mrecos

## DR Set up
# predict on all 65M rows of DRPrimaryv4.csv
# using deployment [BOOKED predcitons] (634728e0ba6623c209179662)
# using model [eXtreme Gradient Boosted Trees Classifier with Early Stopping (learning rate =0.01) (Fast Feature Binning) M60 BP52] (63457d2fcc011ba8bee7ec78)
# from project Test1004 created by Adam

## Post-processing (R & SQL)
# load predictions (result-6356a08dabe595d288af9a3e.csv) into SQLite DB

sqlite_file <- "AMN.sqlite"

mrecos

library(readr)
library(tidyverse)
library(glue)
library(DataRobotColors)

pred <- read.csv("DATA/Predictions/[MDH]_DRPrimaryV3_10pcnt_SF_20221003_eXtreme_Gradient_Boosted_Trees_Classifier_with_Ear_(57)_79.93_IF_-_MDH_Reduced_v2_holdout.csv")

pred_rank <- pred %>% 
  group_by(TRAVELERID) %>% 
  arrange(TRAVELERID, desc(Cross.Validation.Prediction)) %>% 

mrecos

# google distance example
library(tidyverse)
library(googleway)

api_key <- "GET A GOOGLE API KEY"

## set up a data.frame of locations
## can also use 'lat/lon' coordinates as the origin/destination
df_locations <- data.frame(
  origin = c("Melbourne, Australia", "Sydney, Australia","Canberra, Australia")

mrecos

dat <- data.frame(Response = c("Yes","No"), Value = c(0.32,0.68))

ggplot(dat, aes(x = Response, y = Value)) +
  geom_bar(stat = "identity", width = 0.5) +
  xlab("") +
  ylab("") +
  theme_bw() +
  scale_y_continuous(limits = c(0,1)) +
  geom_text(aes(label = paste0(Value*100,"%")), vjust = -1, 
            hjust = 0.5, face = "bold", size = 6) +

mrecos

lancCounty <- st_read("C:/Users/matthew.d.harris/Documents/GitHub/Public-Policy-Analytics-Landing/DATA/Chapter2/LancasterCountyBoundary.geojson") %>%
  st_transform('ESRI:102728')

uga <- st_read("C:/Users/matthew.d.harris/Documents/GitHub/Public-Policy-Analytics-Landing/DATA/Chapter2/Urban_Growth_Boundary.geojson") %>% 
  st_transform('ESRI:102728')     

studyAreaTowns <- st_read("C:/Users/matthew.d.harris/Documents/GitHub/Public-Policy-Analytics-Landing/DATA/Chapter2/StudyAreaTowns.geojson") %>%
  st_transform('ESRI:102728')    

buildings <- st_read("C:/Users/matthew.d.harris/Documents/GitHub/Public-Policy-Analytics-Landing/DATA/Chapter2/LancasterCountyBuildings.geojson") %>% st_transform('ESRI:102728')    

mrecos

library(shiny)
library(shinydashboard)
# devtools::install_github("nik01010/dashboardthemes")
library(dashboardthemes)
library(tidyverse)
library(sf)
library(leaflet)
library(usmap) # us counties and states as table
library(leaflet)

mrecos

##### Your code leading up to this point ######

# initialize active cases
SWDetail$active_cases <- 0
SWDetail[1,"active_cases"] <- 1
# initialize most recent cases
SWDetail$last_case_given <- 0
SWDetail[1,"last_case_given"] <- 1
# resorting this by date
cps_test_assignment <- cps_test_assignment[order(cps_test_assignment$assignment_date),]

mrecos

#Package installs -------------------------------------------------------------
load.fun <- function(x) { 
  x <- as.character(x) 
  if(isTRUE(x %in% .packages(all.available=TRUE))) { 
    eval(parse(text=paste("require(", x, ")", sep=""))) 
    print(paste(c(x, " : already installed; requiring"), collapse=''))
  } else { 
    #update.packages()
    print(paste(c(x, " : not installed; installing"), collapse=''))
    eval(parse(text=paste("install.packages('", x, "')", sep=""))) 

mrecos

library(brms)
library(tidyverse)
library(caret)

rstan_options(auto_write=TRUE)
options(mc.cores=parallel::detectCores ()) # Run on multiple cores
set.seed(3875)

ir <- data.frame (scale (iris[, -5]), Species=iris[, 5])
system.time (b2 <- brm (Species ~ Petal.Length + Petal.Width + Sepal.Length + Sepal.Width,