More complete documentation to follow, though the public API is already documented in the doc
folder.
Running Thyr:
- Install Torch7
- Make sure you have a C++14 compatible compiler
- Modify
compile.sh
to work with your environment - Run
compile.sh
- Clone Plyght into Thyr's folder, making sure you have the Python pre-requisites (numpy, matplotlib, pyqt5 (Mac only)).
- Inspect and/or modify the
ThyrDriver.lua
file (should be well commented) - Create a directory with the same name as the
prefix
variable inThyrDriver.lua
in the working directory (e.g.c7DataHR
in the default case). - Run Plyght (i.e.
python3 Plyght/plyght.py
) and leave in running in background - Run the simulation
th ThyrDriver.lua
(after whatever variant oftorch_activate
is required on your system) - Post-processing of the simulations from the backed up "datacubes" is done using
th ThyrPostProcess.lua
. Setprefix
inThyrPostProcess.lua
to the same string as was used inThyrDriver.lua
.
NB: The default simulation in ThyrDriver.lua is currently very high resolution and will take a long time to compute. To get quick results from Thyr I would recommend numVox = 32
with a refinement factor of 2 in Thyr.create_high_res_footpoints
. To obtain results quickly it is also a good idea to reduce the number of frequencies being computed, especially the higher frequencies.
All files under MIT License (c) 2015-2018 Christopher Osborne, University of Glasgow, with the exception of maf.lua
(c) Bjorn Swenson (MIT License) and the excellent Penlight libraries (c) Steve Donovan (MIT License).