The Bucky model is a spatial SEIR model for simulating COVID-19 at the county level.
The Bucky model currently supports Linux and OSX and includes GPU support for accelerated modeling and processing.
git
must be installed and in your PATH.- GPU support requires a cupy-compatible CUDA installation. See the CuPy docs for details.
Standard installation:
pip install bucky-covid
Bucky will produce multiple folders for historical data and outputs. It's recommended to put these in their own directory, for example ~/bucky
BUCKY_DIR=~/bucky
mkdir $BUCKY_DIR
cd $BUCKY_DIR
The default configuration for bucky is located here. Currently, you can locally modify these options by creating a bucky.yml
in BUCKY_DIR
that will override any of the default options specified in it.
TODO this is WIP and does not work yet:
To use a customized configuration you first need to make a local copy of the bucky configuration. In your working directory:
bucky cfg install-local
To download the required input data to the data_dir
specified in the configuration files (default is $(pwd)/data
:
bucky data sync
To run the model with default settings and produce standard outputs.
bucky run
Equivalently, one can the following command (to provide cli configuration to each part of the process)
bucky run model
bucky run postprocess
bucky viz plot
Each bucky
command has options that can be detailed with the --help
flag. e.g.
$ bucky run model --help
Usage: bucky run model [OPTIONS]
`bucky run model`, run the model itself, dumping raw monte
carlo output to raw_output_dir.
Options:
-d INTEGER Number of days to project forward
[default: 30]
-s INTEGER Global PRNG seed [default: 42]
-n INTEGER Number of Monte Carlo iterations [default:
100]
--runid TEXT UUID name of current run [default:
2022-06-04__08_00_03]
--start-date TEXT Start date for the simulation. (YYYY-MM-DD)
--help Show this message and exit.
Further CLI documentation is available in the documentation.