R

When plotting my typing performance, I wanted to demonstrate my baseline typing performance as well as the number of errors. For this, I discovered some quite useful capabilities in the CRAN package ggplot2.

results<-read.csv("./dvorak_vs_qwerty_typing_test_results.csv", header = TRUE);
png(filename="kinesis_dvorak.png", width=640, height=440, bg="#070707");
ggplot(results, aes(x=Trial, y=Kinesis.DVORAK)) + scale_x_continuous(name="Trial #", limits=c(1, 15), breaks=seq(1,15)) + scale_y_continuous(name="Words per Minute", limits=c(0,100), breaks=seq(0,100,by=10)) + geom_point(aes(x=results$Trial, y=(results$Kinesis.DVORAK + results$Errors)), color="#7b0000", pch=19) + geom_area(color="#004088", fill="#004088", alpha=0.5) + opts(title="Kinesis Keyboard - DVORAK Layout") + theme_bb() + opts(plot.title=theme_text(colour="#ffffff", size=14, vjust=1), axis.title.x=theme_text(colour="#ffffff",size=12,vjust=0), axis.title.y=theme_text(angle=90,colour="#ffffff",size=12), plot.background=theme_rect(colour="#070707", fill="#070707"), panel.background=theme_rect(colour="#666666"), panel.grid.major=theme_line(colour="#444444"), panel.grid.minor=theme_line(colour="#070707"));
dev.off();

Required packages: ggplot2, ggExtra

Basic feature comparisons (e.g. amount of L3 cache):
Xeon 7500-series L3 Cache

png( filename="plot.png", width=640, height=440, bg="#070707" );
par(bg="#070707", fg="white", col.axis="white", col.main="white", col.lab="white", mar=c(3,4,2,2) + 0.1);
plot(xeon$Model, xeon$L3.Cache, ylab="L3 Cache", border=brewer.pal(8, "Dark2"), cex.axis=1.3, cex.lab=1.5);
dev.off();


Two-factor comparisons (e.g. with and without turbo boost):
Xeon 7500-series Processor Frequency

png( filename="plot.png", width=640, height=440, bg="#070707" );
par(bg="#070707", fg="white", col.axis="white", col.main="white", col.lab="white", mar=c(3,4,2,2) + 0.1);
plot(xeon$Model, xeon$Frequency, ylab="Processor Frequency", ylim=c(1.8,2.8), border=brewer.pal(8, "Dark2"), cex.axis=1.3, cex.lab=1.5)
points(xeon$Model, xeon$Frequency..Turbo., col=brewer.pal(8, "Dark2"), pch=19)
dev.off();


Required packages: RColorBrewer

I've loved R since the first time I used it in statistics class. I went several years without using it, but I probably should have pulled it out a few times. It's incredibly powerful and ends up being quite fun for data visualization geeks. Unfortunately, it's not intuitive to a lot of users.

Further, even those of us comfortable with it seem to quickly forget exactly how we displayed the data just the way we wanted last time. I'm going to post the tricks/scripts I write whenever I use R - both for others and myself to re-use. Just click on the tag "R".