search

WarwickMicroscopy / Felix

Felix: Bloch wave method diffraction pattern simulation software
16 stars 10 forks source link

Issues running felixrefine #131

Closed JacobJoseph-gu closed 7 years ago

JacobJoseph-gu commented 7 years ago

It may be a problem on my end, but after compiling the up-to-date felixrefine, it does not seem to run on any of the samples. Old felixrefine used to work on FR-GaAs but I see that it now says on the master branch that it is broken for that sample

rbeanland commented 7 years ago

This problem may be because the input file has changed as different refinement methods have been tested. I have updated felix.inp in the FR-GaAs to the current version and merged the latest code into master, so this should work.
A second possibility is a recent CSC rebuild which removed old mpi modules. To build or run the software the command: module load ompi/2.0.1/gnu/4.9.2 :should be run first. (This needs to be amended in the make file but I haven't yet done so.)

JacobJoseph-gu commented 7 years ago

Thanks, that makes sense. I have just updated my repository to master and the felixrefine output was the following:

felixrefine: Version: master / BUILD / Alpha
Date: 12-01-2017
Status: Parabolic refinement test&debug
on rank= 0 of 2 in total.

D:Refining Isotropic Debye Waller Factors Refining by Parabola Refining atoms 1 2 ReadExperimentalImages( 0 )error reading -5-5-3.img, line 1 ReadExperimentalImages( 1 )error reading -5-5-3.img, line 1 felixrefine( 1 ) error in ReadExperimentalImages felixrefine( 0 ) error in ReadExperimentalImages Note: The following floating-point exceptions are signalling: IEEE_UNDERFLOW_FLAG IEEE_DENORMAL Note: The following floating-point exceptions are signalling: IEEE_UNDERFLOW_FLAG IEEE_DENORMAL

I've successfully done apt-get install environment-modules but am having problems using the commands. I will keep working on that so that I can use the module load command you suggested.

rbeanland commented 7 years ago

This error says that the experimental images are missing.Their names need to be changed to match the expected format. I will try to do this later today

rbeanland commented 7 years ago

Ok, done & updated in master. I haven't tested it, but it should work (famous last words)...

JacobJoseph-gu commented 7 years ago

There may be a mistake on my end but with the updates my output was the following:

felixrefine: Version: master / BUILD / Alpha
Date: 12-01-2017
Status: Parabolic refinement test&debug
on rank= 0 of 1 in total.

D:Refining Isotropic Debye Waller Factors Refining by Parabola Refining atoms 1 2 ReadExperimentalImages( 0 )error reading -5-5-3.img, line 1 felixrefine( 0 ) error in ReadExperimentalImages Note: The following floating-point exceptions are signalling: IEEE_UNDERFLOW_FLAG IEEE_DENORMAL

rbeanland commented 7 years ago

The renamed files were not being tracked in git. I've added them now, so there should be a load of *.img files in your FR-GaAs folder when you update.

JacobJoseph-gu commented 7 years ago

Thanks Richard, that now works. I have also ran convert on lots of the output images and they look good. Closing Issue