facet « WordPress.com Tag Feed

christmas time cz.3 - szkoła przetrwania dla Niego

evi219 — Mon, 23 Nov 2009 16:57:45 +0000

Wyobraź sobie, że siedzisz przy wigilijnym stole. Na twarzy masz uśmiech no.5 przeznaczony specjalnie na takie okazje – dobry uśmiech do złej gry. Z każdą godziną krew w Twoich żyłach coraz bardziej się gotuje. Toczysz walkę z samym sobą. Czy dalej będziesz siedzieć spokojnie tłumiąc w sobie gniew i złość, czy uderzysz ręką w stół i wyjdziesz, kończąc tym samym całą tą farsę?

Oczywiście zostajesz. Tak jak co roku.

W kieszeni otwiera Ci się scyzoryk. Ty zaś ze stoickim spokojem dalej wysłuchujesz narzekań na polityków, historie przebytych chorób, żenujące opowiastki z twojego dzieciństwa, litanie kąśliwych uwag na swój temat oraz krzyki dzieci biegających wokół stołu… W głębi serca obiecujesz sobie, że to już ostatni raz… Za rok to nie będzie tak wyglądać, bo wyjedziesz gdzieś daleko stąd i będziesz mieć wreszcie święty spokój.

Jak przetrwać święta bez konieczności popełnienia krwawej zbrodni…? Panowie, mam dla Was kilka propozycji:

1. Oczywiście najprostszym rozwiązaniem było by uciec na drugi koniec świata. Najlepiej gdzieś, gdzie nikt nie słyszał o istnieniu Świąt Bożego Narodzenia. No ale to niestety graniczyłoby z cudem…

2. Niech Święta będą dla Ciebie nowym wyzwaniem. Kolejnym etapem twojej ulubionej gry komputerowej, który po prostu musisz przejść dalej cokolwiek by się nie działo!

3. Popatrz na to co się dzieje dookoła Ciebie jak na kabaret. Z tą tylko jedyną różnicą, że nie oglądasz tego skeczu z widowni, ale przez przypadek znalazłeś się na scenie.

4. Bądź ponad wszystko. Bądź dumny i wyniosły. Nie daj sobie w kaszę dmuchać. I spoglądaj na innych z politowaniem. Gwarantuję, że szybko dadzą Ci spokój.

5. Potraktuj święta jako test własnej wytrzymałości psychicznej. Wyobraź sobie, jak wielka będzie twoja satysfakcja, że kolejne święta masz za sobą. Będziesz z siebie dumny.

6. Jeśli jednak na własnej skórze przekonałeś się, że akurat w twoim przypadku powyższe metody nie zdały egzaminu, masz jeszcze jedno wyjście… Upij się. Zachlaj do nieprzytomności. Wtedy będzie Ci już wszystko jedno. Nie będziesz się niczym przejmować, nie będziesz mieć złych wspomnień – pomijając ewentualnego kaca jaki Cię dopadnie następnego dnia. Jedynym twoim zmartwieniem później będzie zastanawianie się dlaczego żona kazała Ci spać na kanapie, siostra od tygodni nie odbiera twoich telefonów, a kuzynka mijając Cię na ulicy odwraca wzrok…

P.S. Stosujecie powyższe rady tylko i wyłącznie na własną odpowiedzialność. Reklamacji i zażaleń nie przyjmuję!

Merry Christmas!

Sebastian Boenisch w reprezentacji?

printex — Mon, 09 Nov 2009 12:16:28 +0000

Wszystko wskazuje na to, że jednak Sebastian Boenisch zagra dla reprezentacji Polski w piłce nożnej

Managing and growing a cultural heritage web presence

Mike — Fri, 06 Nov 2009 12:24:13 +0000

I’m absolutely delighted (and only slightly scared) to announce that I’ve been commissioned to write a book for Facet Publishing.

Ever since I started working with museums online, I’ve felt that there is a need for strategic advice to help managers of cultural heritage web presences. There are of course hundreds of thousands of resources if you’ve got technical questions, but not many places where you can ask things like “how should I build my web team and structure my budget?” or “how do I write a strategy or business plan?”.

Facet approached me in July asking whether I’d be interested in authoring something for them, and this seemed like the ideal opportunity to try and answer some of these questions.

My (draft) synposis is as follows:

This book will provide a guide for anyone looking to build or maintain a cultural heritage web presence. It will aim to cater both to those who are single-handedly trying to keep their site running on limited budget and time as well as those who have big teams, large budgets and time to spend.

As well as describing the strategic approaches which are required to develop a successful online presence, the book will contain data and case studies on current practice from large and small cultural heritage institutions. This research will help give the reader an insight into how these institutions manage their websites as well as providing hints and tips on best practice. It will have an accompanying web presence which will provide template downloads and other up-to-date information including links and white papers.

As you’ll see, I have no intention of trying to do this all by myself – over the coming year I’m going to be on the phone to many of you (hide now!) asking how you do what you do, and compiling this into what I hope will be a useful guide.

If you have any ideas about what I should include, or the questions I should be asking – please do get in touch either via this blog or on Twitter at @m1ke_ellis!

robię dobrze facetom

majhu — Wed, 28 Oct 2009 11:36:37 +0000

Jestem laska, co dobrze robi facetom, biorę za to trzy stówy, ekskluzywna dziwka ze mnię i bzykam sie tylko w hotelach. Zadnych moteli i innych gównianych miejsc. CHcesz bzyknąć ekskluzywną dziwkę? masz 300zł za godzinę? dawaj, czekam na Ciebie, bo wbrew pozorom niewielu jest na mnie stać. Paranoiku jebany, chodź, przelecę Cię tak że sperma Ci tyłkiem wyleci. Dawaj!

Łukasz Rząsa

Tomek — Tue, 13 Oct 2009 08:55:18 +0000

Facet Design and Packaging

ambalaj — Sun, 04 Oct 2009 18:52:05 +0000

Dzmitry Samal has made the faceted tin concept on the first picture. And with this concept I would like to bring you into what I think is a fabulous document. This Facet Design Book is a fantastic source of inspiration. Scroll through it and you will probably end up having a bunch of ideas. Beautiful! On the same theme Vitamin Water – new label, with the facet design incorporated in the graphic design. Designed by The Coolhunter. The salt and pepper dispenser, designed by Gregory Buntain. And last Fract Engineered Flowers, by Nimrod Gavish.

Telegraf semtymentalny

dzokero — Sun, 27 Sep 2009 08:31:43 +0000

Na studiach wracało sie z sobotnich imprez przeważnie o 2-3. W dzisiejszych czasach 23 i już w domu

Roll Resistant Coke Can Concept

cokeforbreakfast — Sat, 26 Sep 2009 13:03:07 +0000

faceted rather than round. interesting….

Roll Resistant Coke Can Concept via PSFK.

uwodzic.wordpress - Uwodzenie pięknych kobiet na randce.

pawlooo84 — Thu, 24 Sep 2009 22:37:15 +0000

W jaki sposób uwodzić piękne kobiety oraz dziewczyny, wydające się wysoce niedostępne, co nie koniec

ggplot2: Back-to-back Bar Charts

learnr — Wed, 23 Sep 2009 22:50:20 +0000

On the ggplot2 mailing-list the following question was asked:

How to create a back-to-back bar chart with ggplot2?

For anyone who don’t know what I am talking about, have a look on a recent paper from the EU. I’d like to create plots like the graphs 5,6,18 in the paper.

An example graph from the above report is below:

Let’s create the same graph in ggplot2.

I was not able to find the exact dataset used to plot the graph above, and used instead the Eurostat “EU27 trade by BEC product group since 1999″ dataset which has a very similar data structure.

Access the subset used in this post in here.

> library(ggplot2)

> trade <- read.csv("trade.csv", header = TRUE,
     stringsAsFactors = FALSE)

Calculate monthly total trade balances.

> balance <- ddply(trade, .(Time), summarise,
     balance = sum(EXP - IMP))

Convert data from wide format to long format for plotting.

> trade.m <- melt(trade, id.vars = c("BEC", "Time"))

> ggplot(trade.m, aes(Time)) + geom_bar(subset = .(variable ==
     "EXP"), aes(y = value, fill = BEC), stat = "identity") +
     geom_bar(subset = .(variable == "IMP"),
         aes(y = -value, fill = BEC), stat = "identity") +
     xlab("") + scale_y_continuous("Export  -  Import",
     formatter = "comma")

Add horizontal line at 0 for easier reading, plot the trade balance line and make the x-axis labels more readable.

