I've been learning a bit of statistical computing with R lately on the side from Chris Paciorek's Berkeley course. I just got introduced to knitr and it's damned sweet! It's an R package which takes a LaTeX file with embedded R, and produces a pure LaTeX file (similar to how Rails renders an .html.erb file into a .html file), where the resulting LaTeX file has the output of the R code. It makes it super easy to embed statistical calculations, graphs, and all the good stuff R gives you right into your TeX files. It let's you put math in your math, so you can math while you math.

I've got a little project which:

1. Runs a Python script which will use Selenium to scrape a web page for 2012 NFL passing statistics.
2. "Knits" a TeX file with embedded R that cleans the raw scraped data, produces a histogram of touchown passes for teams, and displays the teams with the least and greatest number of touchdowns.
3. Compiles the resulting TeX file and opens the resulting PDF.
4. Cleans up any temporary work files.

Here's what the pre-"knitted" LaTeX looks like with the embedded R:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

\documentclass{article}
\usepackage{graphicx}
%% begin.rcode setup, include=FALSE
% opts_chunk$set(fig.path='figure/latex-', cache.path='cache/latex-')
%% end.rcode
\begin{document}

After scraping data for all passing TDs in 2012, we get the following histogram for number of TD passes by team.

%% begin.rcode cache=TRUE
% scrape <- read.csv('scrape.csv')
% raw_data <- scrape[scrape[,"X"]!="",]
% tds_for_passers <- transform(raw_data[c("Tm","TD")], TD = as.numeric(as.character(TD)))
% tds_for_teams <- aggregate(tds_for_passers$TD, by=list(Team=tds_for_passers$Tm), FUN=sum)
% hist(tds_for_teams$x)
%% end.rcode

The teams with the greatest and least TDs:

%% begin.rcode
% low_high <- c(which.min(tds_for_teams$x), which.max(tds_for_teams$x))
% tds_for_teams[low_high,"Team"]
%% end.rcode

\end{document}

You can comment out the line in the factory script that deletes the tds2012-out.tex file if you want to see what it looks like post-knit. The resulting TeX file basically contains a ton of new commonad definitions but the meat of it is what it does with your R code. It formats and displays the R code itself, and then it displays the output of the R code. Wherever the output is a graph, you'll see \includegraphics[...]{...}. knitr will do the R computation, render the graphics, create a figures subdirectory and store them there for the \includegraphics to reference. Whenever the output is simply text or mathematical expressions, you'll see the R output translated to pure LaTeX markup.

Pretty cool stuff!