Crate hyperopt_pruners 

hyperopt-pruners

Early-stopping policies for hyperopt-rs. A Pruner inspects a trial’s intermediate reports against the rest of the study and decides whether continuing is worth it. Pruners are queried from the user’s objective via hyperopt_core::TrialContext::should_prune.

• NopPruner — never prunes (the default; keeps the API shape consistent).
• MedianPruner — prune when a trial is worse than the median of others at the same step.
• SuccessiveHalvingPruner — ASHA-style rung promotion (the advanced option).

Structs

MedianPruner

Prunes a trial when its latest intermediate value is worse than the median of other trials’ values at the same step.

NopPruner

A no-op pruner: should_prune always returns false. Use it when early-stopping isn’t wanted but the objective still calls should_prune() so the same code runs with and without pruning.

SuccessiveHalvingPruner

ASHA-style (Asynchronous Successive Halving) pruner — the “advanced” option.