Parallel test execution helper

Source: R/performance.R · Help: ?runHeteroTestsParallel

Executes a set of registered diagnostics in parallel when multiple CPU cores are available, falling back to sequential execution otherwise.

Usage

runHeteroTestsParallel(model, data, tests, n_cores = NULL)

Arguments

model

Fitted model object passed to each diagnostic. Must be compatible with the internal .test_factory registry.

data

Data frame used to fit model and supplied to each diagnostic.

tests

Character vector naming the diagnostics to execute.

n_cores

Optional positive integer giving the number of worker processes to spawn. Defaults to parallel::detectCores() - 1, with a lower bound of 1.

Value

Named list of test results in the order provided by tests. Each is the result runHeteroTests() gives for that test on the same call, its numbers computed the same way, since 0.18.0, and so identical() to it where the BLAS gives the same last bits on every call, as R's reference BLAS does (Intel MKL need not): a result whose data.name is "model", the name of the argument of the function the registry calls the test with, reports the formula of the model instead, and attr(, "extras")$streaming is FALSE, as this function runs the exact tests only. Before, White's test, its bootstrap, the studentized Breusch-Pagan statistic and Szroeter's test reported "model" here. A resampling test draws from a stream of its own (Details), and so does not return the value it returns in runHeteroTests(), which draws from the session's generator. The list is not a hetero_test_suite, and a test that fails stops the call, without the fallbacks of runHeteroTests().

Details

When n_cores exceeds one the diagnostics run on a socket cluster of that many workers (parallel::makeCluster()), started for the call, each of which loads the installed heteroTests. The model and the data travel to the workers with each task. With n_cores = 1, or when the default comes to one worker, the diagnostics run in the session.

Test i draws its random numbers from stream i of a L'Ecuyer-CMRG generator seeded by one draw from the session's generator, on the workers and in the session alike. A seeded call that runs a resampling test, such as "wild_bootstrap", therefore returns the same value each time and whatever n_cores, n_cores = 1 included. It is not the value that test returns when called directly from the same seed, which draws from the session's generator; both are bootstraps of the same data.

See also

cachedTest() for memoised execution of individual diagnostics.