CODEWITHSUNDEEP
rapply
Heisenberg
Heisenberg

Reputation: 8816

An intuitive way to understand MARGIN in sweep and apply

I have always been confused by how MARGIN seems to mean 2 different things in sweep and apply. Consider:

m <- matrix(1:6, ncol = 2)
# The "- 1" operation is applied to all cells in each row
sweep(m, MARGIN = 1, 1, "-")
# The sum operation is applied to each column
apply(m, MARGIN = 1, sum)

Do you have a mnemonic device to understand this seemingly contradictory meaning of MARGIN?

Upvotes: 1

Views: 1140

Answers (2)

profPlum
profPlum

Reputation: 498

Regarding Inconsistency & LyzandeR's Answer: First of all OP you are correct they are indeed inconsistent!

Here is concrete proof/example of inconsistency:

Did you know that apply & sweep can be used to do the same thing? Well here's what happens when you do that:

(A = matrix(1:9, 3))
(B=sweep(A, MARGIN=1, 1:3, FUN='-'))
(C=apply(A, MARGIN=2, 1:3, FUN='-'))
stopifnot(all(B==C)) 
# they're equal, but MARGINS are opposite!!

The simple fact is that apply(A, MARGIN=1, ..., FUN=f) means: copy & apply FUN to each row of A, & sweep(A, MARGIN=1, ..., FUN=f) means: copy & apply FUN to each column of A.

My Mnemonic: I use apply() with negative MARGINs (that is apply(A, MARGIN=2, ...) becomes --> apply(A, MARGIN=-1, ...)), this makes the behavior consistent with sweep (in terms of absolute values) & has the added benefit of making the MARGIN argument consistent with the numpy/torch/matlab axis arguments!! e.g., torch.mean(arr, axis=0) == np.mean(arr, axis=0) == apply(arr, MARGIN=-1, mean) (save of course R's base-1 indexing)

It is easy to remember because you cannot actually use negative MARGINs with sweep()! So if you forget which function needs them it will immediately remind you with an error. ...Also negative numbers come before positive numbers just like (A)pply comes before (S)weep in the alphabet.

Upvotes: 0

LyzandeR
LyzandeR

Reputation: 37889

The MARGIN argument means exactly the same thing in both functions and that is row-wise operation. I have been confused with sweep many times in the past but I think you are confused with apply.

I am printing the matrix below so that it is easy to visually compare with apply and sweep later on:

> m
     [,1] [,2]
[1,]    1    4
[2,]    2    5
[3,]    3    6

First of all the sweep function does a row-wise operation when MARGIN is 1. I will slightly change the third argument so that this is more obvious:

> sweep(m, MARGIN = 1, 1:3, "-")
     [,1] [,2]
[1,]    0    3
[2,]    0    3
[3,]    0    3

In the above case number 1 was deducted from row 1, number 2 from row 2 and number 3 from row 3. So, clearly this is a row-wise operation.

Now let's see below the apply function:

> apply(m, MARGIN = 1, sum)
[1] 5 7 9

Clearly, the matrix has 3 rows and 2 columns. It is easy to imply that this is also a row-wise operation since we have 3 results i.e. the same as the number of rows. This is also confirmed if we check the numbers. Row 1 sums to 5, row 2 to 7 and row 3 to 9.

So, clearly MARGIN in both cases implies a row-wise operation.

Upvotes: 3

Related Questions