Package com.google.genai.types
Class ReinforcementTuningHyperParameters
java.lang.Object
com.google.genai.JsonSerializable
com.google.genai.types.ReinforcementTuningHyperParameters
Hyperparameters for Reinforcement Tuning.
-
Nested Class Summary
Nested ClassesModifier and TypeClassDescriptionstatic classBuilder for ReinforcementTuningHyperParameters. -
Field Summary
Fields inherited from class com.google.genai.JsonSerializable
MAX_READ_LENGTH_PROPERTY -
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionabstract Optional<AdapterSize>Adapter size for Reinforcement Tuning.Batch size for the tuning job.builder()Instantiates a builder for ReinforcementTuningHyperParameters.How often (in steps) to save checkpoints during training.Number of training epochs for the tuning job.How often (in steps) to evaluate the tuning job during training.Deserializes a JSON string to a ReinforcementTuningHyperParameters object.Learning rate multiplier for Reinforcement Learning.The maximum number of tokens to generate per prompt.Number of different responses to generate per prompt during tuning.abstract Optional<ReinforcementTuningThinkingLevel>Indicates the maximum thinking depth.Creates a builder with the same values as this instance.Methods inherited from class com.google.genai.JsonSerializable
fromJsonNode, fromJsonString, objectMapper, setMaxReadLength, stringToJsonNode, toJson, toJsonNode, toJsonString
-
Constructor Details
-
ReinforcementTuningHyperParameters
public ReinforcementTuningHyperParameters()
-
-
Method Details
-
epochCount
Number of training epochs for the tuning job. -
learningRateMultiplier
Learning rate multiplier for Reinforcement Learning. -
adapterSize
Adapter size for Reinforcement Tuning. -
samplesPerPrompt
Number of different responses to generate per prompt during tuning. -
batchSize
Batch size for the tuning job. How many prompts to process at a train step. If not set, the batch size will be determined automatically. -
evaluateInterval
How often (in steps) to evaluate the tuning job during training. If not set, evaluation will run per epoch. -
checkpointInterval
How often (in steps) to save checkpoints during training. If not set, one checkpoint per epoch will be saved. -
maxOutputTokens
The maximum number of tokens to generate per prompt. If not set, defaults to 32768. -
thinkingLevel
Indicates the maximum thinking depth. Use with earlier models shall result in error. -
builder
Instantiates a builder for ReinforcementTuningHyperParameters. -
toBuilder
Creates a builder with the same values as this instance. -
fromJson
Deserializes a JSON string to a ReinforcementTuningHyperParameters object.
-