Parallel Designed Experiment
Parameters
Optional override for the base experiment name used to derive per-design-point experiment names ("null (the default), the base name comes from the template model's auto-generated Experiment_<counter> identity — which is JVM-counter-driven and changes between runs. Callers that want deterministic, human-readable per- point names (e.g. the Experiment app: anchored to the analysis name) should pass an explicit value. Persists to EXPERIMENT.exp_name in the KSL database, so the column values match the on-disk folder / file names.
When true (the default, preserving the original behaviour), every design point gets its own subdirectory under pathToOutputDirectory named <experimentName>_DP_<n>_OutputDir; each subdir contains that point's kslOutput.txt (and any per-point CSV / plot artifacts the model writes). When false, every per-point model writes directly into pathToOutputDirectory and the diagnostic log uses a point-distinguished filename (kslOutput_DP_<n>.txt) so concurrent writers don't clash and re-runs overwrite cleanly. False is the right default for callers that rely on the kslDb for per-point results.
Constructors
Convenience constructor for two-level factor setting maps.
Convenience constructor for a function that creates a fresh model.
Properties
A default value for the number of replications per design point.
The experimental design associated with the executed design points.
The number of design points that have been executed.
The names of the responses or counters in the template model.
Returns the executed runs, one run for each design point simulated.
The active design-point random stream policy.
Functions
Request cancellation of the design point with the given 1-based pointId. If that point's coroutine is currently running (or queued) it will be cancelled; the resulting outcome carries wasCancelled = true so the commit phase fires onDesignPointCancelled instead of treating it as a failure, and skips the database write entirely.
Clears previously executed simulation runs.
Returns a map of design-point label to per-replication observations for responseName.
Returns regression data for the supplied response and linear model.
Returns regression data as a dataframe for the supplied response and linear model.
Performs OLS regression for the supplied response and linear model.
Returns replicated design point information in execution order as a data frame.
Returns a data frame with columns (point, exp_name, rep_id, factor1, factor2, ..., factorN).
Returns a data frame with one response joined to replicated design points.
Returns a data frame with selected responses joined to replicated design points.
Returns a data frame with columns (point, exp_name, rep_id, responseName).
Writes the joined design-point and response results to a CSV file.
Simulates the design points presented by iterator concurrently.
Simulates all design points concurrently.
Makes every design point start from the same random stream block.
Assigns non-overlapping pre-run sub-stream advances to design points.