I have a question. I'm working on building a recommendation system in R, and I'm fairly new to the language. I can't seem to figure the following out.
I have a matrix like:
eventID g_26 g_27 g_28 g_29 g_30 g_31 g_32 g_33 g_34 g_35 g_36 g_37 g_38 g_39 g_40 g_41 g_42 g_43
1: 1010 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0
2: 1016 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
3: 1019 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
4: 1053 1 0 1 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0
5: 1168 0 0 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 0
6: 1188 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
What I´d like to do is replace all values that have 1 to 1/sqrt(total # of 1's in that particular row).
I'm using the Data Table package as well if that makes it easier.
Thanks in advance!
We can specify the columns of interest in .SDcols (-1 implies we selected all the columns except the first column), get the sum of each row in the Subset of Data.table with Reduce and +, take the square root (sqrt), divide by 1, multiply with the Subset of data.table (.SD) and assign (:=) it to the columns of interest
dt[, (2:ncol(dt)) := .SD*1/sqrt(Reduce(`+`, .SD)), .SDcols = -1]