> labels <- gsub("20([0-9]{2})M([0-9]{2})", "\\2\n\\1",
     trade.m$Time)
> last_plot() + geom_line(data = balance, aes(Time,
     balance, group = 1), size = 1) + geom_hline(yintercept = 0,
     colour = "grey90") + scale_x_discrete(labels = labels)

The legend explanation (too large to fit on the plot):

CAP Capital goods
CNS Consumption goods
CTR Consumption goods plus motor spirit and passenger motor cars
INT Intermediate goods

There shouldn’t be any negative numbers on the y-axis, we need a custom formatter to convert these to positive.

> commapos <- function(x, ...) {
     format(abs(x), big.mark = ",", trim = TRUE,
         scientific = FALSE, ...)
 }

> last_plot() + scale_y_continuous(formatter = "commapos")

I have to agree with some of the posters on the mailing list, that this type of chart is not very easy to follow – it is impossible to compare individual values. Another useful way of showing information would be to use facets:

> ggplot(trade.m, aes(Time)) + geom_bar(aes(y = value),
     stat = "identity") + facet_grid(variable ~
     BEC) + scale_x_discrete("", labels = labels) +
     scale_y_continuous(formatter = "comma")

Or combine export and import values on one plot showing the comparative trend between the different goods categories.

> ggplot(trade.m, aes(Time, colour = variable,
     group = variable)) + geom_line(aes(y = value)) +
     facet_grid(BEC ~ .) + scale_x_discrete("",
     labels = labels) + scale_y_continuous(formatter = "comma")

The same graph with varying scales across panels.

> last_plot() + facet_grid(BEC ~ ., scales = "free_y")

Update: As suggested in the comments it would be good to have total values added to the last plot.

In order to do so the totals are first calculated and then appended to the original dataframe.

> total <- ddply(trade.m, .(Time, variable),
     summarise, value = sum(value), BEC = "Total")
> total <- rbind(trade.m, total)

Plotting the new dataframe is just a question of changing the source of the last plot.

> last_plot() %+% total

uwodzic.wordpress - Jakich mężczyzn naprawdę lubią kobiety?

pawlooo84 — Tue, 22 Sep 2009 23:30:07 +0000

Obiegowa opinia głosząca, że kobiety kochają tylko dobrze zbudowanych mężczyzn z okazałym portfelem,

A Guide to Effective Witnessing:

jbutzu — Mon, 21 Sep 2009 02:24:28 +0000

Here is a quick list that I hope will at least be considered by those who wish God to use them in th

Logistics: Currency Rate of Exchange

Asif Mir — Wed, 09 Sep 2009 02:39:54 +0000

Variations in the rate of exchange of the dollar against foreign currencies represent a difference of major importance in international logistics just as they do in any other facet of international operations. The design and operation of an international logistic system is based on a set of units of measure. Thus, as rates of exchange vary, so do the local costs associated with different local currencies; these include transportation, warehousing, inventorying, information processing, and other costs elements.

When variations in rates of exchange become substantial in magnitude, and significantly different between countries, it is usually necessary to re-evaluate the structure of the logistic system, to ensure that it is adjusted to keep it close to its optimal performance. By the same token, management reports must provide explicit identification of variations in costs due to variations in currency rates of exchange, as opposed to variations in performance.

My Consultancy–Asif J. Mir - Management Consultant–transforms organizations where people have the freedom to be creative, a place that brings out the best in everybody–an open, fair place where people have a sense that what they do matters. For details please visit www.asifjmir.com, Line of Sight

Facets

quock — Tue, 08 Sep 2009 23:25:58 +0000

What are actually facets and can they be seen as structured features for bio-models? What is facet classification and who uses it?

Gołe dupy

artmaster2 — Mon, 07 Sep 2009 16:58:48 +0000

Jestem ciekaw dlaczego na kobiety nie raz zwykło się słyszeć określenie, ale dupa. Przecież dupa wiadomo do czego służy. Może facetom po prostu chodzi o to, że ma fjane pośladki. W końcu kształt pośladków ma to do siebie, że Nas przyciagą, a w języku natury oznacza to dokładniej, jestem gotowa do kopulacji.

Ale to tyle jeśli chodzi o język natury. Niemiej bardziej zachęcające jest określenie gołe dupy. Co to by miało oznaczać, sam dokładnie nie wiem. Ale właśnie takich wyrażeń według Google wyszukują niektórzy użytkownicy tej wyszukiwarki internetowej. Przypuszczam, że mają tu na myśli rozebrane panieniki, które np. dają dupy na filmach porno.

NEWS FLASH, NEWS FLASH!!!!

skydi — Mon, 07 Sep 2009 16:48:05 +0000

WOW! 2PLUS7 has passed! $500,000 mark for commission payouts. $520,000 in just 2 MONTHS! This is the
“Incredible System Never Sleeps”. We are getting 400 signups a day! Now can you agree with me that we have something VERY SPECIAL for the MASSES!

Here are some short videos to help you understand what 2plus7 is all about. Enjoy and then Contact Me Today! Don’t Wait! …Diana

Benefits, Benefits, Benefits Overview

Security in this Recession Downturn

Turn Your Life Around in 90 Seconds

How It Works

viewfromtop@gmail.com

Wallpaper* (Thai Edition) - Nov, 2008

cheffykung — Thu, 27 Aug 2009 10:24:13 +0000

Ciągi i pociągi.

autoteliczna — Wed, 26 Aug 2009 23:24:48 +0000

Nie wiem jak to się stało, ale w końcu zaczęliśmy się spotykać. Spotykać na zasadzie, że jesteśmy razem. Zawsze o tym marzyłam. Tego pragnęłam w głębi serca najbardziej na świecie. Marzenie w końcu się spełniło. Po długim czasie oczekiwania, zwykłej przyjaźni oboje uzmysłowiliśmy sobie, że nie możemy bez siebie żyć. Bez naszych rozmów, naszej obecności. Że nie ma drugiej takiej osoby, która tak dobrze by nas rozumiała, przy której nie musieliśmy udawać.

Spotykaliśmy się już jakiś czas i lubiliśmy kochać się w różnych dziwnych miejscach. Było zawsze przy tym tyle zabawy. I ta adrenalina, że ktoś nas zobaczy. Zwykła sypialnia zostawała na te noce, kiedy nocowaliśmy u moich lub jego rodziców. Pewnego letniego popołudnia nie mieliśmy co robić, więc wsiedliśmy do mojego pikaczento i pojechaliśmy się przejechać. Koniec końców wylądowaliśmy na dworcu kolejowym. Jak zwykle naszła nas ochota na drobne igraszki. Odwzorowaliśmy scenę z jednego z polskich seriali komediowych (bodajże “Chłopaki nie płaczą”), w którym facet brutalnie bierze dziewczynę od tyłu na masce samochodu. W tej roli wystąpił wtedy Michał Milowicz. Miałam na sobie getry i mini. Po chwili getrów już nie miałam, a Paweł siłował się z gumką. Nie zajęło mu to dużo czasu i moment później już czułam jak we mnie wnika. Bez gry wstępnej. A po co. Gra wstępna była w samochodzie, kiedy ja prowadziłam, a on całował mnie po szyi, schodząc coraz niżej, a moja ręka, zamiast na dźwignię zmiany biegów, wędrowała od zewnętrznej strony jego kolan ku strefie, do której tylko ja miałam dostęp od jakiegoś, dłuższego już czasu. To wystarczyło. Było cudownie.

Kiedy oboje już ochłonęliśmy, postanowiliśmy się jeszcze trochę pobawić. Jako, że od dłuższego czasu nie mieliśmy okazji odwiedzić tutejszego dworca, spędziliśmy tam trochę czasu. Chodziliśmy po ruinach dawnego dworca. Tam gdzie dawniej były kasy biletowe, teraz rósł sobie krzaczek z takimi ciemnymi owocami, podobnymi do jagód. Nie ryzykowaliśmy i nie próbowaliśmy zjeść tych owoców.

Stał tam też pociąg. Kilka wagonów. Pięć, może sześć. Nie zwracaliśmy na to uwagi. Stał tak dosyć długo, a do ostatniego wagonu były przyczepione schodki, po których można było wejść na jego dach. A skoro tak lubiliśmy zabawy, postanowiliśmy wejść na dach jednego z wagonów pociągu i tam się kochać. Jakie było nasze zaskoczenie, kiedy w momencie, gdy On wdrapał się już na samą górę, pociąg ruszył. Ja zostałam na ziemi.

