Cheatsheet.pdf

The DataCamp course "Data Analysis the data.table Way" provides an overview of the data.table package in R. The general form of a data.table operation is DT[i, j, by], which means "take DT, subset rows using i, then calculate j grouped by by".

Key concepts include:

Creating a data.table: `DT <- data.table(V1=c(1L,2L), V2=LETTERS[1:3], V3=round(rnorm(4),4), V4=1:12)`
Subsetting rows: `DT[3:5,]` or `DT[V2 == "A"]`
Selecting columns: `DT[, V2]` or `DT[, .(V2, V3)]`
Calling functions: `DT[, sum(V1)]` or `DT[, .(sum(V1), sd(V3))]`
Grouping: `DT[, .(V4.Sum = sum(V4)), by=V1]` or `DT[, .(V4.Sum = sum(V4)), by=.(V1, V2)]`
Assigning new column names: `DT[, .(Aggregate = sum(V1), Sd.V3 = sd(V3))]`
Using .N to get the total number of observations: `DT[,.N,by=V1]`
Adding/updating columns by reference: `DT[, V1 := round(exp(V1),2)]` or `DT[, c("V1","V2") := list(round(exp(V1),2), LETTERS[4:6])]`
Removing columns: `DT[, V1 := NULL]` or `DT[, c("V1","V2") := NULL]`
Indexing and keys: `setkey(DT, V2)` and `DT["A"]` or `DT[c("A","C")]`
Using mult and nomatch arguments: `DT["A", mult ="first"]` or `DT[c("A","D"), nomatch = 0]`
Using by=.EACHI: `DT[c("A","C"), sum(V4), by=.EACHI]`