drake (version 6.2.1)

max_useful_jobs: Deprecated function

Description

Do not use this function. drake's parallel algorithm has changed since version 5.1.2, so max_useful_jobs() will give you the wrong idea of how many jobs to use. Instead, use the predict_runtime() function with a sensible value for known_times and default_time to cover any targets not built so far.

Usage

max_useful_jobs(config = drake::read_drake_config(),
  imports = c("files", "all", "none"), from_scratch = FALSE)

Arguments

config

internal configuration list of make(...), produced also with drake_config().

imports

Set the imports argument to change your assumptions about how fast objects/files are imported.

from_scratch

logical, whether to assume the next make() will run from scratch so that all targets are attempted.

Value

A numeric scalar, the maximum number of useful jobs for make(..., jobs = ...).

Details

Deprecated on May 4, 2018.

See Also

predict_runtime()

Examples

Run this code
# NOT RUN {
# Do not use this function. Use predict_runtime() instead.
# Pay special attention to the force_times and default_time
# arguments.
# }

Run the code above in your browser using DataCamp Workspace