pypomp.Pomp.prune

Pomp.prune(n: int = 1, refill: bool = True) → None

Keep the top n parameter sets by log-likelihood.

Discards poorly performing starting points after an estimation run, focusing subsequent work on the most promising candidates. If refill is True, the surviving sets are duplicated to restore the original number of replicates.

Parameters:

• n (int, optional) – Number of top parameter sets to retain. Defaults to 1.

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