W pierwszej chwili mnie wmurowało i nie wiedziałam co zrobić. Pociąg dopiero nabierał prędkości, więc dość szybko zorientowałam się co się dzieje i zaczęłam biec. Chciałam wyprzedzić pociąg i przekazać jakoś maszyniście, żeby zatrzymał pociąg. On leżał bezładnie na dachu. Niestety nie udało mi się to co miałam w planach i rozpędzona już maszyna dość szybko zniknęła za zakrętem.

Zostałam sama na peronie nie wiedząc co robić. Łzy płynęły mi ciurkiem po policzkach. Próbowałam coś krzyczeć, ale nic to nie dało. Wsiadłam szybko do mojego pikaczento i ile fabryka dała pędziłam do najbliższej stacji, jednak kiedy tam dotarłam zobaczyłam, tylko, że pociąg pojechał dalej. Pośpieszny. Zrobiłam to samo. Jeśli ktoś twierdził, że takim samochodem nie da się zamknąć licznika to muszę go wyprowadzić z błędu, bo grubo się mylił. Dojechałam do Piotrkowa. O dziwo, udało mi się dotrzeć na dworzec przed pociągiem. Nie zamknąwszy nawet samochodu, pobiegłam do budki informacyjnej… krzyknęłam, że ten pociąg musi się tutaj zatrzymać, po czym wybiegłam szybko na tory. Za mną wybiegło dwóch jakichś facetów. Jeden był ubrany w mundur, ale raczej nie był policjantem. Próbowali ściągnąć mnie z torów. Jednak ja w całej swojej desperacja, pozwoliłam się tylko przesunąć o metr czy dwa, po czym ich podrapałam i dali mi spokój. Miałam wrażenie, że mijają godziny. Lecz po chwili moim oczom ukazały się światła pociągu. Widziałam już, że zwalnia. Poczułam, że adrenalina, której poziom był już krytycznie wysoki, odpływa gdzieś.

Maszyna zatrzymała się około 10 metrów przede mną. Zeszłam na bok. Po chwili dołączył do mnie trzęsący się jeszcze On. Przytulił się i staliśmy tam tak jeszcze dobrych kilka minut. Ludzie dziwnie się na nas patrzyli, ale kto by zwracał na to uwagę. Nadal przytuleni ruszyliśmy do mojego samochodu. Do naszego samochodu. Wróciliśmy do domu. To popołudnie i ten wieczór długo będziemy pamiętać. I już nigdy nie będziemy próbować wejść na dach pociągu. Jeśli w ogóle jeszcze kiedyś do takiego się któreś z nas zbliży.

ggplot2 Version of Figures in "Lattice: Multivariate Data Visualization with R" (Part 12)

learnr — Tue, 18 Aug 2009 12:48:59 +0000

This is the 12th post in a series attempting to recreate the figures in Lattice: Multivariate Data Visualization with R (R code available here) with ggplot2.

Previous parts in this series: Part 1, Part 2, Part 3, Part 4, Part 5, Part 6, Part 7, Part 8, Part 9, Part 10, Part 11.

Chapter 13 – Advanced Panel Functions

Topics covered:

Built-in panel and accessors functions
Examples

Figure 13.1

> library(lattice)
> library(ggplot2)

> grid <- data.frame(p = 11:30, q = 10)
> grid$k <- with(grid, factor(p/q))

> panel.hypotrochoid <- function(r, d, cycles = 10, density = 30) {
+     if (missing(r))
+         r <- runif(1, 0.25, 0.75)
+     if (missing(d))
+         d <- runif(1, 0.25 * r, r)
+     t <- 2 * pi * seq(0, cycles, by = 1/density)
+     x <- (1 - r) * cos(t) + d * cos((1 - r) * t/r)
+     y <- (1 - r) * sin(t) - d * sin((1 - r) * t/r)
+     panel.lines(x, y)
+ }
> panel.hypocycloid <- function(x, y, cycles = x, density = 30) {
+     panel.hypotrochoid(r = x/y, d = x/y, cycles = cycles,
+         density = density)
+ }
> prepanel.hypocycloid <- function(x, y) {
+     list(xlim = c(-1, 1), ylim = c(-1, 1))
+ }

lattice

> pl <- xyplot(p ~ q | k, grid, aspect = 1, scales = list(draw = FALSE),
+     prepanel = prepanel.hypocycloid, panel = panel.hypocycloid)
> print(pl)

ggplot2

> panel.hypotrochoid.gg <- function(r, d, cycles = 10,
+     density = 30) {
+     if (missing(r))
+         r <- runif(1, 0.25, 0.75)
+     if (missing(d))
+         d <- runif(1, 0.25 * r, r)
+     t <- 2 * pi * seq(0, cycles, by = 1/density)
+     x <- (1 - r) * cos(t) + d * cos((1 - r) * t/r)
+     y <- (1 - r) * sin(t) - d * sin((1 - r) * t/r)
+     data.frame(x, y)
+ }
> panel.hypocycloid.gg <- function(x, y, cycles = x, density = 30) {
+     panel.hypotrochoid.gg(r = x/y, d = x/y, cycles = cycles,
+         density = density)
+ }

Note

panel.lines(x, y) replaced with data.frame(x, y) in the panel.hypotrochoid.gg function.

> df <- ddply(grid, .(p, q, k), function(df) {
+     with(df, panel.hypocycloid.gg(q, p))
+ })

> pg <- ggplot(df, aes(x, y)) + geom_path() + facet_wrap(~k,
+     ncol = 4) + scale_x_continuous("", breaks = NA) +
+     scale_y_continuous("", breaks = NA)
> print(pg)

Figure 13.2

lattice

> set.seed(20070706)

> pl <- xyplot(c(-1, 1) ~ c(-1, 1), aspect = 1, cycles = 15,
+     scales = list(draw = FALSE), xlab = "", ylab = "",
+     panel = panel.hypotrochoid)
> print(pl[rep(1, 42)])

ggplot2

> df2 <- ldply(rep(1:42), function(k) {
+     data.frame(k, panel.hypotrochoid.gg(cycles = 15))
+ })

> pg <- ggplot(df2, aes(x, y)) + geom_path() + facet_wrap(~k,
+     ncol = 6) + scale_x_continuous("", breaks = NA) +
+     scale_y_continuous("", breaks = NA) + opts(panel.margin = 0,
+     strip.text.x = theme_blank())
> print(pg)

Figure 13.3

> library("logspline")

lattice

> prepanel.ls <- function(x, n = 50, ...) {
+     fit <- logspline(x)
+     xx <- do.breaks(range(x), n)
+     yy <- dlogspline(xx, fit)
+     list(ylim = c(0, max(yy)))
+ }
> panel.ls <- function(x, n = 50, ...) {
+     fit <- logspline(x)
+     xx <- do.breaks(range(x), n)
+     yy <- dlogspline(xx, fit)
+     panel.lines(xx, yy, ...)
+ }

> faithful$Eruptions <- equal.count(faithful$eruptions,
+     4)

> pl <- densityplot(~waiting | Eruptions, data = faithful,
+     prepanel = prepanel.ls, panel = panel.ls)
> print(pl)

ggplot2

> fn <- function(data = faithful$eruptions, number = 4,
+     ...) {
+     intrv <<- as.data.frame(co.intervals(data, number,
+         ...))
+     eruptions <- sort(unique(data))
+     intervals <- ldply(eruptions, function(x) {
+         t(as.numeric(x < intrv$V2 & x > intrv$V1))
+     })
+     tmp <- melt(cbind(eruptions, intervals), id.var = 1)
+     tmp[tmp$value > 0, 1:2]
+ }
> faithful2 <- merge(faithful, fn())
> intrv <- with(intrv, paste(V1, V2, sep = "-"))
> faithful2 <- rename(faithful2, c(variable = "erupt"))
> faithful2$erupt <- factor(faithful2$erupt, labels = intrv)

> panel.ls.gg <- function(x, n = 50, ...) {
+     fit <- logspline(x)
+     xx <- do.breaks(range(x), n)
+     yy <- dlogspline(xx, fit)
+     data.frame(xx, yy, ...)
+ }
> a <- ddply(faithful2, .(erupt), function(df) {
+     panel.ls.gg(df$waiting)
+ })

