Environment setup

Install The Julia Programming Language, make sure julia is added to PATH.
In the julialang REPL, run

using Pkg;Pkg.add(["CSV","DataFrames","PGFPlotsX","ProgressMeter"]

To compile the tex into svg plots, need full LaTeX installation in PATH.

Instructions

The code is an implementation of sampling the positive steady state with constant $\beta$, $\mu$, and two-compartment asymmetric interaction kernel $k$. It tries to vary different $\beta$ and $\mu$, and determine if overcompensation exists.

In the shell environment (Powershell/Cmd/Bash), run

julia age-structure-run.jl "Profile Name" [overwrite]

It will read the profile with the name Profile Name.jl under the directory profile/.
In the profile, one specifies the necessary arguments needed for calculation, and provide them in the array of commands cmds. It also provides a string label and the method for storing data.
- A sample profile general.jl is given as follows.
- To run it, use julia age-structure-run.jl general.

βs = [i/10 for i in 1:20]
μs = [i/10 for i in 1:0.5:10]
ks = [1]
label = "general"
cmds = []
for β in βs, μ in μs, k in ks
    cmd = `julia age-structure-base.jl $(β) $(μ) $(k)`
    push!(cmds,cmd)
end
if isfile("fig/$(label).csv") && !overwrite
    println("data exists, skipping calculation...")
    df = CSV.read("fig/$(label).csv",DataFrame)
else
    println("starting calculation...")
    @showprogress pmap(run,cmds)
    println("calculation completed, saving data...")
    df = DataFrame(
        β_0 = Float64[],
        μ_0 = Float64[],
        k_0 = Float64[],
        N = Float64[],
        T = Float64[],
        ∂ᵤN = Float64[],
        ∂ᵤlnN = Float64[],
    )
    for f in readdir("./data/",join=true)
        temp_df = CSV.read(f,DataFrame)
        for i in 1:length(temp_df[:,1])
            push!(df,temp_df[i,:])
        end
    end
    sort!(df,[:β_0,:μ_0])
    CSV.write("fig/$(label).csv",df)
    for f in readdir("./data/",join=true)
        rm(f)
    end
end
println("creating TeX figures...")
include("../age-structure-plot-bm.jl")
println("done.")

age-structure-utils.jl contains the general implementation of functions.
age-structure-base.jl receives the commands and run the calculation.
- It also handles the parameters passed by commandline arguments.
- To use different types of dynamics, please change the way $\beta$, $\mu$, and $k$ are defined through parameters.

Multithreading

By default, in age-structure-run.jl, it uses 8 processors by addprocs(8). Change 8 to another positive integer according to your needs and your machine.

hsianktin/age-structure

Environment setup

Instructions

Multithreading