mcdc-convergence-test 0.0.3

Tests certain mcdc functionality convergence to semi-analytical solutions provided by William Bennett

mcdc_convergence_tests

Runs mcdc on given tests and plots convergence to benchmark data

Dependencies

requires numpy,hdf5,and mcdc

Installation guide

Download the src folder

Quick start guide

The most user-friendly way of running the tests is as follows:

python -c 'import run; run.test([particlenums],"sourcetype",createdata,loud)' --mode=numba

where particlenums are the amounts of particles you'd like to use (e.g. [1e2,1e3,1e4,1e5]), and sourcetype is one of

"plane_IC"
"square_IC"
"square_source"
"gaussian_IC"
"gaussian_source"
"all"

createdata allows you to specify whether you want to run the mcdc simulations again, or use existing data (this is best used when you have run it with createdata = True once, then want to reuse that data without taking the time to compute it) loud determines whether convergence is determined and plotted --mode=numba makes computation much quicker, though this can be omitted and mcdc will run in normal python mode

IMPORTANT NOTE:

With the current implementation of MC/DC, the inclusion of --mode=numba is the only way to run in numba mode. This poses a problem when trying to run multiple simluations from one script, and so we recommend refraining from using the all option to create data, instead specifying each source. The all command can still be used to plot all the data with no problems.