> pg <- ggplot(a, aes(xx, yy)) + geom_line() + facet_grid(~erupt)
> print(pg)

Figure 13.4

> data(Chem97, package = "mlmRev")

lattice

> panel.bwtufte <- function(x, y, coef = 1.5, ...) {
+     x <- as.numeric(x)
+     y <- as.numeric(y)
+     ux <- sort(unique(x))
+     blist <<- tapply(y, factor(x, levels = ux), boxplot.stats,
+         coef = coef, do.out = FALSE)
+     blist.stats <<- t(sapply(blist, "[[", "stats"))
+     blist.out <<- lapply(blist, "[[", "out")
+     panel.points(y = blist.stats[, 3], x = ux, pch = 16,
+         ...)
+     panel.segments(x0 = rep(ux, 2), y0 = c(blist.stats[,
+         1], blist.stats[, 5]), x1 = rep(ux, 2), y1 = c(blist.stats[,
+         2], blist.stats[, 4]), ...)
+ }

> pl <- bwplot(gcsescore^2.34 ~ gender | factor(score),
+     Chem97, panel = panel.bwtufte, layout = c(6, 1),
+     ylab = "Transformed GCSE score")
> print(pl)

ggplot2

> dt <- ddply(Chem97, .(gender, score), function(df) {
+     boxplot.stats(df$gcsescore^2.34)$stats
+ })

> pg <- ggplot(dt, aes(x = gender)) + geom_linerange(aes(ymin = V1,
+     ymax = V2)) + geom_linerange(aes(ymin = V4, ymax = V5)) +
+     geom_point(aes(y = V3)) + facet_grid(~score)
> print(pg)

Figure 13.5

lattice

> data(Cars93, package = "MASS")
> cor.Cars93 <- cor(Cars93[, !sapply(Cars93, is.factor)],
+     use = "pair")
> ord <- order.dendrogram(as.dendrogram(hclust(dist(cor.Cars93))))
> panel.corrgram <- function(x, y, z, subscripts, at, level = 0.9,
+     label = FALSE, ...) {
+     require("ellipse", quietly = TRUE)
+     x <- as.numeric(x)[subscripts]
+     y <- as.numeric(y)[subscripts]
+     z <- as.numeric(z)[subscripts]
+     zcol <- level.colors(z, at = at, ...)
+     for (i in seq(along = z)) {
+         ell <- ellipse(z[i], level = level, npoints = 50,
+             scale = c(0.2, 0.2), centre = c(x[i], y[i]))
+         panel.polygon(ell, col = zcol[i], border = zcol[i],
+             ...)
+     }
+     if (label)
+         panel.text(x = x, y = y, lab = 100 * round(z,
+             2), cex = 0.8, col = ifelse(z < 0, "white",
+             "black"))
+ }

> pl <- levelplot(cor.Cars93[ord, ord], at = do.breaks(c(-1.01,
+     1.01), 20), xlab = NULL, ylab = NULL, colorkey = list(space = "top"),
+     scales = list(x = list(rot = 90)), panel = panel.corrgram,
+     label = TRUE)
> print(pl)

ggplot2

Ellipses are not supported in ggplot2.

Figure 13.6

lattice

> panel.corrgram.2 <- function(x, y, z, subscripts, at = pretty(z),
+     scale = 0.8, ...) {
+     require("grid", quietly = TRUE)
+     x <- as.numeric(x)[subscripts]
+     y <- as.numeric(y)[subscripts]
+     z <- as.numeric(z)[subscripts]
+     zcol <- level.colors(z, at = at, ...)
+     for (i in seq(along = z)) {
+         lims <- range(0, z[i])
+         tval <- 2 * base::pi * seq(from = lims[1], to = lims[2],
+             by = 0.01)
+         grid.polygon(x = x[i] + 0.5 * scale * c(0, sin(tval)),
+             y = y[i] + 0.5 * scale * c(0, cos(tval)),
+             default.units = "native", gp = gpar(fill = zcol[i]))
+         grid.circle(x = x[i], y = y[i], r = 0.5 * scale,
+             default.units = "native")
+     }
+ }

> pl <- levelplot(cor.Cars93[ord, ord], xlab = NULL, ylab = NULL,
+     at = do.breaks(c(-1.01, 1.01), 101), panel = panel.corrgram.2,
+     scales = list(x = list(rot = 90)), colorkey = list(space = "top"),
+     col.regions = colorRampPalette(c("red", "white",
+         "blue")))
> print(pl)

ggplot2

Not supported in +ggplot2+.

Figure 13.7

lattice

> panel.3d.contour <- function(x, y, z, rot.mat, distance,
+     nlevels = 20, zlim.scaled, ...) {
+     add.line <- trellis.par.get("add.line")
+     panel.3dwire(x, y, z, rot.mat, distance, zlim.scaled = zlim.scaled,
+         ...)
+     clines <- contourLines(x, y, matrix(z, nrow = length(x),
+         byrow = TRUE), nlevels = nlevels)
+     for (ll in clines) {
+         m <- ltransform3dto3d(rbind(ll$x, ll$y, zlim.scaled[2]),
+             rot.mat, distance)
+         panel.lines(m[1, ], m[2, ], col = add.line$col,
+             lty = add.line$lty, lwd = add.line$lwd)
+     }
+ }

> pl <- wireframe(volcano, zlim = c(90, 250), nlevels = 10,
+     aspect = c(61/87, 0.3), panel.aspect = 0.6, panel.3d.wireframe = "panel.3d.contour",
+     shade = TRUE, screen = list(z = 20, x = -60))
> print(pl)

ggplot2

ggplot2 currently does not support true 3d surfaces.

Figure 13.8

lattice

> library("maps")
> county.map <- map("county", plot = FALSE, fill = TRUE)

> data(ancestry, package = "latticeExtra")
> ancestry <- subset(ancestry, !duplicated(county))
> rownames(ancestry) <- ancestry$county
> freq <- table(ancestry$top)
> keep <- names(freq)[freq > 10]
> ancestry$mode <- with(ancestry, factor(ifelse(top %in%
+     keep, top, "Other")))
> modal.ancestry <- ancestry[county.map$names, "mode"]

> library("RColorBrewer")
> colors <- brewer.pal(n = nlevels(ancestry$mode), name = "Pastel1")

> pl <- xyplot(y ~ x, county.map, aspect = "iso", scales = list(draw = FALSE),
+     xlab = "", ylab = "", par.settings = list(axis.line = list(col = "transparent")),
+     col = colors[modal.ancestry], border = NA, panel = panel.polygon,
+     key = list(text = list(levels(modal.ancestry), adj = 1),
+         rectangles = list(col = colors), x = 1, y = 0,
+         corner = c(1, 0)))
> print(pl)

ggplot2

> counties <- map_data("county")
> counties$reg <- with(counties, paste(region, subregion,
+     sep = ","))
> co_anc <- merge(counties, ancestry, by.x = "reg", by.y = "county")
> co_anc <- co_anc[order(co_anc$order), ]

> pg <- ggplot(co_anc, aes(long, lat, fill = mode, group = group)) +
+     geom_polygon() + scale_fill_brewer("", palette = "Pastel1")
> print(pg)

Figure 13.9

lattice

> rad <- function(x) {
+     pi * x/180
+ }
> county.map$xx <- with(county.map, cos(rad(x)) * cos(rad(y)))
> county.map$yy <- with(county.map, sin(rad(x)) * cos(rad(y)))
> county.map$zz <- with(county.map, sin(rad(y)))
> panel.3dpoly <- function(x, y, z, rot.mat = diag(4),
+     distance, ...) {
+     m <- ltransform3dto3d(rbind(x, y, z), rot.mat, distance)
+     panel.polygon(x = m[1, ], y = m[2, ], ...)
+ }
> aspect <- with(county.map, c(diff(range(yy, na.rm = TRUE)),
+     diff(range(zz, na.rm = TRUE)))/diff(range(xx, na.rm = TRUE)))

> pl <- cloud(zz ~ xx * yy, county.map, par.box = list(col = "grey"),
+     aspect = aspect, panel.aspect = 0.6, lwd = 0.01,
+     panel.3d.cloud = panel.3dpoly, col = colors[modal.ancestry],
+     screen = list(z = 10, x = -30), key = list(text = list(levels(modal.ancestry),
+         adj = 1), rectangles = list(col = colors), space = "top",
+         columns = 4), scales = list(draw = FALSE), zoom = 1.1,
+     xlab = "", ylab = "", zlab = "")
> print(pl)

ggplot2

ggplot2 currently does not support true 3d surfaces.

