reduce: Summarise data (with local reduction)

In jeblundell/multiplyr: Data Manipulation with Parallelism and Shared Memory Matrices

Description

summarise is used to summarise data on each node: reduce is then used to ensure that there's one overall summary

Usage

reduce(.self, ..., auto_compact = NULL)

reduce_(.self, ..., .dots, auto_compact = NULL)

Arguments

.self	Data frame
...	Additional parameters
auto_compact	Compact data after operation
.dots	Workaround for non-standard evaluation

Details

When data have not been grouped, calling summary(...) will result in each node summarising the data it has available. This means that if there are 3 nodes in the cluster, then there will be 3 summary values. reduce is used to bring all those together to a single value.

Value

Data frame

See Also

Other data manipulations: mutate, nsa, summarise, transmute, within_group, within_node

Examples

dat <- Multiplyr (x = 1:100)
dat %>% summarise (N = length(x))
dat %>% shutdown()

dat <- Multiplyr (x = 1:100)
dat %>% summarise (N = length(x)) %>% reduce(N = sum(N))
dat %>% shutdown()

jeblundell/multiplyr documentation built on May 19, 2019, 12:39 a.m.