Clever Geek Handbook

Assign custom factor values ​​to each symbol group

I have one column of my dataframe that contains some symbols and a vector of factors. I would like to have a coefficient assigned to each group of values, so that the first group of characters gets the first coefficient, the second group gets the second, etc.

Col of data frame + vector of factors:

df$charac

:

          charac
1            0
2            0
3            0
4            1
5            1
6            2
7            2
8            2
9            3
10           4
11           4
12           4

vec_factor

:

[1] 39 42 76 89 68
Levels: 39 42 68 76 89

Expected results:

          charac  factor
1            0      39
2            0      39
3            0      39
4            1      42
5            1      42
6            2      76
7            2      76
8            2      76
9            3      89
10           4      68
11           4      68
12           4      68

Data:

Vector of factors:

structure(c(1L, 2L, 4L, 5L, 3L), .Label = c("39", "42", "68", 
"76", "89"), class = "factor")

col characters:

structure(list(test_vector = c("0", "0", "0", "1", "1", "2", 
"2", "2", "3", "4", "4", "4")), .Names = "test_vector", row.names = c(NA, 
-12L), class = "data.frame")
+3


source to share


3 answers


You can use rleid

from data.table

:

library(data.table)
df$factor<-vec_factor[rleid(df$test_vector)]


Result

 df
 test_vector factor
1            0     39
2            0     39
3            0     39
4            1     42
5            1     42
6            2     76
7            2     76
8            2     76
9            3     89
10           4     68
11           4     68
12           4     68
+1


source


You can do it in R base:

df$factor<- as.factor(df$test_vector)
levels(df$factor) <- levels(vec_factor)

   # test_vector factor
# 1            0     39
# 2            0     39
# 3            0     39
# 4            1     42
# 5            1     42
# 6            2     68
# 7            2     68
# 8            2     68
# 9            3     76
# 10           4     89
# 11           4     89
# 12           4     89

So, first create a type type column and then replace the levels with levels vec_factor

.



OR (thanks @alexis_laz for pointing this out)

df$factor <- factor(df$test_vector, labels = levels(vec_factor))
+2


source


We can do it

df1$factor <- as.character(vec_factor)[as.integer(df1[[1]])+1]
df1$factor
#[1] "39" "39" "39" "42" "42" "76" "76" "76" "89" "68" "68" "68"

Or use match 

df1$factor <- with(df1, vec_factor[match(test_vector, unique(test_vector))])
df1$factor
#[1] 39 39 39 42 42 76 76 76 89 68 68 68
#Levels: 39 42 68 76 89

NOTE. Both methods are inbase R

+1


source






More articles:


All Articles