Big Data Analytics 简明教程

Big Data Analytics - Introduction to R

本节专门向用户介绍 R 编程语言。可从 cran website 下载 R。对于 Windows 用户， install rtools 和 rstudio IDE 很有用。

This section is devoted to introduce the users to the R programming language. R can be downloaded from the cran website. For Windows users, it is useful to install rtools and the rstudio IDE.

R 背后的总体概念是充当已编译语言（例如 C、C++ 和 Fortran）编写的其他软件的界面，并为用户提供一个交互式工具来分析数据。

The general concept behind R is to serve as an interface to other software developed in compiled languages such as C, C++, and Fortran and to give the user an interactive tool to analyze data.

导航到图书 zip 文件的文件夹 bda/part2/R_introduction 并打开 R_introduction.Rproj 文件。这将打开一个 RStudio 会话。然后打开 01_vectors.R 文件。逐行运行脚本并遵循代码中的注释。为了学习，另一个有用的选择就是键入代码，这将帮助您习惯 R 语法。在 R 中，注释是用 # 符号编写的。

Navigate to the folder of the book zip file bda/part2/R_introduction and open the R_introduction.Rproj file. This will open an RStudio session. Then open the 01_vectors.R file. Run the script line by line and follow the comments in the code. Another useful option in order to learn is to just type the code, this will help you get used to R syntax. In R comments are written with the # symbol.

为了在图书中显示运行 R 代码的结果，在评估代码后，将注释返回的结果 R。通过这种方式，您可以在书中复制粘贴代码并直接在 R 中尝试其部分。

In order to display the results of running R code in the book, after code is evaluated, the results R returns are commented. This way, you can copy paste the code in the book and try directly sections of it in R.

# Create a vector of numbers
numbers = c(1, 2, 3, 4, 5)
print(numbers)

# [1] 1 2 3 4 5
# Create a vector of letters
ltrs = c('a', 'b', 'c', 'd', 'e')
# [1] "a" "b" "c" "d" "e"

# Concatenate both
mixed_vec = c(numbers, ltrs)
print(mixed_vec)
# [1] "1" "2" "3" "4" "5" "a" "b" "c" "d" "e"

让我们分析一下之前代码中发生的情况。我们可以看到可以使用数字和字母创建向量。我们无需事先告诉 R 我们想要什么类型的数据类型。最后，我们能够创建带有数字和字母的向量。混合向量已将数字强制转换为字符，我们可以通过可视化如何将值打印在引号内来看到这一点。

Let’s analyze what happened in the previous code. We can see it is possible to create vectors with numbers and with letters. We did not need to tell R what type of data type we wanted beforehand. Finally, we were able to create a vector with both numbers and letters. The vector mixed_vec has coerced the numbers to character, we can see this by visualizing how the values are printed inside quotes.

以下代码显示了函数类返回的不同向量的类型。通常使用类函数来“询问”对象，询问他的类别是什么。

The following code shows the data type of different vectors as returned by the function class. It is common to use the class function to "interrogate" an object, asking him what his class is.

### Evaluate the data types using class

### One dimensional objects
# Integer vector
num = 1:10
class(num)
# [1] "integer"

# Numeric vector, it has a float, 10.5
num = c(1:10, 10.5)
class(num)
# [1] "numeric"

# Character vector
ltrs = letters[1:10]
class(ltrs)
# [1] "character"

# Factor vector
fac = as.factor(ltrs)
class(fac)
# [1] "factor"

R 也支持二维对象。在以下代码中，列出了 R 中使用最流行的两个数据结构的示例：矩阵和数据帧。

R supports two-dimensional objects also. In the following code, there are examples of the two most popular data structures used in R: the matrix and data.frame.

# Matrix
M = matrix(1:12, ncol = 4)
#      [,1] [,2] [,3] [,4]
# [1,]    1    4    7   10
# [2,]    2    5    8   11
# [3,]    3    6    9   12
lM = matrix(letters[1:12], ncol = 4)
#     [,1] [,2] [,3] [,4]
# [1,] "a"  "d"  "g"  "j"
# [2,] "b"  "e"  "h"  "k"
# [3,] "c"  "f"  "i"  "l"

# Coerces the numbers to character
# cbind concatenates two matrices (or vectors) in one matrix
cbind(M, lM)
#     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
# [1,] "1"  "4"  "7"  "10" "a"  "d"  "g"  "j"
# [2,] "2"  "5"  "8"  "11" "b"  "e"  "h"  "k"
# [3,] "3"  "6"  "9"  "12" "c"  "f"  "i"  "l"

class(M)
# [1] "matrix"
class(lM)
# [1] "matrix"

# data.frame
# One of the main objects of R, handles different data types in the same object.
# It is possible to have numeric, character and factor vectors in the same data.frame

df = data.frame(n = 1:5, l = letters[1:5])
df
#   n l
# 1 1 a
# 2 2 b
# 3 3 c
# 4 4 d
# 5 5 e

如前一个示例所示，可以在同一对象中使用不同的数据类型。通常，这是如何在数据库中呈现数据，API 数据的一部分是文本或字符向量，另一部分是数字。分析人员的工作是确定要分配哪种统计数据类型，然后为其使用正确的 R 数据类型。在统计中，我们通常认为变量为以下类型 −

As demonstrated in the previous example, it is possible to use different data types in the same object. In general, this is how data is presented in databases, APIs part of the data is text or character vectors and other numeric. In is the analyst job to determine which statistical data type to assign and then use the correct R data type for it. In statistics we normally consider variables are of the following types −

  1. Numeric

  2. Nominal or categorical

  3. Ordinal

在 R 中，向量可以属于以下类 −

In R, a vector can be of the following classes −

  1. Numeric - Integer

  2. Factor

  3. Ordered Factor

R 为每种变量的统计类型提供数据类型。然而，有序因子很少使用，但可以通过因子函数或 ordered 函数创建。

R provides a data type for each statistical type of variable. The ordered factor is however rarely used, but can be created by the function factor, or ordered.

以下部分讨论了索引的概念。这是一个非常常见的操作，并且处理选择对象的各个部分并对它们进行转换的问题。

The following section treats the concept of indexing. This is a quite common operation, and deals with the problem of selecting sections of an object and making transformations to them.

# Let's create a data.frame
df = data.frame(numbers = 1:26, letters)
head(df)
#      numbers  letters
# 1       1       a
# 2       2       b
# 3       3       c
# 4       4       d
# 5       5       e
# 6       6       f

# str gives the structure of a data.frame, it’s a good summary to inspect an object
str(df)
#   'data.frame': 26 obs. of  2 variables:
#   $ numbers: int  1 2 3 4 5 6 7 8 9 10 ...
#   $ letters: Factor w/ 26 levels "a","b","c","d",..: 1 2 3 4 5 6 7 8 9 10 ...

# The latter shows the letters character vector was coerced as a factor.
# This can be explained by the stringsAsFactors = TRUE argumnet in data.frame
# read ?data.frame for more information

class(df)
# [1] "data.frame"

### Indexing
# Get the first row
df[1, ]
#     numbers  letters
# 1       1       a

# Used for programming normally - returns the output as a list
df[1, , drop = TRUE]
# $numbers
# [1] 1
#
# $letters
# [1] a
# Levels: a b c d e f g h i j k l m n o p q r s t u v w x y z

# Get several rows of the data.frame
df[5:7, ]
#      numbers  letters
# 5       5       e
# 6       6       f
# 7       7       g

### Add one column that mixes the numeric column with the factor column
df$mixed = paste(df$numbers, df$letters, sep = ’’)

str(df)
# 'data.frame': 26 obs. of  3 variables:
# $ numbers: int  1 2 3 4 5 6 7 8 9 10 ...
# $ letters: Factor w/ 26 levels "a","b","c","d",..: 1 2 3 4 5 6 7 8 9 10 ...
# $ mixed  : chr  "1a" "2b" "3c" "4d" ...

### Get columns
# Get the first column
df[, 1]
# It returns a one dimensional vector with that column

# Get two columns
df2 = df[, 1:2]
head(df2)

#      numbers  letters
# 1       1       a
# 2       2       b
# 3       3       c
# 4       4       d
# 5       5       e
# 6       6       f

# Get the first and third columns
df3 = df[, c(1, 3)]
df3[1:3, ]

#      numbers  mixed
# 1       1     1a
# 2       2     2b
# 3       3     3c

### Index columns from their names
names(df)
# [1] "numbers" "letters" "mixed"
# This is the best practice in programming, as many times indeces change, but
variable names don’t
# We create a variable with the names we want to subset
keep_vars = c("numbers", "mixed")
df4 = df[, keep_vars]

head(df4)
#      numbers  mixed
# 1       1     1a
# 2       2     2b
# 3       3     3c
# 4       4     4d
# 5       5     5e
# 6       6     6f

### subset rows and columns
# Keep the first five rows
df5 = df[1:5, keep_vars]
df5

#      numbers  mixed
# 1       1     1a
# 2       2     2b
# 3       3     3c
# 4       4     4d
# 5       5     5e

# subset rows using a logical condition
df6 = df[df$numbers < 10, keep_vars]
df6

#      numbers  mixed
# 1       1     1a
# 2       2     2b
# 3       3     3c
# 4       4     4d
# 5       5     5e
# 6       6     6f
# 7       7     7g
# 8       8     8h
# 9       9     9i