RMarkdown and Knitr

• If you haven't done so already, install RStudio

  • Go to http://www.rstudio.com/products/rstudio/download/ and select the appropriate download for your system.

• Select File \(\rightarrow\) New File \(\rightarrow\) R Markdown from the dropdown menu.

• Give your document a title and author.

• You have three output format options:

  • HTML produces a web-readable document, which will open in any browser (hint: great for emailing but don't use this if you plan on printing your document)
  • PDF, uneditable to receivers and maintains its formatting on any platform.
  • Word documents allow the receiver to edit and comment on your document.

• The front-of-document chunk within the --- is a bit of css code that formats your document. With a bit of research, you can customize this. See http://rmarkdown.rstudio.com/html_document_format.html.

• Text chunks are separated from code chunks. Text chunks can include mathematical formulas, web links, anything you'd normally include in a text editor like Word.

• Code chunks are separated within '''{r} [code chunk] '''. By default, code will echo back as output before the results of the code, but this can be modified with echo = FALSE.

  • See http://rmarkdown.rstudio.com/authoring_rcodechunks.html.
    
  • Hint: It's a good idea to name your chunks - '''{r nameHere}

Emphasis:     *italic* or _italic_ 
              **bold** or __bold__

Headers:      # Header 1    
              ## Header 2   
              ### Header 3  

Unordered lists:    - Item 1
                    - Item 2
                    - Item 3

Ordered lists:      1. Item 1
                    2. Item 2
                    3. Item 3

## Classification With Iris Data Set

We will use R.A. Fisher's classic iris data set to generate a classification tree. 

# ```{r loadData, message = FALSE}
#Load data and required packages
data(iris)
library(caret); library(rattle)
nrow <- nrow(iris); ncol <- ncol(iris)  #inline code
iris[1:4,]
# ```

The data consists of `'r nrow'` rows and `'r ncol'` columns, with no missing values.  

We will use R.A. Fisher's classic iris data set to generate a classification tree.

#Load data and required packages
data(iris)
library(caret); library(rattle)
nrow <- nrow(iris); ncol <- ncol(iris)  # inline code
iris[1:4,]

##   Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1          5.1         3.5          1.4         0.2  setosa
## 2          4.9         3.0          1.4         0.2  setosa
## 3          4.7         3.2          1.3         0.2  setosa
## 4          4.6         3.1          1.5         0.2  setosa

The data consists of 150 rows and 5 columns, with no missing values.

### Analysis

Analysis consisted of fitting a predictive model of iris species based on 
petal length/width and sepal length/width. 

#```{r analysis, message = FALSE}
modFit <- train(Species ~., method = "rpart", data=iris) #Fit model
print(modFit$finalModel)   #Summarize model
#```

The final analysis is presented as a decison tree.

#```{r tree, fig.width = 5, fig.height = 4}

fancyRpartPlot(modFit$finalModel) #Plot decision tree
#```

Analysis consisted of fitting a predictive model of iris species based on petal length/width and sepal length/width. The final model was then plotted as a decision tree.

modFit <- train(Species ~., method = "rpart", data=iris) #Fit model
print(modFit$finalModel)   #Summarize model

## n= 150 
## 
## node), split, n, loss, yval, (yprob)
##       * denotes terminal node
## 
## 1) root 150 100 setosa (0.33333 0.33333 0.33333)  
##   2) Petal.Length< 2.45 50   0 setosa (1.00000 0.00000 0.00000) *
##   3) Petal.Length>=2.45 100  50 versicolor (0.00000 0.50000 0.50000)  
##     6) Petal.Width< 1.75 54   5 versicolor (0.00000 0.90741 0.09259) *
##     7) Petal.Width>=1.75 46   1 virginica (0.00000 0.02174 0.97826) *

The final model was then plotted as a decision tree.

fancyRpartPlot(modFit$finalModel) #Plot decision tree

Full document link: http://rpubs.com/BlackWidoww/52864

• Pandoc Markdown allows for tables, images, citations, etc. See http://rmarkdown.rstudio.com/authoring_pandoc_markdown.html#math

• Math:

  • Contain mathematical notation between two $ characters inline with text (i.e. - within a paragraph) or as its own paragraph between $$.
  • Uses LaTeX (pronounced Lay-teck)
  • http://www.artofproblemsolving.com/Wiki/index.php/LaTeX:Symbols
  • f(x) = $\frac{\lambda x}{1+(ax)^b}$ \(\rightarrow\) \(f(x)\:=\:\frac{\lambda x}{1\:+\:(ax)^b}\)

• Links:

  • Enclose a url between pointy brackets, <url>, for an automatic link.
  • Inline links: [my link](/url)
  • Reference links: Here is [my link][LINK] [LINK]: /url/goes/here

To complete your document you will need to knit together the text and the code chunks into (1) a human-readable document, and (2) a machine-readable document.

1. Save your document by clicking the floppy disk icon or File \(\rightarrow\) Save. Your document will have the file extension .Rmd.

2. Click Knit HTML to create two output documents:

• A .html document, which will open in your favorite web browser.
• A .md document, which contains all the document formatting. In the newer version of RStudio, this won't be saved but you can retain it in Options.
  

RPubs

You can publish an RMarkdown piece to a publicly accessible website using RPubs by clicking Publish on your finished HTML.