Figure 13.10

> library("latticeExtra")
> library("mapproj")
> data(USCancerRates)
> rng <- with(USCancerRates, range(rate.male, rate.female,
+     finite = TRUE))
> nbreaks <- 50
> breaks <- exp(do.breaks(log(rng), nbreaks))
> breaks2 <- c(unique(breaks[1 + (0:(nbreaks - 1)%/%10) *
+     10]), max(breaks) - 0.1)

lattice

> pl <- mapplot(rownames(USCancerRates) ~ rate.male + rate.female,
+     data = USCancerRates, breaks = breaks, map = map("county",
+         plot = FALSE, fill = TRUE, projection = "tetra"),
+     scales = list(draw = T), xlab = "", main = "Average yearly deaths due to cancer per 100000")
> print(pl)

ggplot2

> USCancerRates.df <- namerows(USCancerRates, col.name = "reg")
> co_cancer <- merge(counties, USCancerRates.df, by = c("reg"))
> co_cancer <- co_cancer[order(co_cancer$order), ]
> co_cancer.m <- melt(co_cancer, measure.vars = c("rate.male",
+     "rate.female"), na.rm = TRUE)
> co_cancer.m$fill <- with(co_cancer.m, as.numeric(as.character(cut(value,
+     breaks, labels = comma(breaks[-1])))))

> brewer.div <- colorRampPalette(brewer.pal(11, "Spectral"))

> pg <- ggplot(co_cancer.m, aes(long, lat, group = reg,
+     fill = fill)) + geom_polygon() + coord_map(projection = "tetra") +
+     facet_wrap(~variable, ncol = 1) + scale_fill_gradientn("",
+     colours = brewer.div(nbreaks), trans = "log") + opts(title = "Average yearly deaths due to cancer per 100000")
> print(pg)

Srarytety...

dzokero — Tue, 18 Aug 2009 09:46:45 +0000

Projekt ustawy o parytecie gotowy. Połowę miejsc na listach wyborczych mają zajmować kobiety. A jeśl

Spowiedź...

Hercules — Sat, 15 Aug 2009 14:53:26 +0000

Ta opowieść, która będzie się ciągnęła przez wszystkie wpisy na tym blogu, jest moją spowiedzią.

Mógłbym powiedzieć, patrzcie i uczcie się na moich grzechach. Ale nie powiem tego, gdyż homoseksualizm nie jest złem.

Wiem jedno o spowiedziach.

Są Anonimowe.

Dlatego nazwijcie mnie Herculesem.

Cała ta historia zaczęła się wiele lat temu, dokładnie 21 marca 1950r. W dniu gdy przyszedłem na świat.

Homoseksualistą byłem zawsze. Wtedy gdy bawiłem się z innymi w piaskownicy, wtedy gdy wsiadałem na swój pierwszy rower, wtedy gdy zdawałem maturę, wtedy gdy brałem ślub z “miłością mojego życia”, wtedy gdy asystowałem przy narodzinach mojej córki i wtedy gdy rozwodziłem się.

Tak, dobrze czytasz. Miałem żonę. Mam córkę.

Ożeniłem się 30 grudnia 1991r. Byłem 41-latkiem, a jeszcze nie miałem pewności kim jestem. Oczywiście, wcześniej w moim życiu było kilka epizodów związanych z mężczyznami.
Przelotny seks lub dłuższe związki, ale gdy poznałem domniemaną “miłość mojego życia”, wszystko się zmieniło. Myślałem, że “istnienie facetów w moim życiu uczuciowym” było błędem, zwykłą pomyłką.

Rok później, miałem już zupełnie inne zdanie o mojej żonie. Wiedziałem, iż preferuję mężczyzn, że to oni mnie podniecają, że to z nimi chcę uprawiać seks, że to z jednym z nich chcę spędzić resztę swojego życia. Niestety (a może “stety”) dziecko było już w drodze. Postanowiłem odczekać czas do narodzin.

Gosia urodziła się dokładnie 26 maja 1992r.

Pół roku później było już po rozwodzie.

W 1993r. poznałem mojego obecnego faceta, razem jesteśmy od sierpnia 1994r. Ma na imię Daniel, jest młodszy o 14 lat, moją córkę traktuje tak jak by była i jego dzieckiem.

ggplot2 Version of Figures in "Lattice: Multivariate Data Visualization with R" (Part 11)

learnr — Thu, 13 Aug 2009 09:00:48 +0000

This is the 11th post in a series attempting to recreate the figures in Lattice: Multivariate Data Visualization with R (R code available here) with ggplot2.

Previous parts in this series: Part 1, Part 2, Part 3, Part 4, Part 5, Part 6, Part 7, Part 8, Part 9, Part 10.

Chapter 11 – Manipulating the “trellis” object

Topics covered:

Methods for “trellis” objects
Tukey mean-difference plot
Other specialized manipulations

Figure 11.1

> library(lattice)
> library(ggplot2)

lattice

> dp.uspe <- dotplot(t(USPersonalExpenditure), groups = FALSE,
+     index.cond = function(x, y) median(x), layout = c(1,
+         5), type = c("p", "h"), xlab = "Expenditure (billion dollars)")
> dp.uspe.log <- dotplot(t(USPersonalExpenditure), groups = FALSE,
+     index.cond = function(x, y) median(x), layout = c(1,
+         5), scales = list(x = list(log = 2)), xlab = "Expenditure (billion dollars)")
> plot(dp.uspe, split = c(1, 1, 2, 1), more = TRUE)
> plot(dp.uspe.log, split = c(2, 1, 2, 1), more = FALSE)

ggplot2

> library(ggextra)

> p <- ggplot(melt(USPersonalExpenditure), aes(X2, value,
+     ymin = 0, ymax = value)) + geom_point() + coord_flip() +
+     facet_wrap(~X1, ncol = 1) + ylab("Expenditure (billion dollars)")
> p1 <- p + geom_linerange()
> p2 <- p + scale_y_log2()
> print(arrange(p1, p2))

Figure 11.2

lattice

> state <- data.frame(state.x77, state.region, state.name)
> state$state.name <- with(state, reorder(reorder(state.name,
+     Frost), as.numeric(state.region)))

> dpfrost <- dotplot(state.name ~ Frost | reorder(state.region,
+     Frost), data = state, layout = c(1, 4), scales = list(y = list(relation = "free")))
> plot(dpfrost, panel.height = list(x = c(16, 13, 9, 12),
+     unit = "null"))

ggplot2

> state$state.region <- with(state, reorder(state.region,
+     Frost))

> pg <- ggplot(state, aes(Frost, state.name)) + geom_point() +
+     facet_grid(state.region ~ ., scales = "free_y", space = "free",
+         as.table = F)
> print(pg)

Figure 11.3

lattice

> plot(dpfrost, panel.height = list(x = c(16, 13, 9, 12),
+     unit = "null"))
> print(update(trellis.last.object(), layout = c(1, 1))[2])

ggplot2

> pg <- pg %+% subset(state, state.region == "West") +
+     facet_wrap(~state.region, scales = "free_y")
> print(pg)

Figure 11.4

lattice

> npanel <- 12
> rot <- list(z = seq(0, 30, length = npanel), x = seq(0,
+     -80, length = npanel))
> quakeLocs <- cloud(depth ~ long + lat, quakes, pch = ".",
+     cex = 1.5, panel = function(..., screen) {
+         pn <- panel.number()
+         panel.cloud(..., screen = list(z = rot$z[pn],
+             x = rot$x[pn]))
+     }, xlab = NULL, ylab = NULL, zlab = NULL, scales = list(draw = FALSE),
+     zlim = c(690, 30), par.settings = list(axis.line = list(col = "transparent")))
> pl <- quakeLocs[rep(1, npanel)]
> print(pl)

ggplot2

True 3d not supported in ggplot2.

