Installing LAMMPS

[JayKayNJIT]

Hi, I have successfully gone through the training phase but I am getting stuck at LAMMPS running phase. The first instruction under build-in mode at this page[1], you are saying go to directory "cd $deepmd_source_dir/source/build". But I don't have build directory there, my source directory looks like [2].

Thanks for the code.

[1]
https://github.com/deepmodeling/deepmd-kit/blob/master/doc/install/install-lammps.md

[2]
login-1-224 source >: pwd
[...]/DeePMDkit/deepmd-kit-2.1.0/source
login-1-225 source >: ls
3rdparty/ api_cc/ cmake/ CMakeLists.txt config/ gmx/ install/ ipi/ lib/ lmp/ md/ op/ tests/

[AnguseZhang]

Do you have to install DeePMD-kit's C++ interface from scratch? You can refer to https://docs.deepmodeling.com/projects/deepmd/en/master/getting-started/install.html#install-off-line-packages and https://docs.deepmodeling.com/projects/deepmd/en/master/getting-started/install.html#install-with-conda for an easy installation mode. An executable lmp which is patched with DeePMD-kit will also be installed.

[JayKayNJIT]

Thanks, installation worked fine. But the LAMMPS job is exiting with an error[1], and log file[2].

[1]
login-1-53 LAMMPS_JOB >: conda activate deepmd
(deepmd) login-1-54 LAMMPS_JOB >: ls
frame0.data graph-compress.pb lammps.in log.lammps
(deepmd) login-1-55 LAMMPS_JOB >: lmp -in lammps.in
LAMMPS (29 Sep 2021 - Update 3)
OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98)
using 1 OpenMP thread(s) per MPI task
Loading plugin: deepmd pair style v2.0 by Han Wang
Loading plugin: compute deeptensor/atom v2.0 by Han Wang
Loading plugin: fix dplr v2.0 by Han Wang
Reading data file ...
orthogonal box = (0.0000000 0.0000000 0.0000000) to (31.916280 33.156000 23.204787)
1 by 1 by 1 MPI processor grid
reading atoms ...
267 atoms
read_data CPU = 0.002 seconds
251 atoms in group membrane
16 atoms in group DNA
Summary of lammps deepmd module ...

Info of deepmd-kit:
installed to: /home/j/jk435/miniconda3/envs/deepmd
source: v2.1.1
source branch: HEAD
source commit: c4f0cec
source commit at: 2022-04-16 11:11:16 +0800
surpport model ver.:1.1
build float prec: double
build with tf inc: /home/j/jk435/miniconda3/envs/deepmd/include;/home/j/jk435/miniconda3/envs/deepmd/include
build with tf lib: /home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_cc.so;/home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_framework.so
set tf intra_op_parallelism_threads: 0
set tf inter_op_parallelism_threads: 0
Info of lammps module:
use deepmd-kit at: /home/j/jk435/miniconda3/envs/deepmd2022-05-16 14:42:27.484085: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: SSE4.1 SSE4.2 AVX
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2022-05-16 14:42:27.500263: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance.
Info of model(s):
using 1 model(s): graph-compress.pb
rcut in model: 6
ntypes in model: 6
Displacing atoms ...
Summary of lammps deepmd module ...
Info of deepmd-kit:
installed to: /home/j/jk435/miniconda3/envs/deepmd
source: v2.1.1
source branch: HEAD
source commit: c4f0cec
source commit at: 2022-04-16 11:11:16 +0800
surpport model ver.:1.1
build float prec: double
build with tf inc: /home/j/jk435/miniconda3/envs/deepmd/include;/home/j/jk435/miniconda3/envs/deepmd/include
build with tf lib: /home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_cc.so;/home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_framework.so
set tf intra_op_parallelism_threads: 0
set tf inter_op_parallelism_threads: 0
Info of lammps module:
use deepmd-kit at: /home/j/jk435/miniconda3/envs/deepmdINVALID_ARGUMENT: Tensor model_attr/output_dim:0, specified in either feed_devices or fetch_devices was not found in the Graph
terminate called after throwing an instance of 'deepmd::tf_exception'
what(): DeePMD-kit Error: TensorFlow Error: INVALID_ARGUMENT: Tensor model_attr/output_dim:0, specified in either feed_devices or fetch_devices was not found in the Graph
Aborted
(deepmd) login-1-56 LAMMPS_JOB >: ls
frame0.data graph-compress.pb lammps.in log.lammps

