pypomp.core.pomp.Pomp.prune

Pomp.prune(n: int = 1, refill: bool = True)

Filters the current set of parameter replicates to keep only the top n performers based on their most recent log-likelihood estimates.

This method is commonly used after an estimation run (like pfilter or mif) to discard poorly performing parameter sets and focus subsequent computational effort on the most promising candidates. If refill is enabled, the kept parameters are duplicated to maintain the original number of replicates.

Parameters:

• n (int) – Number of top thetas to keep.

• refill (bool) – If True, repeat the top n thetas to match the previous number of theta sets.