MIPROv2Optimizer
constructor(metaModel: LLModel, autoMode: AutoRunMode = AutoRunMode.LIGHT, numCandidatesOverride: Int? = null, numTrialsOverride: Int? = null, maxBootstrappedDemos: Int = 4, maxTotalDemos: Int = 8, includeLabeledExamples: Boolean = true, proposerConfig: InstructionProposerConfig = InstructionProposerConfig(), storagePath: ResilientPath, randomSeed: Long = 42, parallelism: Int = 1)
Parameters
meta
LLM model for meta-LLM calls (instruction proposal, dataset summarization).
auto
Preset mode controlling defaults (LIGHT/MEDIUM/HEAVY).
num
Manual override for number of candidates per module.
num
Manual override for grid search trials.
max
Max demos per bootstrap run.
max
Total demo slots (bootstrapped and labeled).
include
Whether bootstrap uses labeled fallback.
proposer
Configuration for an instruction proposal.
storage
File path to save/load the optimization artifact.
random
Seed for reproducibility.
parallelism
Max concurrent operations (bootstrap runs, instruction proposals).