CODEWITHSUNDEEP
rdplyr
Shreya Agarwal
Shreya Agarwal

Reputation: 716

How to replace certain values in a specific rows and columns with NA in R?

In my data frame, I want to replace certain blank cells and cells with values with NA. But the cells I want to replace with NAs has nothing to do with the value that cell stores, but with the combination of row and column it is stored in.

Here's a sample data frame DF:

  Fruits   Price   Weight   Number of pieces

  Apples      20      2          10
  Oranges     15      4          16
  Pineapple   40      8           6
  Avocado     60      5          20

I want to replace Pineapple'e weight to NA and Orange's number of pieces to NA. 

DF$Weight[3] <- NA
DF$`Number of pieces`[2] <- NA

This replaces any value that's stored in that position and that may change. I want to use specific row and column names to do this replacement so the position of value becomes irrelevant.

Output:

 Fruits   Price   Weight   Number of pieces

  Apples      20      2          10
  Oranges     15      4          NA
  Pineapple   40      NA           6
  Avocado     60      5          20

But if order of the table is changed, this would replace wrong values with NA.

How should I do this?

Upvotes: 10

Views: 58351

Answers (3)

Ben373
Ben373

Reputation: 971

Since your data structure is 2 dimensional, you can find the indices of the rows containing a specific value first and then use this information.

which(DF$Fruits == "Pineapple")
[1]  3
DF$Weight[which(DF$Fruits == "Pineapple")] <- NA

You should be aware of that which will return a vector, so if you have multiple fruits called "Pineapple" then the previous command will return all indices of them.

Upvotes: 17

Rui Barradas
Rui Barradas

Reputation: 76651

Here is a way using function is.na<-.

is.na(DF$Weight) <- DF$Fruits == "Pineapple"
is.na(DF$`Number of pieces`) <- DF$Fruits == "Oranges"

DF
#     Fruits Price Weight Number of pieces
#1    Apples    20      2               10
#2   Oranges    15      4               NA
#3 Pineapple    40     NA                6
#4   Avocado    60      5               20

Data in dput format. 

DF <-
structure(list(Fruits = structure(c(1L, 3L, 4L, 2L), 
.Label = c("Apples", "Avocado", "Oranges", "Pineapple"), 
class = "factor"), Price = c(20L, 15L, 40L, 60L), 
Weight = c(2L, 4L, 8L, 5L), `Number of pieces` = c(10L, 
16L, 6L, 20L)), class = "data.frame", row.names = c(NA, -4L))

Upvotes: 3

NelsonGon
NelsonGon

Reputation: 13319

library(dplyr)
df %>% 
  mutate(Weight=ifelse(Fruits=="Pineapple",NA,Weight),
         Number=ifelse(Fruits=="Oranges",NA,Number))#use Number of Pieces

Result: Number of pieces was truncated to Number due to reading data.

     Fruits Price Weight Number
1    Apples    20      2     10
2   Oranges    15      4     NA
3 Pineapple    40     NA      6
4   Avocado    60      5     20

Upvotes: 4

Related Questions