RFTHyperParameters ¶

class RFTHyperParameters(TypedDict, closed=True):
    epoch_count: NotRequired["capo_bedrock.types.epoch_count.EpochCount"]
    """<p> Number of training epochs to run during reinforcement fine-tuning. Higher values may improve performance but increase training time. </p>"""
    batch_size: NotRequired["capo_bedrock.types.rft_batch_size.RFTBatchSize"]
    """<p> Number of training samples processed in each batch during reinforcement fine-tuning (RFT) training. Larger batches may improve training stability. </p>"""
    learning_rate: NotRequired["capo_bedrock.types.rft_learning_rate.RFTLearningRate"]
    """<p> Learning rate for the reinforcement fine-tuning. Controls how quickly the model adapts to reward signals. </p>"""
    max_prompt_length: NotRequired[
        "capo_bedrock.types.rft_max_prompt_length.RFTMaxPromptLength"
    ]
    """<p> Maximum length of input prompts during RFT training, measured in tokens. Longer prompts allow more context but increase memory usage and training-time. </p>"""
    training_sample_per_prompt: NotRequired[
        "capo_bedrock.types.rft_training_sample_per_prompt.RFTTrainingSamplePerPrompt"
    ]
    """<p> Number of response samples generated per prompt during RFT training. More samples provide better reward signal estimation. </p>"""
    inference_max_tokens: NotRequired[
        "capo_bedrock.types.rft_inference_max_tokens.RFTInferenceMaxTokens"
    ]
    """<p> Maximum number of tokens the model can generate in response to each prompt during RFT training. </p>"""
    reasoning_effort: NotRequired["capo_bedrock.types.reasoning_effort.ReasoningEffort"]
    """<p> Level of reasoning effort applied during RFT training. Higher values may improve response quality but increase training time. </p>"""
    eval_interval: NotRequired["capo_bedrock.types.rft_eval_interval.RFTEvalInterval"]
    """<p> Interval between evaluation runs during RFT training, measured in training steps. More frequent evaluation provides better monitoring. </p>"""