Prototype lightweight datadependency package for Simulation.jl.
This package uses the PyCall package to wrap around the data_pipeline_api python package.
To do this, it is necessary to build the package before running.
This essentially builds PyCall against the private miniconda environment in Conda.jl
The remaining steps in build will instantiate the root Conda environment from the
python environment.yml
Table of datasets currently in use by Simulation.jl
Preliminary list of parameters/datasets.
| Name of parameter/ dataset | Description | Value | Source | Other info (e.g. stability) |
|---|---|---|---|---|
| p_s | Probability of developing symptoms | 0.96 | http://gabgoh.github.io/COVID/index.html (From Thibaud's original model) | |
| p_h | Probability of hospitalisation | 0.2 | Guess | |
| cfr_home | Case fatality ratio (at home) | 0.1 | Guess | |
| cfr_hospital | Case fatality ratio (at hospital) | 0.1 | Guess | |
| T_lat | Latent period | 5 days | http://gabgoh.github.io/COVID/index.html (From Thibaud's original model) | |
| T_asym | Asymptomatic period | 3 days | http://gabgoh.github.io/COVID/index.html (From Thibaud's original model) | |
| T_sym | Symptomatic period | 5 days | http://gabgoh.github.io/COVID/index.html (From Thibaud's original model) | |
| T_hosp | Hospitalisation period | 5 days | https://www.icnarc.org/Our-Audit/Audits/Cmp/Reports (From Thibaud's original model) | |
| T_rec | Recovery period | 11 days | http://gabgoh.github.io/COVID/index.html (From Thibaud's original model) | |
| mu_1 | Probability of becoming Asymptomatic | 1/T_lat | ||
| mu_2 | Probability of becoming Symptomatic | p_s * 1/T_asym | ||
| hospitalisation | Probability of becoming Hospitalised | p_h * 1/T_sym | ||
| sigma_1 | Probability of Recovery from Asymptomatic | (1 - p_s) * 1/T_asym | ||
| sigma_2 | Probability of Recovery from Symptomatic | (1 - p_h) * (1 - cfr_home) * 1/T_rec | ||
| sigma_hospital | Probability of Recovery from Hospital | (1 - cfr_hosp) * 1/T_hosp | ||
| death_home | Probability of Death at home | cfr_home * 2/T_hosp | ||
| death_hospital | Probability of Death at hospital | cfr_hosp * 1/T_hosp | ||
| ScotlandDensity2011 | Scottish population density at 1km grid | UK census 2011 - A Reeves 'Covid19-ScottishCensusData' repo | ||
| dispersal_dist | Average dispersal distance of virus per disease category | 2.0km per infectious disease category | Guess | Varies depending on grid size |
| mean_pref | Mean temperature preference of virus | 298K | Guess | Currently tuned to fit environment perfectly |
| var_pref | Temperature niche width of virus | 0.1K | Guess | Currently tuned to fit environment perfectly |
| birth | Probability of giving birth per individual | 1.3e-4/day (20-40 year olds), 0 otherwise | Guess | |
| death | Probability of giving natural mortality per individual | 2.7e-5/day | Guess | |
| virus_growth_asymp | Rate of generating virus per asymptomatic individual | 0.1/day | Guess | |
| virus_growth_symp | Rate of generating virus per symptomatic individual | 0.1/day | Guess | |
| beta_force | Force of infection | 10.0/day | Guess | |
| beta_env | Environmental transmission | 10.0/day | Guess |