argument to swapaxes

which I am quite sure is related to the uneven groups.

Here is the full error message:

n_permutations = 128
print('Clustering.')
T_obs, clusters, cluster_p_values, H0 = clu = \
    spatio_temporal_cluster_test(all_data, connectivity=connectivity, n_jobs=2,
                                 threshold=f_threshold, stat_fun=stat_fun,
                                 n_permutations=n_permutations,
                                 buffer_size=None) # clu is an array of T_obs, clusters, cluster_p_values, H0
## -- End pasted text –

Clustering.
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-30-377c44f3e5a4> in <module>()
      5                                  threshold=f_threshold, stat_fun=stat_fun,
      6                                  n_permutations=n_permutations,
----> 7                                  buffer_size=None) # clu is an array of T_obs, clusters, cluster_p_values, H0

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/stats/cluster_level.pyc in spatio_temporal_cluster_test(X, threshold, n_permutations, tail, stat_fun, connectivity, verbose, n_jobs, seed, max_step, spatial_exclude, step_down_p, t_power, out_type, check_disjoint, buffer_size)

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/utils.pyc in verbose(function, *args, **kwargs)
    706         with use_log_level(verbose_level):
    707             return function(*args, **kwargs)
--> 708     return function(*args, **kwargs)
    709
    710

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/stats/cluster_level.pyc in spatio_temporal_cluster_test(X, threshold, n_permutations, tail, stat_fun, connectivity, verbose, n_jobs, seed, max_step, spatial_exclude, step_down_p, t_power, out_type, check_disjoint, buffer_size)
   1414                                    t_power=t_power, out_type=out_type,
   1415                                    check_disjoint=check_disjoint,
-> 1416                                    buffer_size=buffer_size)
   1417     return out
   1418

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/stats/cluster_level.pyc in permutation_cluster_test(X, threshold, n_permutations, tail, stat_fun, connectivity, verbose, n_jobs, seed, max_step, exclude, step_down_p, t_power, out_type, check_disjoint, buffer_size)

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/utils.pyc in verbose(function, *args, **kwargs)
    706         with use_log_level(verbose_level):
    707             return function(*args, **kwargs)
--> 708     return function(*args, **kwargs)
    709
    710

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/stats/cluster_level.pyc in permutation_cluster_test(X, threshold, n_permutations, tail, stat_fun, connectivity, verbose, n_jobs, seed, max_step, exclude, step_down_p, t_power, out_type, check_disjoint, buffer_size)
   1025                                      t_power=t_power, out_type=out_type,
   1026                                      check_disjoint=check_disjoint,
-> 1027                                      buffer_size=buffer_size)
   1028
   1029

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/stats/cluster_level.pyc in _permutation_cluster_test(X, threshold, n_permutations, tail, stat_fun, connectivity, verbose, n_jobs, seed, max_step, exclude, step_down_p, t_power, out_type, check_disjoint, buffer_size)

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/utils.pyc in verbose(function, *args, **kwargs)
    706         with use_log_level(verbose_level):
    707             return function(*args, **kwargs)
--> 708     return function(*args, **kwargs)
    709
    710

/Users/tcford/anaconda/lib/python2.7/site-packages/mne/stats/cluster_level.pyc in _permutation_cluster_test(X, threshold, n_permutations, tail, stat_fun, connectivity, verbose, n_jobs, seed, max_step, exclude, step_down_p, t_power, out_type, check_disjoint, buffer_size)
    710     # Step 1: Calculate T-stat for original data
    711     # -------------------------------------------------------------
--> 712     T_obs = stat_fun(*X)
    713     logger.info('stat_fun(H1): min=%f max=%f' % (np.min(T_obs), np.max(T_obs)))
    714

<ipython-input-22-27a09a4d2798> in stat_fun(*args)
     16
     17 def stat_fun(*args):
---> 18     return f_mway_rm(np.swapaxes(args, 1, 0), factor_levels=factor_levels,
     19                      effects=effects, return_pvals=return_pvals)[0]
     20     # get f-values only.

/Users/tcford/anaconda/lib/python2.7/site-packages/numpy/core/fromnumeric.pyc in swapaxes(a, axis1, axis2)
    500         swapaxes = a.swapaxes
    501     except AttributeError:
--> 502         return _wrapit(a, 'swapaxes', axis1, axis2)
    503     return swapaxes(axis1, axis2)
    504

/Users/tcford/anaconda/lib/python2.7/site-packages/numpy/core/fromnumeric.pyc in _wrapit(obj, method, *args, **kwds)
     45     except AttributeError:
     46         wrap = None
---> 47     result = getattr(asarray(obj), method)(*args, **kwds)
     48     if wrap:
     49         if not isinstance(result, mu.ndarray):

ValueError: bad axis1 argument to swapaxes

Any help would be greatly appreciated,

Talitha

From: <mne_analysis-bounces at nmr.mgh.harvard.edu> on behalf of Talitha Ford <tcford at swin.edu.au>
Reply-To: Discussion and support forum for the users of MNE Software <mne_analysis at nmr.mgh.harvard.edu>
Date: Tuesday, 7 March 2017 at 15:14
To: "mne_analysis at nmr.mgh.harvard.edu" <mne_analysis at nmr.mgh.harvard.edu>
Subject: [Mne_analysis] mixed design ANOVA with spatio-temporal clustering

Hi all,

First off, thanks for the new repeated measure ANOVA capability in spatio-temporal cluster analysis!
I’d like to adapt this script for my data, however it is a 2(between-group) by 2(within-group) mixed design. Unfortunately the groups are uneven, so even if I enter the groups data as if they are a within-subjects factors I get an error (at least that’s how I’m interpreting the ValueError: total size of new array must be unchanged).

The shape (subs, time, vertices) of my groups are:
grp1 (16, 151, 20484)
grp2 (20,151,20484)

Is there a mixed design ANOVA alternative?

Talitha
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20170307/f5a6b2ce/attachment-0001.html 


More information about the Mne_analysis mailing list