Libardo1

library(httr)
library(ggplot2)

add_cat <- function(width = 400, height = 400){
  r <- GET(paste("http://theoldreader.com/kittens", width, height, sep = "/"))
  stop_for_status(r)
  img <- content(r)
  bw <- 0.2989*img[,,1] + 0.5870*img[,,2] + 0.1140*img[,,3]
  lighter <- bw + (0.7 * (1-bw))
  annotation_raster(lighter, xmin=-Inf, xmax=Inf, ymin=-Inf, ymax=Inf)

Libardo1

require(proto)

stat_qqline <- function (mapping = NULL, data = NULL, geom = "abline", position = "identity", 
distribution = qnorm, dparams = list(), na.rm = FALSE, ...) { 
  StatQqline$new(mapping = mapping, data = data, geom = geom, position = position, 
  distribution = distribution, dparams = dparams, na.rm = na.rm, ...)
}
  
StatQqline <- proto(ggplot2:::Stat, {
  objname <- "qqline"

Libardo1

library(plyr)
library(ggplot2)

# from: http://www.codeproject.com/Articles/18149/Draw-a-US-Flag-using-C-and-GDI
star_coords <- function(x, y, r, r1){
  a <- 72 * pi /180
  b <- a/2
  df <- data.frame(
    x = x + c(0, r1*sin(b), r*cos(pi/2 - a), r1*cos(a + b - pi/2), r*sin(b),
      0, -r*sin(b), -r1*cos(a + b - pi/2), - r*cos(pi/2 - a), -r1*sin(b)), 

Libardo1

fitted_variofit <- function (x, max.dist, scaled = FALSE, ...) {
  my.l <- list()
  if (missing(max.dist)) {
    my.l$max.dist <- x$max.dist
    if (is.null(my.l$max.dist)) 
      stop("argument max.dist needed for this object")
  }
  else my.l$max.dist <- max.dist
  if (any(x$cov.model == c("matern", "powered.exponential", 
                           "cauchy", "gencauchy", "gneiting.matern"))) 

Libardo1

library(shiny)
library(ggplot2)

# Define server logic required to generate and plot a random distribution
shinyServer(function(input, output) {
  generate_data <- reactive(
    function(){
      mu <- c(input$mu_1, input$mu_2, input$mu_3)
    tmp <- data.frame(len = rnorm(length(mu) * input$n, 
                                  mean = rep(mu, each = input$n), 

Libardo1

require(proto)

stat_qqline <- function (mapping = NULL, data = NULL, geom = "abline", position = "identity", 
distribution = qnorm, dparams = list(), na.rm = FALSE, ...) { 
  StatQqline$new(mapping = mapping, data = data, geom = geom, position = position, 
  distribution = distribution, dparams = dparams, na.rm = na.rm, ...)
}
  
StatQqline <- proto(ggplot2:::Stat, {
  objname <- "qqline"

Libardo1

#------------------------------------------------------------ 
# REVOLUTION ANALYTICS WEBINAR: INTRODUCTION TO R FOR DATA MINING
# February 14, 2013
# Joseph B. Rickert
# Technical Marketing Manager
#
# BIG DATA with RevoScaleR
#
# Copyright: Revolution Analytics

Libardo1

easterEgg.BadWorder.list={
"4r5e":1,
"5h1t":1,
"5hit":1,
a55:1,
anal:1,
anus:1,
ar5e:1,
arrse:1,
arse:1,

Libardo1

ngrams.tokenizer <- function(x, n = 2) {
  trim <- function(x) gsub("(^\\s+|\\s+$)", "", x)
  terms <- strsplit(trim(x), split = "\\s+")[[1]]
  ngrams <- vector()
  if (length(terms) >= n) {
    for (i in n:length(terms)) {
      ngram <- paste(terms[(i-n+1):i], collapse = " ")
      ngrams <- c(ngrams,ngram)
    }
  }

Libardo1

Markov Chains

A while back I wrote a blog post explaining Markov chains and demonstrating different ways of finding their steady-state distribution in R. Now, I want to play with Markov chains as a graph. I’m going to pull examples from around the internet and answer the same questions in Cypher as the authors do with matrices. This gives me the opportunity to explore more advanced Cypher queries while working with a topic I enjoy very much (stochastic processes and Markov chains). So this is officially just for funsies.

I found three Markov chains online that I’m going to showcase, and they involve the following topics: