Search notes:

R: atomic vectors

(Atomic) vectors are probably the most fundamental data structure in the R programming language.

An atomic vector is different from a one-dimensional array: an array has a dim attribute of length one while a vector has no such attribute.

Thus, dim() applied on a vector returns NULL.

An atomic vector is also different from a list. The elements of a vector are all of the same type while a list can contain elements of different and arbitrary types.

Atomic vectors are constructed with the c function or the vector function.

The three properties of vectors are

type (of which there are six)
length (the number of elements that the vector contains, see function length())
attributes (which add arbitrary metadata)

The six vector types

R has six basic («atomic») vector types:

`typeof(…)`	`mode(…)`	`storage.mode(…)`
logical	logical	logical
integer	numeric	integer
double	numeric	double
character	character	character
complex	complex	complex
raw	raw	raw

Homogeneous

All elements within a vector share the same type, that is: they have the same mode In contrast, lists are heterogeneous.

As vectors can have elements of only one data type, it's not possible to nest vectors within other vectors:

not_nested_vector <- c( 1, 1, 2, c(3, 5, 8), 13, 21);
not_nested_vector;
#
#  1  1  2  3  5  8 13 21

Github repository about-R, path: /data-structures/vector/not-nested.R

Creating vectors

For example with

c(…)
seq(…)
as.vector produces a vector with a given mode and length.
…

1-indexed

The first element of a vector has index 1:

elems <- c('one', 'two', 'three')
elems[1] # one

Iterating over vectors

The apply function family is R's preferred way to call a function on each element of a vector or a list.

Comparing vectors

vec_one <- c(1, 2, 3)
vec_two <- c(0, 2, 4)

vec_one == vec_two
# [1] FALSE  TRUE FALSE

vec_one < vec_two
# [1] FALSE FALSE  TRUE

vec_one == 1
# [1]  TRUE FALSE FALSE

Github repository about-r, path: /vector/comparing.R

Selecting elements

vec <- 1:10

# Note: index is 1 based, not 0 based (as for example in
# Perl or C)
vec[4]
# [1] 4

vec[7:9]
# [1] 7 8 9

# Exclude specific elements (2, 4 and 6)
vec[c(-2,-4,-6)]
# [1]  1  3  5  7  8  9 10

# Select elements with TRUE (and excluding others with FALSE):
vec[c(FALSE, TRUE, FALSE, FALSE, TRUE, TRUE, TRUE, TRUE, FALSE, TRUE)]
# [1]  2  5  6  7  8 10

# Same idea, but generating TRUE/FALSE
vec[vec <= 5]
# [1] 1 2 3 4 5

# Select all elements that are greater than the mean of the vector:
vec[vec > mean(vec)]
# [1]  6  7  8  9 10

# selecting elements by name
names(vec) <- c('one', 'two', 'three', 'four', 'five', 'six', 'seven', 'eight', 'nine', 'ten')
vec[c('two', 'seven', 'three')]
#   two seven three
#     2     7     3

Github repository about-r, path: /vector/selecting_elements.R

Names property

Vectors can have the names property which names each element within the vector.

vec <- 1:5
names(vec) <- c('one', 'two', 'three', 'four', 'five')
vec
#   one   two three  four  five
#     1     2     3     4     5

vec[c('two', 'five')]
#  two five
#    2    5

Github repository about-r, path: /vector/names_property.R

Visualizing vectors

A simple method to visualize the values in a vector is to use barplot.

Printing vectors vertically

At times, the elements of a vector can be easier seen by printing each element on its own line. This is possible with cat(…).

vec <- c('foo', 'bar', 'baz');

cat(vec, sep="\n");
#
# foo
# bar
# baz

Github repository about-r, path: /data-structures/vector/print-vertically.R