arvi1000

library(leaflet)
library(readxl) # for read_excel()
library(ggmap) # for geocode()
library(fields) # for rdist.earth()

### read data from excel ----
# download to default working directory
download.file('https://dl.dropboxusercontent.com/u/12146012/place_names.xlsx',
              destfile='place_names.xlsx')
places <- read_excel('place_names.xlsx')

arvi1000

# This is a quasi replication of the correlation matrix heatmap viz from Seaborn,
# using R and ggplot
# see: http://stanford.edu/~mwaskom/software/seaborn/examples/structured_heatmap.html

library(data.table)
library(ggplot2)

# get data
brain_url <- paste0('https://raw.githubusercontent.com/',
                    'mwaskom/seaborn-data/master/brain_networks.csv')

arvi1000

library(ggplot2)

# random data
set.seed(5)
my_data <- data.frame(y = rnorm(15*5, mean=0, sd=2) + 3,
                      x = rep(letters[1:5], each=5))

# plot: jittered points + boxplot overlay
ggplot(my_data, aes(x=x, y=y)) +
  geom_jitter(aes(color=x, shape=x), width = 0.2) +

arvi1000

# sometimes you just want to visualize a quantity; this shows a given number of dots

library(ggplot2)
library(magrittr)

dot_count <- 300
size <- ceiling(dot_count^.5)

data.frame(x = rep(1:size, each=size),
           y = rep(size:1, size),

arvi1000

# you must: source('z_load_data.R')
stopifnot(exists('dat'))

library(tigris)
library(leaflet)

# load congressional districts
cong <- tigris::congressional_districts(cb=TRUE)

# add state and district number in form of (AK-1)

arvi1000

# I love data.table, but I find the syntax for "mutating" columns a little clunky for such a common task
# I wonder if it would be useful to have a setcols() function? 
# Takes advantage of data.table's pass-by-reference

library(data.table)
setcols <- function(my_dt, cols, my_fun) {
  # validate inputs
  stopifnot('data.table' %in% class(my_dt),
            'character' == class(cols),
            'function' == class(my_fun))

arvi1000

# Batch download BARUG docs, so as to move from Meetup to Github

library(rvest)

# 1) build dataframe of download targets, with url and date ----

# fn to get a page of docs from a given offset
scrapeBarug <- function(offset) {
  # get url for a page containing table of BARUG docs
  # (web shows 5 pages @ offsets 0, 25, 50, 75, 100)

arvi1000

library(lubridate)
library(tidyverse)
# data ----
full_dat <- data.frame(
  dt = mdy(c('9/19/16', '2/27/17', '4/9/18', '9/17/18', '2/25/19', '8/5/2019',
             '1/13/2020', '6/22/2020')),
  num = c(2188,1833,1302,1171,1067,988, 892, 831)
)
obs <- nrow(full_dat)
holdout_days <- 2

arvi1000

library(png)
library(tidyverse)

# 1) read in data ----

to_matrix <- function(png_file) {
  
  # read PNG to raster  (3d array of x * y * color channel)
  the_png <- readPNG(png_file)

arvi1000

library(tidyverse)

# read data
dat_raw <- read_csv('http://infographics.economist.com/databank/Economist_women-research.csv')

# drop stuff that's not the data
dat <- dat_raw[2:13,]

# fix names
names(dat) <-