time_methods: Time methods

In CellBench: Construct Benchmarks for Single Cell Analysis Methods

Description

time_methods() take either lists of datasets or benchmark_timing_tbl objects and applies a list of functions. The output is a benchmark_timing_tbl where each method has been applied to each dataset or preceeding result. Unlike apply_methods(), time_methods() is always single threaded as to produce fair and more consistent timings.

Usage

time_methods(x, fn_list, name = NULL, suppress.messages = TRUE)

## S3 method for class 'list'
time_methods(x, fn_list, name = NULL, suppress.messages = TRUE)

## S3 method for class 'benchmark_timing_tbl'
time_methods(x, fn_list, name = NULL, suppress.messages = TRUE)

Arguments

x	the list of data or benchmark timing tibble to apply methods to
fn_list	the list of methods to be applied
name	(optional) the name of the column for methods applied
suppress.messages	TRUE if messages from running methods should be suppressed

Value

benchmark_timing_tbl object containing results from methods applied, the first column is the name of the dataset as factors, middle columns contain method names as factors and the final column is a list of lists containing the results of applying the methods and timings from calling system.time().

See Also

apply_methods

Examples

datasets <- list(
    set1 = 1:1e7
)

transform <- list(
    sqrt = sqrt,
    log = log
)

time_methods(datasets, transform) %>%
    unpack_timing() # extract timings out of list

CellBench documentation built on Nov. 8, 2020, 5:11 p.m.