problem in setting one column as row id and then removing it

Question

I have a simple problem but I just don't know how to fix it. I have a dataframe like this:

structure(list(ID = c(1006332, 1010660, 
1012960, 1013515, 1014490), ave_ocean = c(1, 0, 0, 0, 0), ave_price_per_sqft = c(1419.69, 
912.18, 600.74, 673.8725, 439.46), ave_year = c(2005, 2009, 1986.4, 
2006.25, 1983), ave_DOM = c(7, 10, 36.1, 10.5, 104), total_num_sold = c(1L, 
1L, 10L, 4L, 1L)), row.names = c(NA, -5L), class = c("tbl_df", 
"tbl", "data.frame"))

                         ID ave_ocean ave_price_per_sqft ave_year ave_DOM total_num_sold
                       <dbl>     <dbl>              <dbl>    <dbl>   <dbl>          <int>
1                    1006332         1              1420.    2005      7                1
2                    1010660         0               912.    2009     10                1
3                    1012960         0               601.    1986.    36.1             10
4                    1013515         0               674.    2006.    10.5              4
5                    1014490         0               439.    1983    104                1

I want to set column ID as row name and then remove it. I use the code below:

row.names(data) <- data$ID
data <-data[,-c(1)]

when I run this it first changes the row name and replaces it with ID but once I run the second line to remove "ID" column it changes row name to the normal row name! Any idea? Any alternative solution?

`Warning message: Setting row names on a tibble is deprecated.` — user2974951, Feb 18 '21 at 06:57
@user2974951 Although I get this warning, it's working. The odd part is when I run the second line of code it turns back to the original row name! I need to do this for clustering. I don't want to keep the ID and want to separate them. that's why I assigned it to row names (they are all unique) — Ross_you, Feb 18 '21 at 06:59

score 2 · Answer 1 · answered Feb 18 '21 at 07:04

2

tibble::column_to_rownames(df, "ID")

read documentation (first sentance) ?tibble::column_to_rownames, also note that return value is data.frame beacuse of that reason.

answered Feb 18 '21 at 07:04

det

5,013
1
8
16

score 0 · Accepted Answer · answered Feb 18 '21 at 07:02

The first line sets the rownames but you'll not be able to see it since you have a tibble.

row.names(data) <- data$ID

#$Warning message:
#Setting row names on a tibble is deprecated. 
data
# A tibble: 5 x 6
#       ID ave_ocean ave_price_per_sqft ave_year ave_DOM total_num_sold
#*   <dbl>     <dbl>              <dbl>    <dbl>   <dbl>          <int>
#1 1006332         1              1420.    2005      7                1
#2 1010660         0               912.    2009     10                1
#3 1012960         0               601.    1986.    36.1             10
#4 1013515         0               674.    2006.    10.5              4
#5 1014490         0               439.    1983    104                1

If you change your data to dataframe you'll see the rownames.

data <- data.frame(data)
df

#             ID ave_ocean ave_price_per_sqft ave_year ave_DOM total_num_sold
#1006332 1006332         1          1419.6900  2005.00     7.0              1
#1010660 1010660         0           912.1800  2009.00    10.0              1
#1012960 1012960         0           600.7400  1986.40    36.1             10
#1013515 1013515         0           673.8725  2006.25    10.5              4
#1014490 1014490         0           439.4600  1983.00   104.0              1

problem in setting one column as row id and then removing it

2 Answers2