error on LAMMPS with DPLR

[ryujh28]

Hi,

All my DeepMD molecular dynamics simulations were done well with LAMMPS.
This time, I'm trying to run a molecular dynamics simulation with LAMMPS with DPLR potential, and I'm facing an error code given below.

I already tried to fix it by recompiling the DeepMD and LAMMPS.
The dw model and energy model were retrained as well.

I used DeePMD-kit v2.1.1

Do you have any idea to fix this error?
Thanks for reading

Error code is presented below

Your simulation uses code contributions which should be cited:

• USER-DEEPMD package:
  The log file lists these citations in BibTeX format.

CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE

PPPM initialization ...
using 12-bit tables for long-range coulomb (../kspace.cpp:342)
G vector (1/distance) = 0.4
grid = 20 20 20
stencil order = 5
estimated absolute RMS force accuracy = 0.059750338
estimated relative force accuracy = 0.0041494313
using double precision MKL FFT
3d grid and FFT values/proc = 24389 8000
WARNING: Interaction between types 1 and 3 is set with deepmd, but will be ignored.
Deepmd model has only 2 types, it only computes the mulitbody interaction of types: 1-2. (../pair_deepmd.cpp:1077)
WARNING: Interaction between types 2 and 3 is set with deepmd, but will be ignored.
Deepmd model has only 2 types, it only computes the mulitbody interaction of types: 1-2. (../pair_deepmd.cpp:1077)
WARNING: Interaction between types 3 and 3 is set with deepmd, but will be ignored.
Deepmd model has only 2 types, it only computes the mulitbody interaction of types: 1-2. (../pair_deepmd.cpp:1077)
Generated 0 of 3 mixed pair_coeff terms from geometric mixing rule
Neighbor list info ...
update every 1 steps, delay 0 steps, check yes
max neighbors/atom: 2000, page size: 100000
master list distance cutoff = 8
ghost atom cutoff = 8
binsize = 4, bins = 3 3 3
1 neighbor lists, perpetual/occasional/extra = 1 0 0
(1) pair deepmd, perpetual
attributes: full, newton on
pair build: full/bin
stencil: full/bin/3d
bin: standard
Setting up Verlet run ...
Unit style : metal
Current step : 0
Time step : 0.001
INTERNAL: 2 root error(s) found.
(0) INTERNAL: Operation received an exception: DeePMD-kit Error: CUDA Assert, in file /home/tools/LAMMPS-DeePMD-OCTP/deepmd-kit-v2.1.4/source/op/custom_op.cc:17
[[{{node ProdEnvMatA}}]]
[[o_force/_29]]
(1) INTERNAL: Operation received an exception: DeePMD-kit Error: CUDA Assert, in file /home/tools/LAMMPS-DeePMD-OCTP/deepmd-kit-v2.1.4/source/op/custom_op.cc:17
[[{{node ProdEnvMatA}}]]
0 successful operations.
0 derived errors ignored.
ERROR: DeePMD-kit Error: TensorFlow Error: INTERNAL: 2 root error(s) found.
(0) INTERNAL: Operation received an exception: DeePMD-kit Error: CUDA Assert, in file /home/tools/LAMMPS-DeePMD-OCTP/deepmd-kit-v2.1.4/source/op/custom_op.cc:17
[[{node ProdEnvMatA}]]
[[o_force/_29]]
(1) INTERNAL: Operation received an exception: DeePMD-kit Error: CUDA Assert, in file /home/tools/LAMMPS-DeePMD-OCTP/deepmd-kit-v2.1.4/source/op/custom_op.cc:17
[[{node ProdEnvMatA}]]
0 successful operations.
0 derived errors ignored. (../pair_deepmd.cpp:390)
Last command: run ${step_total} upto

[goodluck-xyz]

I meet the same error. Have you solved this error?

[goodluck-xyz]

I haven't figure out what's wrong. After I load the deepmd module, the lammps run successfully. However, I ran another lmp progroam successfully without load deepmd module. If I compute msd and rdf , is loading deepmd module necessary?

[njzjz]

Usually TensorFlow will log the version and the location of CUDA libraries. Could you find it?

[goodluck-xyz]

error.log
you mean this?

[wxm0703]

I meet the same error too. Have you solved this error?

[goodluck-xyz]

Maybe mpiru -np 4 lmp_mpi -i input.lammps will help you.

[ChunyiZh]

You may also check to see whether this error is caused by the DPLR model or by the software by conducting the DPLR example calculation here: https://github.com/CSIprinceton/workshop-july-2022/tree/main/hands-on-sessions/day-2/5-deep-potential-long-range/3-md
It includes the DPLR model and the input script. If this simulation can be run correctly on your machine, there may be some problems related to your model or script.