--- title: "1.2 R Packages and the Tidyverse" --- ```{r} #| label: setup #| include: false knitr::opts_chunk$set( echo = TRUE, message = FALSE, warning = FALSE, fig.width = 6, fig.height = 3.6, fig.align = "center", dpi = 120, error = TRUE, fig.path = "_figs/r-packages-and-the-tidyverse/" ) library(tidyverse) ``` Installing packages, loading them, the tidyverse, and the pipe. ## 1. What Is a Package? R already contains many functions: ```{r} #| label: chunk-1 mean() sum() sqrt() seq() data.frame() ``` These functions come with **Base R** and the standard R installation. However, not every useful function is included in Base R. R can be extended using **packages**. > A **package** is a collection of functions, data sets, documentation, and other tools designed for particular tasks. Thousands of R packages are available for statistics, visualization, machine learning, data manipulation, and many other applications. --- ## 2. Installing a Package Before using a package for the first time, we usually need to **install** it. For example: ```{r} #| label: chunk-2 #| eval: false install.packages("ggplot2") ``` The package name must be inside quotation marks: ```{r} #| label: chunk-3 #| eval: false install.packages("ggplot2") ``` not: ```{r} #| label: chunk-4 #| eval: false install.packages(ggplot2) ``` Installation downloads the package and places it on your computer. ### Usually, you install a package only once. Think of installing an R package like installing an app on your computer. --- ## 3. Loading a Package Installing a package does **not** automatically make its functions available in the current R session. We load an installed package using: ```{r} #| label: chunk-5 library(ggplot2) ``` Notice the difference: ```{r} #| label: chunk-6 #| eval: false install.packages("ggplot2") # Install library(ggplot2) # Load ``` A useful rule to remember is: > **Install once; load each new R session.** For example, you may install `ggplot2` once: ```{r} #| label: chunk-7 #| eval: false install.packages("ggplot2") ``` Then, whenever you start a new R session and want to use it: ```{r} #| label: chunk-8 library(ggplot2) ``` --- ## 4. What Is an R Session? An **R session** is the period during which R is currently running. When you close and restart R or RStudio, you begin a new session. Packages that you loaded previously are generally not automatically loaded into the new session. Therefore, your script might begin with: ```{r} #| label: chunk-9 library(ggplot2) library(dplyr) ``` but you should **not normally put** this in a script that runs every time: ```{r} #| label: chunk-10 #| eval: false install.packages("ggplot2") ``` Installation is something you do when needed, not every time the script runs. --- ## 5. Using a Function from a Package Suppose we load `ggplot2`: ```{r} #| label: chunk-11 library(ggplot2) ``` We can now use functions provided by that package: ```{r} #| label: chunk-12 ggplot() ``` Another way to use a function is: ```{r} #| label: chunk-13 ggplot2::ggplot() ``` The notation: ```{r} #| label: chunk-14 package::function() ``` means: > "Use this function specifically from this package." For example: ```{r} #| label: chunk-15 dplyr::filter() ``` means: > Use the `filter()` function from the `dplyr` package. This can be useful because different packages can sometimes contain functions with the same name. --- ## 6. Getting Help R has a built-in help system. To read the documentation for a function: ```{r} #| label: chunk-16 ?mean ``` or: ```{r} #| label: chunk-17 help(mean) ``` For a function from a package: ```{r} #| label: chunk-18 ?ggplot2::ggplot ``` You can also search the help system: ```{r} #| label: chunk-19 help.search("mean") ``` A shortcut is: ```{r} #| label: chunk-20 ??mean ``` Therefore: ?function → documentation for a function ??topic → search R help for a topic 7. What Packages Are Currently Loaded? You can see information about the current R session using: ```{r} #| label: chunk-21 sessionInfo() ``` You can also see the current search path: ```{r} #| label: chunk-22 search() ``` For example, after: ```{r} #| label: chunk-23 library(ggplot2) ``` try: ```{r} #| label: chunk-24 search() ``` and notice that `ggplot2` appears in the search path. --- ## 8. What Packages Are Installed? To see installed packages: ```{r} #| label: chunk-25 installed.packages() ``` This returns considerable information. To see just their names: ```{r} #| label: chunk-26 rownames(installed.packages()) ``` To check whether a particular package is available: ```{r} #| label: chunk-27 requireNamespace("ggplot2", quietly = TRUE) ``` --- ## 9. What Is the Tidyverse? In this course, we will learn both **Base R** and the **tidyverse**. The tidyverse is a collection of R packages designed to work together for data science. We can install it using: ```{r} #| label: chunk-28 #| eval: false install.packages("tidyverse") ``` and load it using: ```{r} #| label: chunk-29 library(tidyverse) ``` Loading `tidyverse` loads several important packages. Some of the most commonly used are: Package Main purpose `ggplot2` Data visualization `dplyr` Data manipulation `tidyr` Reshaping and tidying data `readr` Reading data files `tibble` Modern data frames `stringr` Working with