edoaltamura / run-from-spreadsheet

Generate a set of SWIFT parameter files and submission scripts from a spreadsheet

General functionalities

Generates runs based on a spreadsheet provided, with individual directories.

The spreadsheet should be organised as follows:

Run ID	EagleSection:Param	EagleSection:Param2	...	Comment
Hello	0.1	0.2	...	Good run!
...	...	...	...	...

This will then generate directories for each of the Run IDs, changing the selected parameters to the new values based on a 'base' parameter file.

You can also generate submission scripts, with a template that uses the following special parameters

• $SWIFT_PATH - the path to the SWIFT executable
• $RUN_ID - ID for the run
• $PARAMETER_FILE - the parameter file name

Don't forget to change the directory paths in your parameter file to include a relative change to one directory higher (e.g. ../output_list.txt).

Finally, you can create a submission script to run VELOCIraptor on your outputs in postprocessing by using the create_vr_slurm_scripts.py file. Note that this assumes that you run it from the same directory as the others (i.e. one above the Run ID directories that create_parameter_files.py created).

Calibration of zoom simulations

This repository, forked from Josh's pipeline for calibrating EAGLE-XL periodic cosmological boxes, is adapted to be used for setting up zoom simulations. The listing functionality of the spreadsheet allows to run a whole sample of group or clusters with different sets of subgrid parameters. Although runs at different resolutions could be gathered in the same .csv file, we do recommend the use of separate files, since the set-up pipeline would need to handle significantly different sets of parameters (e.g. particle softenings, particle splitting threshold and other resolution-dependent quantities).

The spreadsheet includes explicit specification of the initial conditions file of the individual run, in order to allow for different objects in the sample to appear in the same spreadsheet.

Once the spreadsheet is selected, make sure to look at the launch_setup.sh script and edit the directory where you wish to set-up the runs. After that, just run the script as

source launch_setup.sh

Note: although the parameter file handling can be adapted for dark-matter-only simulations, the launch_setup.sh is optimized for hydro runs. Set-up directories for DMO simulations can instead be easily generated using the pipeline for initial conditions.

Reference model

The runs are compared to a reference subgrid model, commonly identified as Ref. This calibration set of simulations is created such as the sample is firstly run with the Ref model, in order to then explore departures from it.