Package software.amazon.awssdk.services.bedrock.model

Interface RFTConfig.Builder

All Superinterfaces:

Buildable, CopyableBuilder<RFTConfig.Builder,RFTConfig>, SdkBuilder<RFTConfig.Builder,RFTConfig>, SdkPojo

Enclosing class:

RFTConfig

@Mutable
@NotThreadSafe
public static interface RFTConfig.Builder
extends SdkPojo, CopyableBuilder<RFTConfig.Builder,RFTConfig>

Method Summary

Modifier and Type	Method	Description
default RFTConfig.Builder	graderConfig(Consumer<GraderConfig.Builder> graderConfig)	Configuration for the grader that evaluates model responses and provides reward signals during RFT training.
RFTConfig.Builder	graderConfig(GraderConfig graderConfig)	Configuration for the grader that evaluates model responses and provides reward signals during RFT training.
default RFTConfig.Builder	hyperParameters(Consumer<RFTHyperParameters.Builder> hyperParameters)	Hyperparameters that control the reinforcement fine-tuning training process, including learning rate, batch size, and epoch count.
RFTConfig.Builder	hyperParameters(RFTHyperParameters hyperParameters)	Hyperparameters that control the reinforcement fine-tuning training process, including learning rate, batch size, and epoch count.

Methods inherited from interface software.amazon.awssdk.utils.builder.CopyableBuilder

copy

Methods inherited from interface software.amazon.awssdk.utils.builder.SdkBuilder

applyMutation, build

Methods inherited from interface software.amazon.awssdk.core.SdkPojo

equalsBySdkFields, sdkFieldNameToField, sdkFields

Method Detail

graderConfig

RFTConfig.Builder graderConfig(GraderConfig graderConfig)

Configuration for the grader that evaluates model responses and provides reward signals during RFT training.

Parameters:

graderConfig - Configuration for the grader that evaluates model responses and provides reward signals during RFT training.

Returns:

Returns a reference to this object so that method calls can be chained together.

graderConfig

default RFTConfig.Builder graderConfig(Consumer<GraderConfig.Builder> graderConfig)

Configuration for the grader that evaluates model responses and provides reward signals during RFT training.

This is a convenience method that creates an instance of the GraderConfig.Builder avoiding the need to create one manually via GraderConfig.builder().

When the Consumer completes, SdkBuilder.build() is called immediately and its result is passed to graderConfig(GraderConfig).

Parameters:

graderConfig - a consumer that will call methods on GraderConfig.Builder

Returns:

Returns a reference to this object so that method calls can be chained together.

See Also:

graderConfig(GraderConfig)

hyperParameters

RFTConfig.Builder hyperParameters(RFTHyperParameters hyperParameters)

Hyperparameters that control the reinforcement fine-tuning training process, including learning rate, batch size, and epoch count.

Parameters:

hyperParameters - Hyperparameters that control the reinforcement fine-tuning training process, including learning rate, batch size, and epoch count.

Returns:

Returns a reference to this object so that method calls can be chained together.

hyperParameters

default RFTConfig.Builder hyperParameters(Consumer<RFTHyperParameters.Builder> hyperParameters)

Hyperparameters that control the reinforcement fine-tuning training process, including learning rate, batch size, and epoch count.

This is a convenience method that creates an instance of the RFTHyperParameters.Builder avoiding the need to create one manually via RFTHyperParameters.builder().

When the Consumer completes, SdkBuilder.build() is called immediately and its result is passed to hyperParameters(RFTHyperParameters).

Parameters:

hyperParameters - a consumer that will call methods on RFTHyperParameters.Builder

Returns:

Returns a reference to this object so that method calls can be chained together.

See Also:

hyperParameters(RFTHyperParameters)