raise RuntimeError("Failed to retrieve any thermo_style-output")

[YukeLi-chem]

No description provided.

[YuuuXie]

Hi @YukeLi-chem , this might be a result from ase parser's failure to read the lammps log file from the latest lammps version. What you can do is to revise ase's parser code. We have been using the bash commands as below:

ase_file="$(dirname $(python3 -c 'import ase; print(ase.__file__)'))/calculators/lammpsrun.py"
sed -i 's/line.startswith(_custom_thermo_mark)/line.strip\(\).startswith\("Step"\)/g' $ase_file

[Teng-Martin-Ma]

Hi @YukeLi-chem , this might be a result from ase parser's failure to read the lammps log file from the latest lammps version. What you can do is to revise ase's parser code. We have been using the bash commands as below:

ase_file="$(dirname $(python3 -c 'import ase; print(ase.__file__)'))/calculators/lammpsrun.py"
sed -i 's/line.startswith(_custom_thermo_mark)/line.strip\(\).startswith\("Step"\)/g' $ase_file

This solution exactly works and no retrivement issue anymore. However, the check_mgp_match will assert error due to the large forces discrepancy between gp and lammps.

[YuuuXie]

Hi @Teng-Martin-Ma, here are a few suggestions you could try:

1. do not use restart or warm start, try a fresh start with all previous output files deleted.
2. set export OMP_NUM_THREADS=<number_of_cpus_on_a_single_node> (from @YukeLi-chem)