This code includes models for the reduced and enriched Zika models. The parameters of the discrepancy model are calibrated using QUESO.
Dependencies: QUESO (https://github.com/libqueso/queso), Eigen
My Makefile is included for reference.
The steps to run the code are as follows:
rm -r outputData
./bin/zika_ip inputs/mhInput.inp
Remove or move existing 'outputData' directory. This directory is created during the inverse problem. 'mhInput.inp' is an input file for QUESO.
To plot the time series results:
cd postprocessing
./post_proc.sh
python3 compute-percents.py
python3 percent-time-series.py
You can also run
./bin/gen_data
to test the forward model alone.
Notes:
You can ignore 'americo' and 'data' directories.
'rep_factor' is set to 1 within src/compute.cpp, and must be changed by hand with a recompile if needed.
In future versions, it should be set as an input that propagates to post-processing, but this is not implemented yet.
To cite this code, you can use:
@misc{morrison2020zikacode,
doi = {10.5281/ZENODO.3666845},
url = {https://zenodo.org/record/3666845},
author = {Morrison, Rebecca E.},
title = {rebeccaem/zika: Initial release},
publisher = {Zenodo},
version = {0.1.0},
year = {2020}
}