datalove

For the code below, if we want to select records have a field with colour = Red and length > 30...

<root>
<record id="1">
  <field id="colour" value="Red" />
  <field id="size"   value="Small" />
  <field id="length" value=25 />
</record>

datalove

library(XML)
library(ggplot2)

url  <- "http://www.kaggle.com/c/acquire-valued-shoppers-challenge/leaderboard"
tree <- htmlTreeParse(url)
tbl  <- readHTMLTable(pagetree, stringsAsFactors = FALSE)[[1]]

colnames(tbl) <- gsub("[^a-zA-Z0-9#]","", colnames(tbl))
tbl$Score     <- as.numeric(tbl$Score)

datalove

# calculation as per:
# http://www.bmj.com/content/343/bmj.d2304

get_odds_p <- function(est, cil, ciu) {  
  se <- (log(ciu) - log(cil))/(2*1.96)
  z  <- abs(log(est)/se)  
  p  <- exp(-0.717*z - 0.416*z^2)
  p
}

datalove

rm(list=ls()) 

rcode <- "x <- 1+1; y <- 2+2"

# running `eval` evaluates `rcode` into the parent environment
eval(parse(text = rcode))

print(x)
print(y)

datalove

# Use autodetected proxy settings
setInternet2(TRUE)

# Get the SpotfireSPK package from the Spotfire Stats Server
install.packages("SpotfireSPK", repos = "http://MySpotfireServer:8080/SplusServer/update/TERR")

# Get a package you want to deploy from CRAN
install.packages("nortest", repos = "http://cran.us.r-project.org")

# Create a Debian Control File

datalove

TBD

datalove

#######################################################
#
# Expression Functions in Spotfire can take arbitrarily
# many columns as input. Columns will be passed to TERR
# in order as 'input1', 'input2', etc. 
# 
# This shows how to capture an arbitrary number of 
# columns and to put them into a data frame.
#
#######################################################

datalove

###################################################################
# Takes an arbitrarily long list of input columns and returns a 
# boolean indicating whether or not each row is an outlier.
###################################################################

# create vector of inputs
inputs <- grep("^input[0-9]+$",ls(), value = TRUE)

# capture columns as a matrix
x <- sapply(inputs, function(y) {eval(parse(text = y))})

datalove

########################################################
# Takes an arbitrarily long list of input columns and 
# returns a boolean indicating whether or not each row 
# is an outlier. 
#
# The function uses the critical value for Mahalanobis 
# Distance calculated from an upper tailed ChiSq 
# distribution with p=0.001.
########################################################

datalove

decode <- function(x, ...) {

  odds <- function(x) { unlist(x[1:length(x) %% 2 == 1][1:floor(length(x)/2)]) }
  even <- function(x) { unlist(x[1:length(x) %% 2 == 0]) }
  last <- function(x) { unlist(if(length(x) %% 2 == 1) tail(x,1)) }
  
  interpret_args <- function(x) { if(is.call(x)) {eval(x)} else if(is.name(x)) {as.character(x)} else {x} }
  
  args <- eval(substitute(alist(...)))
  args <- lapply(args, interpret_args)