Correction of QM Output

The QCORR module focuses on the analysis and correction of the output files of QM calculations. Here we refer to correction as:

  • Generate new inputs from calculations that have an error termination.

  • Generate new inputs for minima containing a small imaginary frequency

  • Ensure that all provided files have the same level of theory, grid size, program, version, etc.

The following scheme shows how QCORR works and how it sorts the calculations.

QCORR_scheme

Analyzing the output files

For these tasks we will be using the QCORR module --qcorr

Then we list the files that we want to analyze. In this case we are going to analyze Gaussian16 output files. We are going to assume that we have our files in the folder 'calculations' therefore --files "calculations/*.log"

Warning

Please notice that shell wildcard arguments need to be provided as strings. --files "calculations/*.sdf" should be provided instead of --files calculations/*.sdf. This feature might change in future to follow the usual conventions.

We can specify the --freq_conv "opt=(calcfc,maxstep=5)" which will attempt to fix calculations whose optimization ended normally but whose frequency calculation did not.

Finally we run the analysis of the files.

python -m aqme --qcorr --files "calculations/*.log"

Optionally we may indicate the extension of the initial input files --isom_type com as well as the folder where those files are --isom_inputs folder.

python -m aqme --qcorr --files "calculations/*.log" --isom_type com --isom_inputs folder

If we instead wanted to skip the checks and generate the .json files containing information about our calculations we can use the --fullcheck False keyword.

python -m aqme --qcorr --fullcheck False --files "calculations/*.log"