USPEX restart not working after numGenerations has been reached

[Franziska Heß]

Hello! 

I have run into an issue that others have reported here as well, but I did not find a satisfactory answers. My USPEX calculation has reached the numGenerations, but while the results look promising, they are not yet satisfactory, so I tried to continue the calculation by increasing numGenerations and setting 25 : pickUpGen and 1 : pickUpFolder, which were the folders used in the last step. I also removed the file USPEX_IS_DONE.

However, after resubmitting the job, the file USPEX_IS_DONE immediately reappears, a folder results2 is not created and the job stops after that. In the file 'log', the following output is written:

Wed Jul 28 16:03:55 CEST 2021
/bin/bash: synclient: command not found
*********************************************************
*                                                       *
  _|    _|     _|_|_|   _|_|_|     _|_|_|_|   _|      _|
  _|    _|   _|         _|    _|   _|           _|  _|   
  _|    _|     _|_|     _|_|_|     _|_|_|         _|     
  _|    _|         _|   _|         _|           _|  _|   
    _|_|     _|_|_|     _|         _|_|_|_|   _|      _|
*                                                       *
** USPEX v10.5.0                         Oganov's Lab! **
*********************************************************
 
Warning: File 'POSCAR' not found.
> In USPEXDone (line 4)
  In Start (line 72)
  In USPEX (line 39)
Warning: File 'POSCAR_order' not found.
> In USPEXDone (line 5)
  In Start (line 72)
  In USPEX (line 39)
USPEX IS DONE!
 

I am unsure which POSCAR files it is looking for or what these error messages mean, but apparently it does not realize that I changed numGenerations. Here is my complete input file after increasing numGenerations:

PARAMETERS EVOLUTIONARY ALGORITHM
******************************************
*      TYPE OF RUN AND SYSTEM            *
******************************************
USPEX : calculationMethod (USPEX, VCNEB, META)
300   : calculationType (dimension: 0-3; molecule: 0/1; varcomp: 0/1)
1     : AutoFrac

% optType
1
% EndOptType

% atomType
Ru O H
% EndAtomType

% numSpecies
5 10 5
% EndNumSpecies
******************************************
*               POPULATION               *
******************************************
20    : populationSize (how many individuals per generation)
30    : initialPopSize
100    : numGenerations (how many generations shall be calculated)
15     : stopCrit
0     : reoptOld
0.6   : bestFrac
25 : pickUpGen
1 : pickUpFolder
******************************************
*          VARIATION OPERATORS           *
******************************************
0.50  : fracGene (fraction of generation produced by heredity)
0.30  : fracRand (fraction of generation produced randomly from space groups)
0.20  : fracAtomsMut (fraction of the generation produced by softmutation)
0.00  : fracLatMut (fraction of the generation produced by softmutation)
0.00  : fracPerm
*****************************************
*   DETAILS OF AB INITIO CALCULATIONS   *
*****************************************
% abinitioCode
1 1 1 1  
% ENDabinit

% KresolStart
0.13 0.10 0.07 0.05
% Kresolend

% commandExecutable
mpirun -np 8 ~/bin/vasp.md
% EndExecutable

1       : whichCluster (0: no-job-script, 1: local submission, 2: remote submission)
5      : numParallelCalcs
0.00001 : ExternalPressure

How can I make it continue the calculation in this situation?

Thank you for your time!

[Franziska Heß]

So in case anyone else comes here with the same problem, it appears that you have to run

USPEX --clean

before restarting the job. This information not mentioned in the manual https://uspex-team.org/online_utilities/uspex_manual_release/EnglishVersion/uspex_manual_english/input_restart.html.