Figure 11.5

> data(Chem97, package = "mlmRev")

lattice

> ChemQQ <- qq(gender ~ gcsescore | factor(score), Chem97,
+     f.value = ppoints(100), strip = strip.custom(style = 5))
> pl <- tmd(ChemQQ)
> print(pl)

ggplot2

> q <- function(x, probs = ppoints(100)) {
+     data.frame(q = probs, value = quantile(x, probs))
+ }
> Chem97.q <- ddply(Chem97, c("gender", "score"), function(df) q(df$gcsescore))
> Chem97.df <- recast(Chem97.q, score + q ~ gender, id.var = 1:3)

> pg <- ggplot(Chem97.df, aes(M, F)) + geom_point(aes(x = (M +
+     F)/2, y = F - M)) + facet_wrap(~score) + xlab("mean") +
+     ylab("difference")
> print(pg)

Figure 11.6

> library("latticeExtra")
> data(biocAccess)

lattice

> baxy <- xyplot(log10(counts) ~ hour | month + weekday,
+     biocAccess, type = c("p", "a"), as.table = TRUE,
+     pch = ".", cex = 2, col.line = "black")
> dimnames(baxy)$month <- month.name[1:5]
> pl <- useOuterStrips(baxy)
> print(pl)

ggplot2

> pg <- ggplot(biocAccess, aes(hour, log10(counts))) +
+     geom_point(colour = "steelblue", size = 1) + geom_line(stat = "summary",
+     fun.y = mean) + facet_grid(weekday ~ month)
> print(pg)

ggplot2 Version of Figures in "Lattice: Multivariate Data Visualization with R" (Part 10)

learnr — Tue, 11 Aug 2009 12:00:27 +0000

This is the 10th post in a series attempting to recreate the figures in Lattice: Multivariate Data Visualization with R (R code available here) with ggplot2.

Previous parts in this series: Part 1, Part 2, Part 3, Part 4, Part 5, Part 6, Part 7, Part 8, Part 9.

Chapter 10 – Data Manipulation and Related Topics

Topics covered:

Non-standard evaluation in the context of lattice
The extended formula interface
Combining data sources, subsetting
Shingles
Ordering levels of categorical variables
Controlling the appearance of strips

Figure 10.1

> library(lattice)
> library(ggplot2)

> Titanic1 <- as.data.frame(as.table(Titanic[, , "Adult",
+     ]))

lattice

> pl <- barchart(Class ~ Freq | Sex, Titanic1, groups = Survived,
+     stack = TRUE, auto.key = list(title = "Survived",
+         columns = 2))
> print(pl)

ggplot2

> pg <- ggplot(Titanic1, aes(Class, Freq, fill = Survived)) +
+     geom_bar(stat = "identity") + coord_flip() + facet_grid(~Sex)
> print(pg)

Figure 10.2

lattice

> Titanic2 <- reshape(Titanic1, direction = "wide", v.names = "Freq",
+     idvar = c("Class", "Sex"), timevar = "Survived")
> names(Titanic2) <- c("Class", "Sex", "Dead", "Alive")

> pl <- barchart(Class ~ Dead + Alive | Sex, Titanic2,
+     stack = TRUE, auto.key = list(columns = 2))
> print(pl)

ggplot2

> pg <- ggplot(Titanic1, aes(Class, Freq, fill = Survived)) +
+     geom_bar(stat = "identity") + coord_flip() + facet_grid(~Sex) +
+     scale_fill_discrete("", labels = c("Dead", "Alive"))
> print(pg)

Figure 10.3

> data(Gcsemv, package = "mlmRev")

lattice

> pl <- xyplot(written ~ course | gender, data = Gcsemv,
+     type = c("g", "p", "smooth"), xlab = "Coursework score",
+     ylab = "Written exam score", panel = function(x,
+         y, ...) {
+         panel.xyplot(x, y, ...)
+         panel.rug(x = x[is.na(y)], y = y[is.na(x)])
+     })
> print(pl)

ggplot2

> pg <- ggplot(Gcsemv, aes(course, written)) + geom_point() +
+     geom_smooth(method = "loess", se = F) + geom_rug() +
+     facet_grid(~gender) + xlab("Coursework score") +
+     ylab("Written exam score")
> print(pg)

Figure 10.4

lattice

> pl <- qqmath(~written + course, Gcsemv, type = c("p",
+     "g"), outer = TRUE, groups = gender, auto.key = list(columns = 2),
+     f.value = ppoints(200), ylab = "Score")
> print(pl)

ggplot2

> pg <- ggplot(melt(Gcsemv, id.vars = 1:3, na.rm = TRUE)) +
+     geom_point(aes(sample = value, colour = gender),
+         stat = "qq", quantiles = ppoints(200)) + facet_grid(~variable)
> print(pg)

Figure 10.5

> set.seed(20051028)
> x1 <- rexp(2000)
> x1 <- x1[x1 > 1]
> x2 <- rexp(1000)

lattice

> pl <- qqmath(~data, make.groups(x1, x2), groups = which,
+     distribution = qexp, aspect = "iso", type = c("p",
+         "g"))
> print(pl)

Note

make.groups() function in lattice package “combines two or more vectors, possibly of different lengths, producing a data frame with a second column indicating which of these vectors that row came from”. Another alternative would be to use combine() function contained in gtools package.

ggplot2

> pg <- ggplot(make.groups(x1, x2)) + geom_point(aes(sample = data,
+     colour = which), stat = "qq", distribution = qexp) +
+     coord_equal()
> print(pg)

Figure 10.6

> beavers <- make.groups(beaver1, beaver2)
> beavers$hour <- with(beavers, time%/%100 + 24 * (day -
+     307) + (time%%100)/60)

lattice

> pl <- xyplot(temp ~ hour | which, data = beavers, groups = activ,
+     auto.key = list(text = c("inactive", "active"), columns = 2),
+     xlab = "Time (hours)", ylab = "Body Temperature (C)",
+     scales = list(x = list(relation = "sliced")))
> print(pl)

ggplot2

> pg <- ggplot(beavers, aes(hour, temp, colour = factor(activ))) +
+     geom_point() + facet_grid(~which, scales = "free_x") +
+     xlab("Time (hours)") + ylab("Body Temperature (C)")
> print(pg)

Figure 10.7

> data(USAge.df, package = "latticeExtra")

lattice

> pl <- xyplot(Population ~ Age | factor(Year), USAge.df,
+     groups = Sex, type = c("l", "g"), auto.key = list(points = FALSE,
+         lines = TRUE, columns = 2), aspect = "xy", ylab = "Population (millions)",
+     subset = Year %in% seq(1905, 1975, by = 10))
> print(pl)

ggplot2

> pg <- ggplot(USAge.df, aes(Age, Population, colour = Sex)) +
+     geom_line(subset = .(Year %in% seq(1905, 1975, by = 10))) +
+     facet_wrap(~Year, ncol = 4) + ylab("Population (millions)")
> print(pg)

Note

ggplot2 doesn’t have the equivalent of aspect="xy" in lattice, which “tries to compute the aspect based on the 45 degree banking rule”.

Figure 10.8

lattice

> pl <- xyplot(Population ~ Year | factor(Age), USAge.df,
+     groups = Sex, type = "l", strip = FALSE, strip.left = TRUE,
+     layout = c(1, 3), ylab = "Population (millions)",
+     auto.key = list(lines = TRUE, points = FALSE, columns = 2),
+     subset = Age %in% c(0, 10, 20))
> print(pl)

ggplot2

> pg <- ggplot(USAge.df, aes(Year, Population, colour = Sex)) +
+     geom_line(subset = .(Age %in% c(0, 10, 20))) + facet_grid(Age ~
+     .) + ylab("Population (millions)")
> print(pg)

Figure 10.9

lattice

> pl <- xyplot(Population ~ Year | factor(Year - Age),
+     USAge.df, groups = Sex, subset = (Year - Age) %in%
+         1894:1905, type = c("g", "l"), ylab = "Population (millions)",
+     auto.key = list(lines = TRUE, points = FALSE, columns = 2))
> print(pl)

ggplot2

> USAge.df$YearAge <- with(USAge.df, Year - Age)

> pg <- ggplot(USAge.df, aes(Year, Population, colour = Sex)) +
+     geom_line(subset = .((YearAge) %in% 1894:1905)) +
+     facet_wrap(~YearAge) + ylab("Population (millions)")
> print(pg)

Figure 10.10

lattice

> pl <- xyplot(stations ~ mag, quakes, jitter.x = TRUE,
+     type = c("p", "smooth"), xlab = "Magnitude (Richter)",
+     ylab = "Number of stations reporting")
> print(pl)

ggplot2

> pg <- ggplot(quakes, aes(mag, stations)) + geom_jitter(position = position_jitter(width = 0.01),
+     shape = 1) + geom_smooth(method = "loess", se = F) +
+     xlab("Magnitude (Richter)") + ylab("Number of stations reporting")
> print(pg)

