In practice, the data set is rarely complete, and in many cases, the sample will contain several missing values NA, which is troublesome in data analysis and mining.
R language can improve the omit value of samples by na.fail and na.omit.
The

na. Fail (& lt; Vector a>) : If vector A contains at least 1 NA, error is returned; If NA is excluded, return the original vector ana. Omit (& LT; Vector a>) : Return the vector aattr (na.omit (& LT); Vector a>) , “na.action”) : returns the subscript of na in vector a. Na: determines whether the element in the vector is na

Example:

data< – c (1, 2, NA, 2,4,2,10, NA, 9)
data. NA. Omit< – na. Omit (data)
data. Na. Omit the
[1] 1 2 2 and 4 2 10 9
attr (, “na. The action”)
3 8 [1]
attr (” class “)
[1] “omit”

attr (data. Na. Omit, “na. The action”)
3 8 [1]
attr (” class “)
[1] “omit”

can also be used! X mode conveniently deletes NA. Such as:

a< – c (1, 2, 3, NA, NA, 2, NA, 5)
a[!is.na(a)]
[1] 1 2 3 2 5

.
which is for na na is used to determine whether the element in the vector, returns the result: c (FALSE, FALSE, FALSE, TRUE, TRUE, FALSE, TRUE, FALSE), namely the elements within a as na, its corresponding subscript elements is TRUE, otherwise is FALSE. ! X is the non-logical operator,! Is. Na (a) means that the element inside a is not Na, and its subscript element is TRUE and FALSE conversely. After indexing through A [! Is.na (a)], the element that is not Na in A can be taken out and filtered.
The functions Na. fail and Na. omit can be applied not only to vectors but also to matrices and data boxes.
Example:

data < – read.table(text=”
a b c d e f
NA 1 1 1 1 1
1 NA 1 1 1 1
1 1 NA 1 1 1
1 1 1 NA 1 1
1 1 1 1 NA 1
1 1 1 1 1 NA”,header=T)
na.omit(data)
data
> [1] a b c d e f
< 0 line & gt; (or 0-length row.names)

— — — — — — — — — — — — — — — — — — — — —
the author: SThranduil
source: CSDN
,
https://blog.csdn.net/SThranduil/article/details/71710283 copyright statement: this article original articles for bloggers, reproduced please attach link to blog!