strings `forcats` Working with categorical data/factors `purrr` Functional programming and iteration We will encounter these packages gradually throughout the course. --- ## 10. Base R vs. Tidyverse Often, the same task can be performed using either Base R or tidyverse tools. For example, suppose: ```{r} #| label: chunk-30 students <- data.frame( name = c("John", "Sara", "David"), grade = c(75, 92, 85) ) ``` Using **Base R**, we could select students with grades above 80: ```{r} #| label: chunk-31 students[students$grade > 80, ] ``` or, with `subset()`: ```{r} #| label: chunk-32 subset(students, grade > 80) ``` `subset()` lets you write the condition with the column name. In dplyr, two functions sit next to it: - `filter()` is the closest match for keeping rows - `select()` is the closest match for keeping columns (`subset(students, select = name)`) ```{r} #| label: chunk-33 library(dplyr) filter(students, grade > 80) ``` Both the Base R and dplyr approaches are valid. Throughout this course, we will learn how to work with data using both approaches. --- ## 11. The Pipe One important idea in modern R and the tidyverse is the **pipe**. Suppose: ```{r} #| label: chunk-34 x <- c(10, 20, 30, 40) ``` We could calculate its mean using: ```{r} #| label: chunk-35 mean(x) ``` Base R provides the pipe: ```{r} #| label: chunk-36 x |> mean() ``` You can read this approximately as: > Take `x`, **then** calculate its mean. The tidyverse historically popularized another pipe: ```{r} #| label: chunk-37 x %>% mean() ``` The `%>%` pipe is provided by `magrittr` and is commonly encountered in tidyverse code. In this course, you may encounter both: ```{r} #| label: chunk-38 #| eval: false |> # Base R pipe %>% # traditional tidyverse pipe ``` The pipe becomes especially useful when several operations are performed in sequence. For example: ```{r} #| label: chunk-39 students |> subset(grade > 80) |> head() ``` `subset()` is Base R. The dplyr line that does the same row-keeping is `filter(grade > 80)`. Note 2.1 goes through `subset()`, `filter()`, and `select()` more carefully. Later, we will use pipes extensively with tidyverse functions. --- ## 12. A Typical R Workflow A simple R project might follow this pattern: ### Step 1 — Install required packages Usually done only once: ```{r} #| label: chunk-40 #| eval: false install.packages("tidyverse") ``` ### Step 2 — Load packages Done when beginning a new R session: ```{r} #| label: chunk-41 library(tidyverse) ``` ### Step 3 — Import or create data We can create a small data set in code: ```{r} #| label: chunk-42 students <- data.frame( name = c("John", "Sara", "David"), grade = c(75, 92, 85) ) ``` We can also read a file. Download [students.csv](files/students.csv) and place it in your working directory. Then: ```{r} #| label: chunk-43 #| eval: false students <- read.csv("students.csv") ``` With the tidyverse, an equivalent approach is: ```{r} #| label: chunk-44 #| eval: false students <- readr::read_csv("students.csv") ``` `read.csv()` is Base R. `read_csv()` comes from the `readr` package. Note **1.3** is the full treatment: relative paths, delimiters, `read.table()` / `read_csv()`, inspecting the import, and exporting. ### Step 4 — Explore the data ```{r} #| label: chunk-45 #| eval: false head(students) str(students) summary(students) View(students) ``` ### Step 5 — Manipulate the data ```{r} #| label: chunk-46 students |> dplyr::filter(grade > 80) ``` ### Step 6 — Analyze or visualize the data For example: ```{r} #| label: chunk-47 mean(students$grade) ``` or later: ```{r} #| label: chunk-48 ggplot(students, aes(x = name, y = grade)) + geom_col() ``` --- ## 13. Important Package Commands to Remember ```{r} #| label: chunk-49 #| eval: false install.packages("package") ``` **Install** a package. ```{r} #| label: chunk-50 library(package) ``` **Load** an installed package. ```{r} #| label: chunk-51 package::function() ``` Use a particular **function from a particular package**. ```{r} #| label: chunk-52 ?function ``` Read a function's **documentation**. ```{r} #| label: chunk-53 ??topic ``` **Search** R's help system. ```{r} #| label: chunk-54 installed.packages() ``` See the packages that are **installed**. ```{r} #| label: chunk-55 sessionInfo() ``` See information about the current **R session**. ```{r} #| label: chunk-56 search() ``` See packages currently attached to the **search path**. --- ## 14. The Big Picture A useful way to think about the R ecosystem is: R │ ┌───────────┴───────────┐ │ │ Base R Packages │ │ Built-in functions Additional functions │ │ mean(), seq(), etc. ┌──────┴─────────┐ │ │ tidyverse Other packages │ ┌────────────┼────────────┐ │ │ │ dplyr ggplot2 tidyr │ │ │ Data Visualization Reshaping manipulation data The central distinction to remember is: > **R is the programming language. A package extends R by providing additional functions and tools.** And: > **Base R and tidyverse are not two different programming languages. The tidyverse is a collection of packages that we use within R.** For packages, remember the three most important commands: ```{r} #| label: chunk-57 #| eval: false install.packages("tidyverse") # Get it library(tidyverse) # Load it dplyr::filter() # Use a specific function ```