How to get the newly created Dummy variable columns in fastDummies in R?

Question

I'm using the R package fastDummies to create dummy variables for categorical variables. How can I select the names of the newly created columns with the dummy variables instead of getting the whole dataframe?

Example:

# Create random datframe
vec1<-sample(1:100, 50, replace=TRUE) # continuous variable
vec2<-sample(1:100, 50, replace=TRUE) # continuous variable
vec3<-sample(1:3, 50, replace=TRUE)  
vec4<-sample(1:3, 50, replace=TRUE)

mydata<-data.frame(vec1, vec2, vec3, vec4)
mydata$vec3<-factor(mydata$vec3)  # categorical variable with 3 levels
mydata$vec4<-factor(mydata$vec4)  # categorical variable with 3 levels

# Create dummy variables
library(fastDummies)
dummys<-dummy_columns(mydata, select_columns=c("vec3", "vec4"), remove_first_dummy = TRUE)

# Now dummys will contain this:
 head(dummys)

# output:
#  vec1 vec2 vec3 vec4 vec3_2 vec3_3 vec4_2 vec4_3
#1   40   59    3    2      0      1      1      0
#2   55    3    1    2      0      0      1      0
#3   26   55    1    3      0      0      0      1
#4   38   29    1    2      0      0      1      0
#5   33   54    2    1      1      0      0      0
#6   45   26    2    2      1      0      1      0

Now I want to select the newly created dummy variable columns without specifying them manually:

# NOT BY checking
colnames(mydata)
mydata$vec3_2
# etc...

But just by selecting the newly created dummy variable columns created by dummy_columns with whatever dataset used, since dummy_columns itself gives the whole dataset but I just want the names of the newly created dummy variable columns (so in this example: vec3_2, vec3_3, vec4_2, and vec4_3).

Anyone knows how to do this? Or even using another package/function to do this?

Answer 1

Well, a very simple use, is to do:

attach(mydata)

and then, you can simply use the vec3_2 for example, but, firstly, you could even did:

vec3<-factor(vec3)
vec4<-factor(vec4)

Without the necessity to use the symbol $ , actually it's quite simple.