Package com.google.genai.types

Class PreferenceOptimizationHyperParameters

java.lang.Object

com.google.genai.JsonSerializable

com.google.genai.types.PreferenceOptimizationHyperParameters

@InternalApi
public abstract class PreferenceOptimizationHyperParameters
extends JsonSerializable

Hyperparameters for Preference Optimization. This data type is not supported in Gemini API.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	PreferenceOptimizationHyperParameters.Builder	Builder for PreferenceOptimizationHyperParameters.

Field Summary

Fields inherited from class com.google.genai.JsonSerializable

MAX_READ_LENGTH_PROPERTY

Constructor Summary

Constructors

Constructor	Description
PreferenceOptimizationHyperParameters()

Method Summary

Modifier and Type	Method	Description
abstract Optional<AdapterSize>	adapterSize()	Optional.
abstract Optional<Double>	beta()	Optional.
static PreferenceOptimizationHyperParameters.Builder	builder()	Instantiates a builder for PreferenceOptimizationHyperParameters.
abstract Optional<Long>	epochCount()	Optional.
static PreferenceOptimizationHyperParameters	fromJson(String jsonString)	Deserializes a JSON string to a PreferenceOptimizationHyperParameters object.
abstract Optional<Double>	learningRateMultiplier()	Optional.
abstract PreferenceOptimizationHyperParameters.Builder	toBuilder()	Creates a builder with the same values as this instance.

Methods inherited from class com.google.genai.JsonSerializable

setMaxReadLength, stringToJsonNode, toJson

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

PreferenceOptimizationHyperParameters

public PreferenceOptimizationHyperParameters()

Method Details

adapterSize

public abstract Optional<AdapterSize> adapterSize()

Optional. Adapter size for preference optimization.

beta

public abstract Optional<Double> beta()

Optional. Weight for KL Divergence regularization.

epochCount

public abstract Optional<Long> epochCount()

Optional. Number of complete passes the model makes over the entire training dataset during training.

learningRateMultiplier

public abstract Optional<Double> learningRateMultiplier()

Optional. Multiplier for adjusting the default learning rate.

builder

public static PreferenceOptimizationHyperParameters.Builder builder()

Instantiates a builder for PreferenceOptimizationHyperParameters.

toBuilder

public abstract PreferenceOptimizationHyperParameters.Builder toBuilder()

Creates a builder with the same values as this instance.

fromJson

public static PreferenceOptimizationHyperParameters fromJson(String jsonString)

Deserializes a JSON string to a PreferenceOptimizationHyperParameters object.