How to remove only rows that have all NA in R? [duplicate]

问题

This question already has answers here:

Remove rows with all or some NAs (missing values) in data.frame (16 answers)

Removing empty rows of a data file in R (5 answers)

How to remove rows where columns satisfy certain condition in data frame (2 answers)

Closed last year.

I have a data frame where only some of the rows have all NA. How do I loop through and delete all these rows? I tried na.omit() but that doesn't work. In this example I need rows 3 and 5 removed

x1 <- c("Bob", "Mary","","Jane","")
x2 <- c("Bob","Mary","","Jane","")
x3 <- c("Bob", "Mary","","Jane","")
x4 <- c("Bob","Mary","","Jane","")

df <- data.frame(x1,x2,x3,x4)

df <- df %>% na.omit()

回答1:

Here is one option; first you need to define the NA pattern.

df[df == ""] <- NA # define NA pattern
df[rowSums(is.na(df)) != ncol(df), ] # result
# Try with x1 <- c("Bob", "Mary","","","") 

回答2:

> df[rowSums(df=="")!=ncol(df), ]
    x1   x2   x3   x4
1  Bob  Bob  Bob  Bob
2 Mary Mary Mary Mary
4 Jane Jane Jane Jane

来源：https://stackoverflow.com/questions/51214357/how-to-remove-only-rows-that-have-all-na-in-r