Figure 10.11

> quakes$Mag <- equal.count(quakes$mag, number = 10, overlap = 0.2)

lattice

> ps.mag <- plot(quakes$Mag, ylab = "Level", xlab = "Magnitude (Richter)")
> bwp.quakes <- bwplot(stations ~ Mag, quakes, xlab = "Magnitude",
+     ylab = "Number of stations reporting")
> plot(bwp.quakes, position = c(0, 0, 1, 0.65))
> plot(ps.mag, position = c(0, 0.65, 1, 1), newpage = FALSE)

ggplot2

> Layout <- grid.layout(nrow = 2, ncol = 1, heights = unit(c(1,
+     2), c("null", "null")))
> grid.show.layout(Layout)
> vplayout <- function(...) {
+     grid.newpage()
+     pushViewport(viewport(layout = Layout))
+ }
> subplot <- function(x, y) viewport(layout.pos.row = x,
+     layout.pos.col = y)
> pr <- function() {
+     vplayout()
+     print(p1, vp = subplot(1, 1))
+     print(p2, vp = subplot(2, 1))
+ }

> fn <- function(data = quakes$mag, number = 4, ...) {
+     intrv <<- as.data.frame(co.intervals(data, number,
+         ...))
+     mag <- sort(unique(data))
+     intervals <- ldply(mag, function(x) {
+         t(as.numeric(x < intrv$V2 & x > intrv$V1))
+     })
+     tmp <- melt(cbind(mag, intervals), id.var = 1)
+     tmp[tmp$value > 0, 1:2]
+ }
> quakes.ordered <- merge(quakes, fn(number = 10, overlap = 0.2))
> intrv <- with(intrv, paste(V1, V2, sep = "-"))
> quakes.ordered <- rename(quakes.ordered, c(variable = "magnitude"))
> quakes.ordered$magnitude <- factor(quakes.ordered$magnitude,
+     labels = intrv)

> p1 <- ggplot(fn(number = 10, overlap = 0.2), aes(variable,
+     mag)) + stat_summary(aes(xmin = as.numeric(variable) -
+     0.4, xmax = as.numeric(variable) + 0.4), fun.ymin = min,
+     fun.ymax = max, geom = "rect") + coord_flip()
> p2 <- ggplot(quakes.ordered, aes(as.numeric(magnitude),
+     stations, group = magnitude)) + geom_boxplot() +
+     xlab("Magnitude") + ylab("Number of stations reporting")
> pr()

Figure 10.12

lattice

> pl <- bwplot(sqrt(stations) ~ Mag, quakes, scales = list(x = list(limits = as.character(levels(quakes$Mag)),
+     rot = 60)), xlab = "Magnitude (Richter)", ylab = expression(sqrt("Number of stations")))
> print(pl)

ggplot2

> pg <- ggplot(quakes.ordered, aes(magnitude, sqrt(stations),
+     group = magnitude)) + geom_boxplot() + xlab("Magnitude (Richter)") +
+     ylab(expression(sqrt("Number of stations"))) + opts(axis.text.x = theme_text(angle = 45,
+     hjust = 1, colour = "grey50"))
> print(pg)

Note

ggplot2 doesn’t have the equivalent of aspect="xy" in lattice, which “tries to compute the aspect based on the 45 degree banking rule”.

Figure 10.13

lattice

> pl <- qqmath(~sqrt(stations) | Mag, quakes, type = c("p",
+     "g"), pch = ".", cex = 3, prepanel = prepanel.qqmathline,
+     aspect = "xy", strip = strip.custom(strip.levels = TRUE,
+         strip.names = FALSE), xlab = "Standard normal quantiles",
+     ylab = expression(sqrt("Number of stations")))
> print(pl)

ggplot2

> pg <- ggplot(quakes.ordered, aes(sample = sqrt(stations))) +
+     geom_point(stat = "qq") + facet_wrap(~magnitude,
+     nrow = 2) + scale_x_continuous("Standard normal quantiles") +
+     scale_y_continuous(expression(sqrt("Number of stations")))
> print(pg)

Note

ggplot2 doesn’t have the equivalent of aspect="xy" in lattice, which “tries to compute the aspect based on the 45 degree banking rule”.

Figure 10.14

lattice

> pl <- xyplot(sqrt(stations) ~ mag, quakes, cex = 0.6,
+     panel = panel.bwplot, horizontal = FALSE, box.ratio = 0.05,
+     xlab = "Magnitude (Richter)", ylab = expression(sqrt("Number of stations")))
> print(pl)

ggplot2

> pg <- ggplot(quakes.ordered, aes(factor(mag), sqrt(stations))) +
+     geom_boxplot() + xlab("Magnitude (Richter)") + ylab(expression(sqrt("Number of stations")))
> print(pg)

Figure 10.15

> state.density <- data.frame(name = state.name, area = state.x77[,
+     "Area"], population = state.x77[, "Population"],
+     region = state.region)
> state.density$density <- with(state.density, population/area)

lattice

> pl <- dotplot(reorder(name, density) ~ density, state.density,
+     xlab = "Population Density (thousands per square mile)")
> print(pl)

ggplot2

> pg <- ggplot(state.density, aes(density, reorder(name,
+     density))) + geom_point() + xlab("Population Density (thousands per square mile)")
> print(pg)

Figure 10.16

lattice

> state.density$Density <- shingle(state.density$density,
+     intervals = rbind(c(0, 0.2), c(0.2, 1)))

> pl <- dotplot(reorder(name, density) ~ density | Density,
+     state.density, strip = FALSE, layout = c(2, 1), levels.fos = 1:50,
+     scales = list(x = "free"), between = list(x = 0.5),
+     xlab = "Population Density (thousands per square mile)",
+     par.settings = list(layout.widths = list(panel = c(2,
+         1))))
> print(pl)

ggplot2

> state.density$pos <- state.density$density > 0.2

> pg <- ggplot(state.density, aes(density, reorder(name,
+     density))) + geom_point() + facet_grid(~pos, scales = "free_x") +
+     xlab("Population Density (thousands per square mile)") +
+     opts(strip.background = theme_blank(), strip.text.x = theme_blank())
> print(pg)

Note

It is not possible to change the facet width in ggplot2.

Figure 10.17

lattice

> cutAndStack <- function(x, number = 6, overlap = 0.1,
+     type = "l", xlab = "Time", ylab = deparse(substitute(x)),
+     ...) {
+     time <- if (is.ts(x))
+         time(x)
+     else seq_along(x)
+     Time <- equal.count(as.numeric(time), number = number,
+         overlap = overlap)
+     xyplot(as.numeric(x) ~ time | Time, type = type,
+         xlab = xlab, ylab = ylab, default.scales = list(x = list(relation = "free"),
+             y = list(relation = "free")), ...)
+ }

> pl <- cutAndStack(EuStockMarkets[, "DAX"], aspect = "xy",
+     scales = list(x = list(draw = FALSE), y = list(rot = 0)))
> print(pl)

ggplot2

> library(zoo)

> cutAndStack_g <- function(data, number = 4, overlap = 0.1) {
+     ylab = deparse(substitute(data))
+     data <- as.data.frame(data)
+     data$id <- if (is.ts(data$x))
+         time(data$x)
+     else seq_along(data$x)
+     intrv <<- as.data.frame(co.intervals(data$id, number,
+         overlap))
+     x <- sort(unique(data$id))
+     intervals <- ldply(x, function(x) {
+         t(as.numeric(x < intrv$V2 & x > intrv$V1))
+     })
+     tmp <- melt(cbind(x, intervals), id.var = 1)
+     tmp <- tmp[tmp$value > 0, 1:2]
+     tmp <- rename(tmp, c(x = "id"))
+     stock <- merge(data, tmp)
+     stock$variable <- factor(stock$variable, labels = with(intrv,
+         paste(as.yearmon(V1), as.yearmon(V2), sep = " - ")))
+     p <- ggplot(stock, aes(id, x)) + geom_line() + facet_wrap(~variable,
+         scales = "free", ncol = 1, as.table = FALSE) +
+         scale_x_continuous("", breaks = NA) + ylab(ylab)
+     print(p)
+ }

> cutAndStack_g(data = EuStockMarkets[, "DAX"], number = 6,
+     overlap = 0.1)

Note

ggplot2 doesn’t have the equivalent of aspect=”xy” in lattice, which “tries to compute the aspect based on the 45 degree banking rule”.

