This is just some simple code to write, run and analyze input files for HOOMD using GSD and Freud. Adding it to my profile so that I can update my local, laptop and cluster versions more easily.
These tasks are broken down accordingly:
Writing Files: Parameters: -x_a = particle fraction of particle whose activity is varied -Pe_a = activity of species being varied
1.)Enter the above parameters and how to change them (stepsize)
2.)Using these inputs, generate simulation aspect of infile
Note: HOOMD scripts allow all of this to be done in one file (so analysis will have to be edited as well) Note: you can access system values within a script (like particle number) via a snapshot
3.)Params cascade into analysis methods...
Note: How should we be saving this output (need hard data)
4.)Run infiles with SLURM manager (batch submission)
Jump for joy as your computer does your job :)