[2]
LAMMPS (29 Sep 2021 - Update 3)
OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98)
using 1 OpenMP thread(s) per MPI task
plugin load libdeepmd_lmp.so
Loading plugin: deepmd pair style v2.0 by Han Wang
Loading plugin: compute deeptensor/atom v2.0 by Han Wang
Loading plugin: fix dplr v2.0 by Han Wang
units metal
boundary p p p
atom_style atomic

#read_data 23_12.data # 1st Mo, 2nd S
#read_data DFTrelaxed.data extra/atom/types 4 group membrane
#read_data Conformer3D_CID_135398634_Guanine.data offset 2 0 0 0 0 add append group DNA
read_data frame0.data
Reading data file ...
orthogonal box = (0.0000000 0.0000000 0.0000000) to (31.916280 33.156000 23.204787)
1 by 1 by 1 MPI processor grid
reading atoms ...
267 atoms
read_data CPU = 0.002 seconds
group membrane type 1 2
251 atoms in group membrane
group DNA type 3 4 5 6
16 atoms in group DNA

mass 1 95.95 # Mo
mass 2 32.065 # S
mass 3 15.999 # O
mass 4 14.0067 # N
mass 5 12.0107 # C
mass 6 1.00784 # H

pair_style deepmd graph-compress.pb
Summary of lammps deepmd module ...

Info of deepmd-kit:
installed to: /home/j/jk435/miniconda3/envs/deepmd
source: v2.1.1
source branch: HEAD
source commit: c4f0cec
source commit at: 2022-04-16 11:11:16 +0800
surpport model ver.:1.1
build float prec: double
build with tf inc: /home/j/jk435/miniconda3/envs/deepmd/include;/home/j/jk435/miniconda3/envs/deepmd/include
build with tf lib: /home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_cc.so;/home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_framework.so
set tf intra_op_parallelism_threads: 0
set tf inter_op_parallelism_threads: 0
Info of lammps module:
use deepmd-kit at: /home/j/jk435/miniconda3/envs/deepmdpair_coeff * *

#displacing DNA Nucleobases and generating frames
#write_dump all cfg frame0.data mass type xs ys zs modify element Mo S O N C H
#write_dump all custom frame0.data id type x y z
label loop
variable steps equal round($(lz)/0.5)
variable steps equal round(23.204787034399998902/0.5)
variable a loop ${steps}
variable a loop 46
displace_atoms DNA move 0 0 0.5
Displacing atoms ...
#write_dump all cfg frame${a}.data mass type xs ys zs modify element Mo S O N C H
compute dipole all deeptensor/atom graph-compress.pb
Summary of lammps deepmd module ...

Info of deepmd-kit:
installed to: /home/j/jk435/miniconda3/envs/deepmd
source: v2.1.1
source branch: HEAD
source commit: c4f0cec
source commit at: 2022-04-16 11:11:16 +0800
surpport model ver.:1.1
build float prec: double
build with tf inc: /home/j/jk435/miniconda3/envs/deepmd/include;/home/j/jk435/miniconda3/envs/deepmd/include
build with tf lib: /home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_cc.so;/home/j/jk435/miniconda3/envs/deepmd/lib/libtensorflow_framework.so
set tf intra_op_parallelism_threads: 0
set tf inter_op_parallelism_threads: 0
Info of lammps module:
use deepmd-kit at: /home/j/jk435/miniconda3/envs/deepmd

[njzjz]

I don't find any error message..

[JayKayNJIT]

I am sorry, I edited the original post which includes the error message now. Thanks.

[njzjz]

compute dipole all deeptensor/atom graph-compress.pb

Looks like you are trying to load an energy model to compute the dipole. Instead, you need to train a dipole model. There is an example. See also the doc.

[JayKayNJIT]

Perfect, it's working now. Thanks a lot!