可选的更有效的循环选项

Question

Are there any bug-free and faster alternatives to this for loop code?

for(i in 1:length(Mergedf)) 
{if (Mergedf[i,"z"]==0) 
{Mergedf[i,"cntry_origin"] <-Mergedf[i,"V2"]} 
  print(Mergedf[i,"cntry_origin"])
}

Mergdf if the result of a complex logic.

Thanks

Answer 1

What you're attempting to do is a very basic R 'subset and replace' problem. If you're used to writing in a language like VBA then a for loop seems natural for this. However, where R "excels" (pun intended) is the ability to vectorise these kinds of operations so it does it in one step, without the need to loop through the entire data set.

The code here is all written in Base R

Consider the example data

set.seed(1)
MergedDF <- data.frame("z" = c(0,1,2,3,0,1,2,3),
                       "cntry_origin" = letters[1:8],
                       "V2" = rnorm(8,0,1),
                       stringsAsFactors = FALSE)

#   z cntry_origin         V2
# 1 0            a -0.6264538
# 2 1            b  0.1836433
# 3 2            c -0.8356286
# 4 3            d  1.5952808
# 5 0            e  0.3295078
# 6 1            f -0.8204684
# 7 2            g  0.4874291
# 8 3            h  0.7383247

Filtering a data.frame for a given condition

To get all the rows where z == 0

MergedDF[MergedDF$z == 0, ]

#   z cntry_origin         V2
# 1 0            a -0.6264538
# 5 0            e  0.3295078

Selecting specific columns

To get all the values in column cntry_origin where z==0 , there are two equivalent statements:

MergedDF[MergedDF$z == 0, "cntry_origin"]
# [1] "a" "e"

## Or
MergedDF[MergedDF$z == 0, ]$cntry_origin
# [1] "a" "e" 

To get the value of column V2 where z==0

MergedDF[MergedDF$z == 0, ]$V2
#[1] -0.6264538  0.3295078

Replacing one column with different values

To replace the column cntry_origin with V2 , where z==0 it's just a matter of assigning one to the other

MergedDF[MergedDF$z == 0, ]$cntry_origin <- MergedDF[MergedDF$z == 0, ]$V2

MergedDF

#   z       cntry_origin         V2
# 1 0 -0.626453810742332 -0.6264538
# 2 1                  b  0.1836433
# 3 2                  c -0.8356286
# 4 3                  d  1.5952808
# 5 0  0.329507771815361  0.3295078
# 6 1                  f -0.8204684
# 7 2                  g  0.4874291
# 8 3                  h  0.7383247

---

The equivalent data.table code would be

library(data.table)
setDT(MergedDF)[z==0, cntry_origin := V2]