LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] lmp.extract_compute returns Null
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] lmp.extract_compute returns Null

From: Axel Kohlmeyer <akohlmey@...24...>
Date: Wed, 18 Apr 2018 08:27:51 -0400

On Wed, Apr 18, 2018 at 12:18 AM, Patrick Aselin <aselip@...707...> wrote:
> Hello everyone!
> I am working on a course project to make a very simple interface to LAMMPS
> through the PyLammps package working with Python3.6 and the latest LAMMPS
> release. My hope is to be able to run a simulation without having to save
> files (dump, RDF, MSD, etc). So far I have succeeded by setting up LAMMPS
> with
> lmp = lammps()
> L = PyLammps(ptr=lmp)
> and then collecting thermo data from L.runs[...] and coordinates by using
> lmp.gather_atoms("x", 1,3). My problem is with trying to find collect the
> MSD and RDF computes from the provided methods. After I run the program
> below, I have been trying to read the RDF using variations of the command
> lmp.extract_compute("myRDF",0,1), but only receive types <lammps.LP_c_double
> at 0x7f3b88156510> that can't be subscripted as they seem to be Null.

i don't understand why you are trying to extract data from the
compute. shouldn't you instead extract the averaged data from fix 4

anyway, your code is not correct. compute rdf computes a global array
(with 2*n+1 columns with n being the number of pairs to types to be
these will be in a 2d-double array, so you would have to do something like this:

for i in range(0,50):
  print("%8.3f %g %g" % (rdfone[i][0],rdfone[i][1],rdfone[i][2]))

for the fix, the API for getting access is slightly different, as the
indexing is already done from within the code. so the equivalent
output from above would be created with:

for i in range(0,50):
  print("%8.3f %g %g" %

in both cases, you have three columns: the value r, g(r), Int_0^r(g(r))

> I'm
> not sure what I'm doing wrong here as these commands seem to follow the
> manuals. Is what I am trying to do even possible? Would it be better to just
> collect the coordinate data and then use another Python function to compute
> these?

it is possible. it looks like you need to look a bit closer at the


> Thanks,
> Patrick
> from lammps import lammps, PyLammps
> lmp = lammps()
> L = PyLammps(ptr=lmp)
> L.units("lj")
> L.atom_style(" atomic")
> L.atom_modify("map array")
> L.lattice("fcc 0.8442")
> L.region("box block 0 4 0 4 0 4")
> L.create_box("1 box")
> L.create_atoms("1 box")
> L.mass("1 1.0")
> L.velocity("all create 1.44 87287 loop geom")
> L.pair_style("lj/cut 2.5")
> L.pair_coeff("1 1 1.0 1.0 2.5")
> L.neighbor("0.3 bin")
> L.neigh_modify("delay 0 every 20 check no")
> L.fix("1 all nve")
> L.compute("msdcal all msd") #Compute msd
> L.compute("myRDF all rdf 50") # Compute rdf
> L.fix("4 all ave/time 100 1 100 c_myRDF[*] file mode vector")
> L.fix("5 all ave/time 1 1 1 c_msdcal[4] file mode scalar")
> L.thermo(50)
> L.thermo_style("custom step temp ke pe etotal vol press pxx pyy pzz lx ly lz
> enthalpy")
> L.dump("id all atom 50 dump.melt")
> lmp.extract_compute("myRDF",0,1) #returns null
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites,!
> _______________________________________________
> lammps-users mailing list

Dr. Axel Kohlmeyer  akohlmey@...24...
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.