Figure 10.18

lattice

> bdp1 <- dotplot(as.character(variety) ~ yield | as.character(site),
+     barley, groups = year, layout = c(1, 6), auto.key = list(space = "top",
+         columns = 2), aspect = "fill")
> bdp2 <- dotplot(variety ~ yield | site, barley, groups = year,
+     layout = c(1, 6), auto.key = list(space = "top",
+         columns = 2), aspect = "fill")
> plot(bdp1, split = c(1, 1, 2, 1))
> plot(bdp2, split = c(2, 1, 2, 1), newpage = FALSE)

ggplot2

> Layout <- grid.layout(ncol = 3, widths = unit(c(2, 2,
+     0.75), c("null", "null", "null")))
> vplayout <- function(...) {
+     grid.newpage()
+     pushViewport(viewport(layout = Layout))
+ }
> subplot <- function(x, y) viewport(layout.pos.row = x,
+     layout.pos.col = y)
> pr <- function() {
+     vplayout()
+     print(p_left, vp = subplot(1, 1))
+     print(p_right, vp = subplot(1, 2))
+     print(legend, vp = subplot(1, 3))
+ }

> p <- ggplot(barley, aes(yield, variety, colour = year)) +
+     geom_point() + facet_wrap(~site, ncol = 1) + ylab("")
> legend <- p + opts(keep = "legend_box")
> p_right <- p + opts(legend.position = "none")
> barley$site <- factor(barley$site, levels = sort(levels(barley$site)))
> barley$variety <- factor(barley$variety, levels = sort(levels(barley$variety)))
> p_left <- p_right %+% barley
> pr()

Figure 10.19

> state.density <- data.frame(name = state.name, area = state.x77[,
+     "Area"], population = state.x77[, "Population"],
+     region = state.region)
> state.density$density <- with(state.density, population/area)

lattice

> pl <- dotplot(reorder(name, density) ~ 1000 * density,
+     state.density, scales = list(x = list(log = 10)),
+     xlab = "Density (per square mile)")
> print(pl)

ggplot2

> pg <- ggplot(state.density, aes(density * 1000, reorder(name,
+     density))) + geom_point() + scale_x_log10() + xlab("Density (per square mile)") +
+     ylab("")
> print(pg)

Figure 10.20

> state.density$region <- with(state.density, reorder(region,
+     density, median))
> state.density$name <- with(state.density, reorder(reorder(name,
+     density), as.numeric(region)))

lattice

> pl <- dotplot(name ~ 1000 * density | region, state.density,
+     strip = FALSE, strip.left = TRUE, layout = c(1, 4),
+     scales = list(x = list(log = 10), y = list(relation = "free")),
+     xlab = "Density (per square mile)")
> print(pl)

ggplot2

> pg <- ggplot(state.density, aes(1000 * density, name)) +
+     geom_point() + facet_grid(region ~ ., scales = "free_y") +
+     scale_x_log10() + xlab("Density (per square mile)") +
+     ylab("")
> print(pg)

Figure 10.21

> library("latticeExtra")

lattice

> print(pl)
> print(resizePanels())

ggplot2

> pg <- pg + facet_grid(region~., scales = "free_y", space = "free", as.table = FALSE)
> print(pg)

Figure 10.22

> data(USCancerRates, package = "latticeExtra")

lattice

> pl <- xyplot(rate.male ~ rate.female | state, USCancerRates,
+     aspect = "iso", pch = ".", cex = 2, index.cond = function(x,
+         y) {
+         median(y - x, na.rm = TRUE)
+     }, scales = list(log = 2, at = c(75, 150, 300, 600)),
+     panel = function(...) {
+         panel.grid(h = -1, v = -1)
+         panel.abline(0, 1)
+         panel.xyplot(...)
+     })
> print(pl)

ggplot2

> med <- ddply(USCancerRates, .(state), summarise, med = median(rate.female -
+     rate.male, na.rm = TRUE))
> med$state <- with(med, reorder(state, med))
> USCancerRates$state <- factor(USCancerRates$state, levels = levels(med$state))

> pg <- ggplot(USCancerRates, aes(rate.female, rate.male)) +
+     geom_point(size = 1) + geom_abline(intercept = 0,
+     slope = 1) + scale_x_log2(breaks = c(75, 150, 300,
+     600), labels = c(75, 150, 300, 600)) + scale_y_log2(breaks = c(75,
+     150, 300, 600), labels = c(75, 150, 300, 600)) +
+     facet_wrap(~state, ncol = 7)
> print(pg)

Note

For some strange reason the order of facets is different compared to lattice plot, although the formula to set the levels is the same!

Figure 10.23

> data(Chem97, package = "mlmRev")

lattice

> strip.style4 <- function(..., style) {
+     strip.default(..., style = 4)
+ }

> pl <- qqmath(~gcsescore | factor(score), Chem97, groups = gender,
+     type = c("l", "g"), aspect = "xy", auto.key = list(points = FALSE,
+         lines = TRUE, columns = 2), f.value = ppoints(100),
+     strip = strip.style4, xlab = "Standard normal quantiles",
+     ylab = "Average GCSE score")
> print(pl)

ggplot2

> pg <- ggplot(Chem97, aes(sample = gcsescore, colour = gender)) +
+     geom_point(stat = "qq", quantiles = ppoints(100)) +
+     facet_wrap(~score) + scale_x_continuous("Standard Normal Quantiles") +
+     scale_y_continuous("Average GCSE Score") + theme_bw()
> print(pg)

Note

ggplot2 doesn’t have the equivalent of aspect="xy" in lattice, which “tries to compute the aspect based on the 45 degree banking rule”.

Figure 10.24

lattice

> strip.combined <- function(which.given, which.panel,
+     factor.levels, ...) {
+     if (which.given == 1) {
+         panel.rect(0, 0, 1, 1, col = "grey90", border = 1)
+         panel.text(x = 0, y = 0.5, pos = 4, lab = factor.levels[which.panel[which.given]])
+     }
+     if (which.given == 2) {
+         panel.text(x = 1, y = 0.5, pos = 2, lab = factor.levels[which.panel[which.given]])
+     }
+ }

> pl <- qqmath(~gcsescore | factor(score) + gender, Chem97,
+     f.value = ppoints(100), type = c("l", "g"), aspect = "xy",
+     strip = strip.combined, par.strip.text = list(lines = 0.5),
+     xlab = "Standard normal quantiles", ylab = "Average GCSE score")
> print(pl)

ggplot2

> Chem97 <- transform(Chem97, sc_gen = paste(score, gender))

> library(gtools)
> Chem97$sc_gen <- factor(Chem97$sc_gen, levels = mixedsort(unique(Chem97$sc_gen)))

> pg <- ggplot(Chem97, aes(sample = gcsescore)) + geom_point(stat = "qq",
+     quantiles = ppoints(100)) + facet_wrap(~sc_gen, nrow = 2) +
+     scale_x_continuous("Standard normal quantiles") +
+     scale_y_continuous("Average GCSE score")
> print(pg)

Note

ggplot2 doesn’t have the equivalent of aspect="xy" in lattice, which “tries to compute the aspect based on the 45 degree banking rule”.

Figure 10.25

> morris <- barley$site == "Morris"
> barley$year[morris] <- ifelse(barley$year[morris] ==
+     "1931", "1932", "1931")

lattice

> pl <- stripplot(sqrt(abs(residuals(lm(yield ~ variety +
+     year + site)))) ~ site, data = barley, groups = year,
+     jitter.data = TRUE, auto.key = list(points = TRUE,
+         lines = TRUE, columns = 2), type = c("p", "a"),
+     fun = median, ylab = expression(abs("Residual Barley Yield")^{
+         1/2
+     }))
> print(pl)

ggplot2

> pg <- ggplot(barley, aes(site, sqrt(abs(residuals(lm(yield ~
+     variety + year + site)))), colour = year)) + geom_jitter(position = position_jitter(w = 0.1)) +
+     geom_line(stat = "summary", fun.y = median, aes(group = year)) +
+     ylab(expression(abs("Residual Barley Yield")^{
+         1/2
+     }))
> print(pg)