A Split-Apply-Combine strategy to parallelize the evaluation of a function.
big_parallelize(
X,
p.FUN,
p.combine = NULL,
ind = cols_along(X),
ncores = nb_cores(),
...
)An object of class FBM.
The function to be applied to each subset matrix.
It must take a Filebacked Big Matrix as first argument and
ind, a vector of indices, which are used to split the data.
For example, if you want to apply a function to X[ind.row, ind.col],
you may use X[ind.row, ind.col[ind]] in a.FUN.
Function to combine the results with do.call.
This function should accept multiple arguments (...). For example, you
can use c, cbind, rbind. This package also provides function plus
to add multiple arguments together. The default is NULL, in which case
the results are not combined and are returned as a list, each element being
the result of a block.
Initial vector of subsetting indices. Default is the vector of all column indices.
Number of cores used. Default doesn't use parallelism. You may use nb_cores.
Extra arguments to be passed to p.FUN.
Return a list of ncores elements, each element being the result of
one of the cores, computed on a block. The elements of this list are then
combined with do.call(p.combine, .) if p.combined is given.
This function splits indices in parts, then apply a given function to each part and finally combine the results.
if (FALSE) # CRAN is super slow when parallelism.
X <- big_attachExtdata()
### Computation on all the matrix
true <- big_colstats(X)
#> Error in as.list.environment(parent.frame()): object 'X' not found
big_colstats_sub <- function(X, ind) {
big_colstats(X, ind.col = ind)
}
# 1. the computation is split along all the columns
# 2. for each part the computation is done, using `big_colstats`
# 3. the results (data.frames) are combined via `rbind`.
test <- big_parallelize(X, p.FUN = big_colstats_sub,
p.combine = 'rbind', ncores = 2)
#> Error in ncol(x): object 'X' not found
all.equal(test, true)
#> Error in all.equal(test, true): object 'test' not found
### Computation on a part of the matrix
n <- nrow(X)
#> Error in nrow(X): object 'X' not found
m <- ncol(X)
#> Error in ncol(X): object 'X' not found
rows <- sort(sample(n, n/2)) # sort to provide some locality in accesses
#> Error in sample(n, n/2): object 'n' not found
cols <- sort(sample(m, m/2)) # idem
#> Error in sample(m, m/2): object 'm' not found
true2 <- big_colstats(X, ind.row = rows, ind.col = cols)
#> Error in as.list.environment(parent.frame()): object 'X' not found
big_colstats_sub2 <- function(X, ind, rows, cols) {
big_colstats(X, ind.row = rows, ind.col = cols[ind])
}
# This doesn't work because, by default, the computation is spread
# along all columns. We must explictly specify the `ind` parameter.
tryCatch(big_parallelize(X, p.FUN = big_colstats_sub2,
p.combine = 'rbind', ncores = 2,
rows = rows, cols = cols),
error = function(e) message(e))
#> Error in ncol(x): object 'X' not found
# This now works, using `ind = seq_along(cols)`.
test2 <- big_parallelize(X, p.FUN = big_colstats_sub2,
p.combine = 'rbind', ncores = 2,
ind = seq_along(cols),
rows = rows, cols = cols)
#> Error in assert_one_int(total_len): object 'cols' not found
all.equal(test2, true2)
#> Error in all.equal(test2, true2): object 'test2' not found