google-genai-kotlin/com.google.genai.kotlin.types/SupervisedHyperParameters

SupervisedHyperParameters

@Serializable
data class SupervisedHyperParameters(val adapterSize: AdapterSize? = null, val batchSize: Long? = null, val epochCount: Long? = null, val learningRate: Double? = null, val learningRateMultiplier: Double? = null)

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

Constructors

Link copied to clipboard
constructor(adapterSize: AdapterSize? = null, batchSize: Long? = null, epochCount: Long? = null, learningRate: Double? = null, learningRateMultiplier: Double? = null)

Properties

Link copied to clipboard
val adapterSize: AdapterSize? = null

Optional. Adapter size for tuning.

Link copied to clipboard
val batchSize: Long? = null

Optional. Batch size for tuning. This feature is only available for open source models.

Link copied to clipboard
val epochCount: Long? = null

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

Link copied to clipboard
val learningRate: Double? = null

Optional. Learning rate for tuning. Mutually exclusive with learning_rate_multiplier. This feature is only available for open source models.

Link copied to clipboard
val learningRateMultiplier: Double? = null

Optional. Multiplier for adjusting the default learning rate. Mutually exclusive with learning_rate. This feature is only available for 1P models.