Dear LAMMPS users,
Hi. Tried google but find no answer. One quick question. Is there a way to exclude the defined include command? I know there are "similar" commands such as uncompute, delete, or clear commands. My intention is to compute the displacements of couple grouped atoms within different calculations. For instance, run 10000 steps -> find displacements within these 1000 steps, then run another 20000 steps, find the displacements within these 2000 steps, etc. First workaround, I believe I can simply subtract the displacements accordingly. If I am going to sequentially run ten different runs with some tweaks, I wish I can start over the new displacement calculation. Second workaround, I can also uncompute then compute again. But since I want to calculate multiple displacements, uncompute these in that file then include the file again seems not efficient. Thanks for your suggestion.
Best,
F.-C.
no, there is not an “uninclude” command.
There is a clear command which allows one input script
to essentially start over.
Steve
Thanks a lot for your prompt response.
In that case, it seems like I can only clear (all) and include all the files again even I only want to modify one include file though. Thanks.
Cheers,
F.-C.
LAMMPS has no idea what was in a previously included file of commands.
It just processes commands, one at a time, as if they
were all in a single input file. If the include file had
a velocity command that set atom velocities or deleted atoms,
how would you imagine that an “uninclude” command would unset the
velocities or undelete the deleted atoms?
Steve
LAMMPS has no idea what was in a previously included file of commands.
It just processes commands, one at a time, as if they
were all in a single input file. If the include file had
a velocity command that set atom velocities or deleted atoms,
how would you imagine that an "uninclude" command would unset the
velocities or undelete the deleted atoms?
how about writing an include file that reverses the commands from the
first include.
on top of that, if the desire is to return to a previous state, there
are multiple options (write_restart/read_restart,
write_data/read_data).
also check out fix store/state, since you are only interested in
specific displacement, that should suffice.
finally, it may be even easier to do the whole analysis in
post-processing. it seems simple enough.
axel.
Dear Both,
Great. For the time being, fix store/state will do the trick (so does post-processing). Thanks for the discussion and suggestion.
Cheers,
F.-C.