Some problems when using USPEX2022.1.1 & USPEX10.5

团山通讯

unread,

Jan 9, 2023, 9:21:14 PM1/9/23

to USPEX

I am a beginner in USPEX and want to use USPEX2022.1.1 for small molecule crystal structure prediction. VASP and LAMMPS have been used for structure relaxation and energy calculations. Though corresponding examples (EX01-3D_Si_vasp & EX04-3D_C_lammps) can work successfully, I have met different kinds of problems in small molecule crystal structure prediction (I tried methane).

When using with VASP, USPEX run successfully for the first generation. All random structures can be successfully relaxed on the first stage (as in EX01-3D_Si_vasp, 6 stages of relaxation were kept), but the work terminated in “Looking for duplicates.” (See attached files)

When using with LAMMPS, USPEX even cannot generate the first generation of random structures with information “ValueError: could not broadcast input array from shape (6,) into shape (7,)”. (See attached files)

I have attached all related files (including input.uspex, log, but there is no OUTPUT in the simulation with LAMMPS) and looking forward to any kinds of answers.

By the way, I have tried USPEX10.5. It seems that I have successfully installed it because there were no problems reported. On the contrary, it told me that:

USPEX is installed successfully!

You can check USPEX by "USPEX -h"

EnjoY USPEX!

But afterwards, I met problems when using USPEX10.5 no matter with VASP or without any external codes (EX13-3D_special_quasirandom_structure_TiCoO). The following is my screen copy when performing EX13:

(base) txj@ubuntu:~/EX13$ USPEX -r &

[1] 2113

(base) txj@ubuntu:~/EX13$ Submission dir copied to the current directory.

/bin/bash: synclient: command not found

*********************************************************

* *

_| _| _|_|_| _|_|_| _|_|_|_| _| _|

_| _| _| _| _| _| _| _|

_| _| _|_| _|_|_| _|_|_| _|

_| _| _| _| _| _| _|

_|_| _|_|_| _| _|_|_|_| _| _|

* *

** USPEX v10.5.0 Oganov's Lab! **

*********************************************************

Read Seeds ...

seed number 1 has been successfully added

End of pickup Seeds

Read AntiSeeds ...

Structure1 step1 at CalcFold1

Relaxation is done.

Error using python_uspex (line 91)

System error: Traceback (most recent call last):

File "/home/txj/USPEX105/application/archive/src/FunctionFolder/spacegroup/spacegroup.py", line 3, in <module>

import spglib

ModuleNotFoundError: No module named 'spglib'

Command executed: python3 -W ignore /home/txj/USPEX105/application/archive/src/FunctionFolder/spacegroup/spacegroup.py structure.json

Error in anasym_stokes (line 55)

Error in WriteIndividualOutput_300 (line 21)

Error in ReadJobs (line 105)

Error in LocalRelaxation (line 52)

Error in EA_300 (line 13)

Error in Start (line 64)

Error in USPEX (line 39)

MATLAB:python:ExecutionError

[1]+ Done USPEX -r

Any answer will be appreciated.

Thanks a lot.

CH4-LAMMPS.rar

CH4-VASP.rar

Pavel Bushlanov

unread,

Jan 12, 2023, 7:55:45 AM1/12/23

to USPEX

Please try adding blank line to the end of MOL file.

团山通讯

unread,

Jan 12, 2023, 10:35:10 PM1/12/23

to USPEX

Dear Pavel,

Thank you very much for your suggestion.

I have performed several times according to your suggestion and some other modifications. When using with VASP, USPEX gave the same error as before. But when using with LAMMPS, USPEX gave different error information (still cannot generate the first generation of random structures) as below:

2023-01-13 10:59:04,929 - numexpr.utils - INFO - NumExpr defaulting to 4 threads.

2023-01-13 10:59:34,558 - USPEX.Target - INFO - Following utilities was not initialized: ['PowderSpectrumAnalyzer', 'SingleCrystalSpectrumAnalyzer'].

2023-01-13 10:59:34,559 - USPEX.Target - INFO - Permutation does not work when number of symbols in calculation is 1.

2023-01-13 10:59:34,560 - USPEX.Target - INFO - AddAtom does not currently work in molecular regime.

2023-01-13 10:59:34,560 - USPEX.Target - INFO - RemoveAtom does not currently work in molecular regime.

2023-01-13 10:59:34,560 - USPEX.Target - INFO - TeleportAtom does not currently work in molecular regime.

2023-01-13 10:59:34,560 - USPEX.Target - INFO - RandSymPyXtal does not currently work in molecular regime.

2023-01-13 10:59:34,561 - USPEX.main - ERROR -

Traceback (most recent call last):

File "/home/txj/anaconda3/lib/python3.8/site-packages/USPEX/main.py", line 107, in _run

asyncio.get_event_loop().run_until_complete(GenerationController.createController().run())

File "/home/txj/anaconda3/lib/python3.8/site-packages/USPEX/GenerationController.py", line 89, in createController

stages.append(GenerationController.knownStages['execute'](**stage))

File "/home/txj/anaconda3/lib/python3.8/site-packages/USPEX/Stages/Executor.py", line 58, in __init__

self._interface = self.knownInterfaces[type](tag, **kwargs)

File "/home/txj/anaconda3/lib/python3.8/site-packages/USPEX/Stages/Interfaces/LAMMPS_Interface.py", line 64, in __init__

assert all([os.path.exists(lib) for lib in libs])

AssertionError

That’s all.

Thanks again for your kindly replies.

Pavel Bushlanov

unread,

Jan 16, 2023, 7:21:05 AM1/16/23

to USPEX

Looks like you did not put proper specific files for lammps. Look into uspex-lammps example.

团山通讯

unread,

Jan 21, 2023, 5:12:01 AM1/21/23

to USPEX

Dear Pavel,

Thank you very much for your suggestion.

You are right. I had missed the Tersoff potential file SiC.tersoff, because I wanted to use the cvff potential. Unfortunately, I had not wrote the potential correctly.

After realizing my problems at your reminding, I have rewrote the input.uspex and potential files. USPEX run can now successfully generate random structures, but failed in relaxation. I have not succeeded in this respect until now.

Would you please to give me an example input.uspex file and potential file that can be used for molecular crystal structure prediction containing C H O N elements? My email address: t...@qlu.edu.cn.