create script for generating postprocessed truth data

And bundle into single script for DRY

Author: SamuelBrand1

pipeline/scripts/create_postprocessing_dataframes.jl

@@ -0,0 +1,14 @@
+using EpiAwarePipeline, EpiAware, JLD2, DrWatson, DataFramesMeta, CSV
+
+## Define scenarios
+scenarios = ["measures_outbreak", "smooth_outbreak", "smooth_endemic", "rough_endemic"]
+
+if !isfile(plotsdir("plotting_data/predictions.csv"))
+    @info "Prediction dataframe does not exist, generating now"
+    include("create_prediction_dataframe.jl")
+end
+
+if !isfile(plotsdir("plotting_data/truthdata.csv"))
+    @info "Truth dataframe does not exist, generating now"
+    include("create_truth_dataframe.jl")
+end

pipeline/scripts/create_prediction_dataframe.jl

@@ -1,9 +1,3 @@
-using EpiAwarePipeline, EpiAware, AlgebraOfGraphics, JLD2, DrWatson, DataFramesMeta,
-      Statistics, Distributions, DrWatson, CSV
-
-## Define scenarios
-scenarios = ["measures_outbreak", "smooth_outbreak", "smooth_endemic", "rough_endemic"]
-
 ## Define true GI means
 true_gi_means = [2.0, 10.0, 20.0]
 

pipeline/scripts/create_truth_dataframe.jl

@@ -0,0 +1,11 @@
+truth_df = mapreduce(vcat, scenarios) do scenario
+    truth_data_files = readdir(datadir("truth_data")) |>
+                       strs -> filter(s -> occursin("jld2", s), strs) |>
+                               strs -> filter(s -> occursin(scenario, s), strs)
+    mapreduce(vcat, truth_data_files) do filename
+        D = load(joinpath(datadir("truth_data"), filename))
+        make_truthdata_dataframe(D, scenario)
+    end
+end
+
+CSV.write(plotsdir("plotting_data/truthdata.csv"), truth_df)