trinker

# R Function to Draw Edward Tufte-style Slopeplots

# TODO:
## deal with overlapping numeric values: if more than one label falls at a position, offset based on rank() of the subsequent column values for those rows
## overlapping rownames (same algorithm as above, but hard code for 1st and 2nd value)


slopegraph <-
function(
df,

trinker

pstat <- require(pacman)
bostat <- require(BreakoutDetection)
if (!pstat) devtools::install_github("trinker/pacman")
if (!bostat) devtools::install_github("twitter/BreakoutDetection")
p_load(BreakoutDetection, dplyr, XML, qdap)

doc <- htmlParse("http://dasonk.com/2014/07/18/Lil-Jon-Analysis/")
text <- read.transcript(text = head(sapply(getNodeSet(doc, "//p"), xmlValue), -1))
sents <- sent_detect(text[[1]])

trinker

A function developed by Bryan Goodrich for reading in .docx files:

```{r}
read_docx <- function (file, skip = 0) {
    tmp <- tempfile()
    if (!dir.create(tmp)) 
        stop("Temporary directory could not be established.")
    unzip(file, exdir = tmp)
    xmlfile <- file.path(tmp, "word", "document.xml")
    doc <- XML::xmlTreeParse(xmlfile, useInternalNodes = TRUE)

trinker

# Makefile to use knitr for presentations

SLIDES := Day1Part1-Introduction.html Day1Part1-session2.html Day1Part1-session3.html Day1Part2-session1.Rmd

all: $(SLIDES)

clean:
  rm -rf *.md mplus.*

%.html: %.Rmd

trinker

function(package, name, qpath, path, ...){
    message(sprintf("Attempting to tweak %s...", package))

    if (TRUE)  message(sprintf("%s has been tweaked!", package))
}

trinker

if (!require("pacman")) install.packages("pacman")
pacman::p_load(babynames, dplyr)

lapply(seq(1, 50000, by=35), function(n){

    babynames %>%
        filter(year > 1940 & year < 1980) %>%
        group_by(name, sex)  %>% 
        summarize(n=sum(n))%>% 
        ungroup %>%

trinker

if (!require("pacman")) install.packages("pacman"); library(pacman)
p_load(dplyr, qdap, qdapRegex, qdapDictionaries, ggplot2, scales)

nms <- unlist(qdapRegex::TC(sample(NAMES[[1]], 26)))
dat <- data_frame(
    year = rep(1997:2000, each = length(nms)),
    observation = rep(nms, length(1997:2000)),
    score = sample(50:100, length(nms)*length(1997:2000), TRUE)
)

trinker

L1 <- list(
    a = c(2, 15, 23, 19, 3, 2, 3, 27, 20, 11, 27, 10, 19, 10, 13, 10),
    b1 = c(22, 9, 5, 10, 5, 1, 24, 2, 10, 9, 7, 3, 12, 24, 10, 9)
)
L2 <- list(
    i = c(16, NA, 17, NA, 2, NA, 2, NA, 10, NA, 15, NA, 6, NA, 9, NA),
    b2 = c(11, 27, 14, 5, 5, 7, 8, 24, 8, 3, 6, 15, 22, 6, 1, 1)
)
#L3 <- list( #bad wheel
#    j = c(3, NA, 8, NA, 10, NA, 14, NA, 11, NA, 8, NA, 12, NA, 11, NA),

trinker

load(url("http://dl.dropbox.com/u/61803503/nycounty.RData"))
head(ny); head(key)  #view the data set from my drop box

if (!require("pacman")) install.packages("pacman")
p_install_version("ggtree", '1.0.14')
p_load(ggplot2, ggtree, dplyr, tidyr, sp, maps, pipeR, grid, XML, qdapRegex, magrittr)

getLabelPoint <- function(county) {Polygon(county[c('long', 'lat')])@labpt}

df <- map_data('county', 'new york')                 # NY region county data

trinker

#' Convert the output of a topicmodels Latent Dirichlet Allocation to JSON
#' for use with LDAvis
#'
#' @param fitted Output from a topicmodels \code{LDA} model.
#' @param corpus Corpus object used to create the document term
#' matrix for the \code{LDA} model. This should have been create with
#' the tm package's \code{Corpus} function.
#' @param doc_term The document term matrix used in the \code{LDA}
#' model. This should have been created with the tm package's 
#' \code{DocumentTermMatrix} function.