deepmodeling
[Feature Request]_deepmd Pair style incompatible with Dihedral charmm _
Summary
The error: ERROR: Dihedral charmm is incompatible with Pair style (src/MOLECULE/dihedral_charmm.cpp:381) when trying to run pair style deepmd with a structure that utilizes CHARMM dihedral and angle coefficientes.
Detailed Description
DeePMD-kit Version
DeePMD-kit v2.1.1
TensorFlow Version
tensorflow-2.9.0
Python Version, CUDA Version, GCC Version, LAMMPS Version, etc
No response
Details
Hello,
I am using the latest version of deepmd-kit pulled from conda. so DeePMD-kit v2.1.1.
I get the following error when using LAMMPS:
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 = (-32.521667 -32.521667 -32.521667) to (0.50000000 32.521667 0.50000000)
1 by 1 by 1 MPI processor grid
reading atoms ...
6035 atoms
scanning bonds ...
4 = max bonds/atom
scanning angles ...
14 = max angles/atom
scanning dihedrals ...
34 = max dihedrals/atom
reading bonds ...
6110 bonds
reading angles ...
10327 angles
reading dihedrals ...
15761 dihedrals
Finding 1-2 1-3 1-4 neighbors ...
special bond factors lj: 0 0 0
special bond factors coul: 0 0 0
4 = max # of 1-2 neighbors
8 = max # of 1-3 neighbors
18 = max # of 1-4 neighbors
22 = max # of special neighbors
special bonds CPU = 0.003 seconds
read_data CPU = 0.036 seconds
Summary of lammps deepmd module ...
>>> Info of deepmd-kit:
installed to: /home/ykr/miniconda3/envs/deepmd_cpu
source: v2.1.1-1-gdff6aa33
source branch: HEAD
source commit: dff6aa33
source commit at: 2022-05-21 20:41:28 -0400
surpport model ver.:1.1
build float prec: double
build with tf inc: /home/ykr/miniconda3/envs/deepmd_cpu/include;/home/ykr/miniconda3/envs/deepmd_cpu/include
build with tf lib: /home/ykr/miniconda3/envs/deepmd_cpu/lib/libtensorflow_cc.so;/home/ykr/miniconda3/envs/deepmd_cpu/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/ykr/miniconda3/envs/deepmd_cpu2022-05-25 11:22:26.176922: I tensorflow/core/platform/cpu_feature_guard.cc:193] 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 AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2022-05-25 11:22:26.180734: 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.
2022-05-25 11:22:26.382716: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:354] MLIR V1 optimization pass is not enabled
>>> Info of model(s):
using 1 model(s): aPS_from_20mono75chains_no_dt_more_data_5mil_compressed.pb
rcut in model: 8
ntypes in model: 2
CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
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
ERROR: Dihedral charmm is incompatible with Pair style (src/MOLECULE/dihedral_charmm.cpp:381)
Last command: run 10000000
Is there a way we could get deepmd pair style to play nice with charmm dihedral and angle coefficients. I ask because I am using CHARMM GUI Polymer builder and the structure created has those coefficients. and I am not sure on how to convert them to for example harmonic, which I have verified plays nice with deepmd pair style.
Further Information, Files, and Links
No response
I have read https://docs.lammps.org/dihedral_charmm.html. It looks that this style requires the potential to be pair wise. However, deep potential is a many-body potential.
