In leerichardson/spew: SPEW Framework for Generating Synthetic Ecosystems

#!/bin/env/Rscript
args <- commandArgs(TRUE)
input_dir <- args[1]
input_dir <- paste0(input_dir, "/")
#print(input_dir)
library(spew)
#TODO change this line
#source("~/spew/R/diagnostics.R")
#source("~/spew/R/plot.R")
library(ggplot2)
library(ggmap)
library(data.table)

# Data directory
data_dir <- file.path(input_dir, "../eco")
image_dir <- file.path(input_dir, "../diags")

r toupper(basename(input_dir))

Maps

if( grepl("OUTPUT_", toupper(basename(input_dir)))){
    region_name <- toupper(gsub("output_", "",
    basename(input_dir)))
} else{
    region_name <- toupper(basename(input_dir))
}

This is r region_name with a subsample of the households displayed as points.

# Plot the region

# Get the correct reigon file from the data folder
files <- list.files(data_dir)
hh_files <- files[grepl("household", files)]
region_file <- hh_files[grepl(region_name, toupper(hh_files))]
stopifnot(length(region_file) == 1)
region_path <- file.path(data_dir, region_file)

# Plot the total region colored with different people
spew:::plot_region(region_path, map_title = region_name, savePlot = TRUE, K = 10^5, zoom = 3)

library(png)
library(grid)
image_path <- file.path(image_dir, paste0(region_name, ".png"))
#print(image_path)
img <- readPNG(image_path)
 grid.raster(img)

Household Population

# Household Pop
filenames <- list.files(file.path(input_dir, "eco"))
full_paths <- file.path(input_dir, "eco", filenames)
full_paths <- full_paths[grepl("household", full_paths)]
big_list <- lapply(full_paths, function(path) spew:::summary_diags2(type="hh", path))
#spew:::summary_diags2(type="hh", path))

summary_df <- do.call("rbind", lapply(big_list, "[[", 1))

There are r prettyNum(sum(summary_df$nRecords), big.mark=",") households in r toupper(basename(input_dir)).

library(knitr)
kable(summary_df, col.names=c("Region Name", "Number of Households"))

library(ggplot2)
colnames(summary_df) <- c("Region", "nRecords")
rose <- ggplot(summary_df, aes(x=Region, y=nRecords, fill=Region)) + geom_bar(stat="identity") + ggtitle("Household Counts")
rose <- rose + theme(axis.line=element_blank(),axis.text.x=element_blank(),
          axis.text.y=element_blank(),
          axis.title.x=element_blank(),
          axis.title.y=element_blank(),panel.border=element_blank())
print(rose)

Rose chart of households The longer the radius, the more households in that region.

There are r length(big_list[[1]]$all_names) attributes to the household population with the following names. Please see the quickstart guide to find the meanings and codes for the attributes.

print(big_list[[1]]$all_names)

Person Population

# Person Pop
filenames <- list.files(file.path(input_dir, "eco"))
full_paths <- file.path(input_dir, "eco", filenames)
full_paths <- full_paths[grepl("people", full_paths)]
big_list <- lapply(full_paths, 
    function(path) spew:::summary_diags2(type="p", path)) #spew:::summary_diags(type="hh", path))


summary_df <- do.call("rbind", lapply(big_list, "[[", 1))
featuresList <- do.call(rbind, lapply(lapply(big_list, "[[", 2), '[[', 1))
colnames(summary_df) <- c("Region", "nRecords")

df <- cbind(summary_df, featuresList)
colnames(df)[3:4] <- c("Males", "Females")

There are r prettyNum(sum(summary_df$nRecords), big.mark=",") people in r toupper(basename(input_dir)).

library(knitr)
kable(df, col.names=c("Region Name", "Number of Households", "Males", "Females"))

featuresList <- do.call(rbind, lapply(lapply(big_list, "[[", 2), '[[', 1))
summary_df <- do.call("rbind", lapply(big_list, "[[", 1))
colnames(summary_df) <- c("Region", "nRecords")


library(reshape2)

df <- cbind(summary_df, featuresList)
colnames(df)[3:4] <- c("Males", "Females")
mfratio <- df[,3]/df[,2]
fmratio <- df[,4]/df[,2]
df <- cbind(summary_df, mfratio, fmratio)[,-2]
colnames(df)[2:3] <- c("Male", "Female")
df.melt <- melt(df[, c(1,3,2)], id.vars="Region", varnames=c("Male", "Female"))
colnames(df.melt)[2:3] <- c("Sex", "Percentage")
p <- ggplot(data=df.melt, aes(x=Region, y=Percentage, fill=Sex)) + geom_bar(stat="identity") + coord_flip() + ggtitle("Ratio of Males to Females")
print(p)

There are r length(big_list[[1]]$all_names) attributes to the person population with the following names. Please see the quickstart guide to find the meanings and codes for the attributes.

print(big_list[[1]]$all_names)

Generation Information

This report was generated on r Sys.time() by spew, an R package used to generate populations throughout the world. Please see our spew Github repo and our previously generated regions at epimodels.org. We are a part of the Infomatics Service Group MIDAS branch at Carnegie Mellon University and University of Pittsburgh and are supported by 1 U24 GM110707-01 NIH/NIGMS grant. Please send your comments and suggestions to [e-mail].

leerichardson/spew documentation built on May 21, 2019, 1:39 a.m.