R：ddply仅返回矩阵列的第一列

Question

I have a data.frame D where some columns are matrices, eg

> head(round(D$equationRTs, 1))
     [,1] [,2] [,3] [,4] [,5] [,6] [,7]
[1,]  2.9  2.1  3.2  3.5   NA   NA   NA
[2,]  2.8   NA   NA   NA   NA   NA   NA
[3,]  3.4  2.4   NA   NA   NA   NA   NA
[4,]  2.7  2.9  1.9   NA   NA   NA   NA
[5,]  3.6  2.6  2.4  2.4  3.4  2.8   NA
[6,]  2.4  2.0  3.3  2.8  2.8  2.6  3.6
...

> dim(D$equationRTs)
[1] 11277     7

> typeof(D$equationRTs)
[1] "double"

However, when I do a ddply to subset D :

my_function = function(df) {
    # Let's see what ddply passes to this function:
    print(head(round(df$equationRTs, 1)))
    print(dim(df$equationRTs))
    print(typeof(df$equationRTs))
}

D = ddply(D, .(id), my_function)

it appears that only the first column is passed to my_function as vector:

[1] 2.9 2.8 3.4 2.7 3.6 2.4
NULL
[1] "double"

Column 2-6 are just gone. What's going on here and how do I make the matrix column stay intact when subsets are passed to my_function ?

Bonus: it seems that ddply is doing something like D$equationRTs[id==x] which indeed returns the first column of the matrix whereas D$equationRTs[id==x, ] returns the matrix.

Answer 1

You can simply manually code what ddply does (or should do...).

So replace the old

D = ddply(D, .(id), my_function)

with

for(id in levels(D$id)) {
    D[D$id == id, ] = check_quality(D[D$id == id, ])
}

The manual version has the desired behavior.