GEPAOptimizer
constructor(reflectionModel: LLModel, storagePath: ResilientPath, maxIterations: Int, minibatchSize: Int, componentSelection: ComponentSelection, enableCrossover: Boolean, maxMergeInvocations: Int = 5, randomSeed: Long = 42, labelExtractor: (TrainSetItem<Input, InputLabel>) -> String)
Parameters
reflection
LLM model for reflection / instruction proposal. Calls are routed through the training DSL's tracked executePrompt, so timing and consumption are recorded under the appropriate iteration stage.
storage
Path for saving the optimization artifact.
max
Maximum number of evolution iterations.
minibatch
Number of training items sampled per iteration for reflection.
component
How to select which modules to update each iteration.
enable
Whether to attempt merging complementary candidates during the loop.
max
Maximum number of crossover attempts per optimization run.
random
Seed for reproducibility.
label
Converts a dataset item's label to a string for the reflection LM.