how to flatten values in this column to separate columns in R

Question

I have some data which looks as follows

"ID","PROD"
"1001658",6619
"100288",11843
"100288",20106
"1004303",921

I need to convert it into a format like

"ID","PROD_6619","PROD_11843","PROD_20106","PROD_921"
"1001658",1,0,0,0
"100288",0,1,1,0
"1004303",0,0,0,1

Basically where each value in the column PROD from the original data set is in a separate column of it's own. Note that the above dataset is only a sample and I cannot hard code to be "PROD_6619","PROD_11843","PROD_20106","PROD_921". It could be much more.

I have tried writing this iteratively using a for loop and it's very slow for my huge data set.

Can you suggest me an alternative in R

Look up `dcast` in the `reshape2` package – shadowtalker Jun 27 '15 at 05:50 — shadowtalker, Jun 27 '15 at 05:50

score 6 · Answer 1 · answered Jun 27 '15 at 04:30

6

You can just use table for something like this.

Example:

table(mydf)
##          PROD
## ID        921 6619 11843 20106
##   100288    0    0     1     1
##   1001658   0    1     0     0
##   1004303   1    0     0     0

answered Jun 27 '15 at 04:30

A5C1D2H2I1M1N2O1R2T1

190,393
28
405
485

score 0 · Answer 2 · answered Jun 27 '15 at 12:29

Here is another approach with the dcast of the reshape2 package.

library(reshape2)

dcast(dat, ID ~ PROD, length )
Using PROD as value column: use value.var to override.
       ID 921 6619 11843 20106
1  100288   0    0     1     1
2 1001658   0    1     0     0
3 1004303   1    0     0     0

how to flatten values in this column to separate columns in R

2 Answers2