Dear SPMers,
I have two linux computer with the same operating system (FedoraCore3),
same Matlab (version 7.0.1), but different SPM5. On one the last update
I installed was Update_546, on the other I installed also the Updates
573 and 665 (the difference is due to the fact that on the computer with
the older version I am doing the analysis for a longitudinal study and I
didn't want to add an extra variable for the analysis).
The problem is the following.
On the computer with all the upgrades I do the preprocessing of some
simple motor fMRI data, and everything is ok.
Then I set up the statistical analysis, save it, run it, and SPM5
(spm_spm.m version 649) stops while doing the "REML estimation" with the
following error (please see bottom of e-mail).
I then transfer to the computer with the older version the .m file I had
set up for the analysis, all the smoothed data, make sure that all the
path to the smoothed data and results directory are set correctly and
run the statistical analysis. On this computer the spm_spm.m version 541
runs smoothly until the end and I've got very nice results (please see
the bottom of the e-mail for the message given by this version).
Could somebody please explain to me why the older version works smoothly
and the new one stops considering tha the data, the vector of onset and
all the rest are exactly the same?
This has happened three times so far, at the beginning I thought it was
a strange dataset, but now it is getting a bit worrying.
Many thanks,
Laura
SPM5: spm_spm (v$Rev: 649 $) 15:50:15 - 08/05/2007
========================================================================
..........
Plane 38/38 , block 13/15 : ...read & mask data
Plane 38/38 , block 14/15 : ...read & mask data
Plane 38/38 , block 15/15 : ...done
Temporal non-sphericity (over voxels) : ...REML estimation
ReML Block - 1
ReML Iteration : 1
...6.310982e+03
ReML Iteration : 2
...2.958229e-01
ReML Iteration : 3
...6.878209e-04
ReML Block - 2
ReML Iteration : 1 ...NaN
Warning: Matrix is close to singular or badly scaled.
Results may be inaccurate. RCOND = -1.000000e+00.
> In spm_reml at 95
In spm_spm at 855
In spm_config_fmri_est>run_est at 390
In spm_jobman>run_struct1 at 1474
In spm_jobman>run_struct1 at 1482
In spm_jobman>run_struct1 at 1482
In spm_jobman>run_struct at 1436
Error running job:
Error using ==> svd
NaN or Inf prevents convergence.
--------------------------
Done.
>>
SPM5: spm_spm (v$Rev: 541 $) 16:33:41 - 08/05/2007
========================================================================
..........
Plane 38/38 , block 13/15 : ...read & mask data
Plane 38/38 , block 14/15 : ...read & mask data
Plane 38/38 , block 15/15 : ...done
Temporal non-sphericity (over voxels) : ...REML estimation
ReML Block - 1
ReML Iteration : 1
...6.310982e+03
ReML Iteration : 2
...2.958229e-01
ReML Iteration : 3
...6.878209e-04
ReML Block - 2
ReML Iteration : 1
...1.302987e+04
ReML Iteration : 2
...8.871181e-01
ReML Iteration : 3
...9.419980e-03
ReML Block - 3
ReML Iteration : 1
...1.581325e+03
ReML Iteration : 2
...1.612759e-01
ReML Iteration : 3
...4.063128e-06
ReML Block - 4
ReML Iteration : 1
...1.094623e+04
ReML Iteration : 2
...2.732539e+00
ReML Iteration : 3
...4.790670e-04
ReML Block - 5
ReML Iteration : 1
...6.853962e+03
ReML Iteration : 2
...8.357067e-01
ReML Iteration : 3
...1.563362e-03
ReML Block - 6
ReML Iteration : 1
...3.078626e+03
ReML Iteration : 2
...6.136454e-01
ReML Iteration : 3
...8.926073e-04
**********************************************************************
This email is confidential and is intended solely for the person or entity to whom it is addressed. If this is not you, please forward the message to [log in to unmask] We have scanned this email before sending it, but cannot guarantee that malicious software is absent and we shall carry no liability in this regard.
We advise that information intended to be kept confidential should not be sent by email. We also advise that health concerns should be discussed with a medical professional in person or by telephone. NHS Direct can also provide advice. We shall not be liable for any failure to follow this advice. University College London Hospitals NHS Foundation Trust (UCLH).
|