Hi Grigory,
I´ve never seen this happening with the tutorial data. Sigma2 is a squared
error and should never by negative. It could indeed be caused by the issue
described in the link on the Wiki, or other communication errors. Did you
try setting the segmentation offloading off?
Sjors
> Dear Sjors,
>
> I've just started learning Relion 1.2 using the tutorial you provided, but
> the run is crushing after several iteration (both 2d and 3d
> classification)
> with the following error:
>
> =============================
>> Expectation iteration 6 of 25
>> 60/ 60 sec
>> ............................................................~~(,_,">
>> Maximization ...
>> 000/??? sec ~~(,_,">
>> [oo] DIRECT_A1D_ELEM(sigma2, i)= -468.511
>> BackProjector::reconstruct: ERROR: unexpectedly small, yet non-zero
>> sigma2
>> value, this should not happen...a
>> File: src/backprojector.cpp line: 668
>> --------------------------------------------------------------------------
>> MPI_ABORT was invoked on rank 2 in communicator MPI_COMM_WORLD
>> with errorcode 1.
>
>
> I'm not sure if this can be
> related<http://www2.mrc-lmb.cam.ac.uk/relion/index.php/FAQs#My_runs_keep_crashing_at_seemingly_random_points_in_the_refinement>to
> the work of OpenMPI on our cluster...
>
> Best regards,
> Grigory Sharov,
>
> Institut de Génétique et de Biologie Moléculaire et Cellulaire
> Departement of Structural Biology and Genomics
> 1, rue Laurent Fries
> 67404 Illkirch, France
> tel. 03 88 65 57 48
> e-mail: [log in to unmask]
>
--
Sjors Scheres
MRC Laboratory of Molecular Biology
Francis Crick Avenue, Cambridge Biomedical Campus
Cambridge CB2 0QH, U.K.
tel: +44 (0)1223 267061
http://www2.mrc-lmb.cam.ac.uk/groups/scheres
|