Interface RFTHyperParameters.Builder

All Superinterfaces:
Buildable, CopyableBuilder<RFTHyperParameters.Builder,RFTHyperParameters>, SdkBuilder<RFTHyperParameters.Builder,RFTHyperParameters>, SdkPojo
Enclosing class:
RFTHyperParameters
@Mutable @NotThreadSafe public static interface RFTHyperParameters.Builder extends SdkPojo, CopyableBuilder<RFTHyperParameters.Builder,RFTHyperParameters>

  • Method Details

    • epochCount

      RFTHyperParameters.Builder epochCount(Integer epochCount)

      Number of training epochs to run during reinforcement fine-tuning. Higher values may improve performance but increase training time.

      Parameters:
      epochCount - Number of training epochs to run during reinforcement fine-tuning. Higher values may improve performance but increase training time.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • batchSize

      RFTHyperParameters.Builder batchSize(Integer batchSize)

      Number of training samples processed in each batch during reinforcement fine-tuning (RFT) training. Larger batches may improve training stability.

      Parameters:
      batchSize - Number of training samples processed in each batch during reinforcement fine-tuning (RFT) training. Larger batches may improve training stability.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • learningRate

      RFTHyperParameters.Builder learningRate(Float learningRate)

      Learning rate for the reinforcement fine-tuning. Controls how quickly the model adapts to reward signals.

      Parameters:
      learningRate - Learning rate for the reinforcement fine-tuning. Controls how quickly the model adapts to reward signals.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • maxPromptLength

      RFTHyperParameters.Builder maxPromptLength(Integer maxPromptLength)

      Maximum length of input prompts during RFT training, measured in tokens. Longer prompts allow more context but increase memory usage and training-time.

      Parameters:
      maxPromptLength - Maximum length of input prompts during RFT training, measured in tokens. Longer prompts allow more context but increase memory usage and training-time.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • trainingSamplePerPrompt

      RFTHyperParameters.Builder trainingSamplePerPrompt(Integer trainingSamplePerPrompt)

      Number of response samples generated per prompt during RFT training. More samples provide better reward signal estimation.

      Parameters:
      trainingSamplePerPrompt - Number of response samples generated per prompt during RFT training. More samples provide better reward signal estimation.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • inferenceMaxTokens

      RFTHyperParameters.Builder inferenceMaxTokens(Integer inferenceMaxTokens)

      Maximum number of tokens the model can generate in response to each prompt during RFT training.

      Parameters:
      inferenceMaxTokens - Maximum number of tokens the model can generate in response to each prompt during RFT training.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • reasoningEffort

      RFTHyperParameters.Builder reasoningEffort(String reasoningEffort)

      Level of reasoning effort applied during RFT training. Higher values may improve response quality but increase training time.

      Parameters:
      reasoningEffort - Level of reasoning effort applied during RFT training. Higher values may improve response quality but increase training time.
      Returns:
      Returns a reference to this object so that method calls can be chained together.
      See Also:

    • reasoningEffort

      RFTHyperParameters.Builder reasoningEffort(ReasoningEffort reasoningEffort)

      Level of reasoning effort applied during RFT training. Higher values may improve response quality but increase training time.

      Parameters:
      reasoningEffort - Level of reasoning effort applied during RFT training. Higher values may improve response quality but increase training time.
      Returns:
      Returns a reference to this object so that method calls can be chained together.
      See Also:

    • evalInterval

      RFTHyperParameters.Builder evalInterval(Integer evalInterval)

      Interval between evaluation runs during RFT training, measured in training steps. More frequent evaluation provides better monitoring.

      Parameters:
      evalInterval - Interval between evaluation runs during RFT training, measured in training steps. More frequent evaluation provides better monitoring.
      Returns:
      Returns a reference to this object so that method calls can be chained together.