Interface SingleReinforcementTuningRewardConfig

Single reinforcement tuning reward config.

interface SingleReinforcementTuningRewardConfig {
    autoraterScorer?: ReinforcementTuningAutoraterScorer;
    cloudRunRewardScorer?: ReinforcementTuningCloudRunRewardScorer;
    codeExecutionRewardScorer?: ReinforcementTuningCodeExecutionRewardScorer;
    parseResponseConfig?: ReinforcementTuningParseResponseConfig;
    rewardName?: string;
    stringMatchRewardScorer?: ReinforcementTuningStringMatchRewardScorer;
}

Properties

autoraterScorer? cloudRunRewardScorer? codeExecutionRewardScorer? parseResponseConfig? rewardName? stringMatchRewardScorer?

Properties

OptionalautoraterScorer

autoraterScorer?: ReinforcementTuningAutoraterScorer

Scores parsed responses for autorater use cases by using a model to compute the reward.

OptionalcloudRunRewardScorer

cloudRunRewardScorer?: ReinforcementTuningCloudRunRewardScorer

Scores parsed responses by calling a Cloud Run service.

OptionalcodeExecutionRewardScorer

codeExecutionRewardScorer?: ReinforcementTuningCodeExecutionRewardScorer

Scores parsed responses for code execution use cases.

OptionalparseResponseConfig

parseResponseConfig?: ReinforcementTuningParseResponseConfig

Defines how to parse sample response.

OptionalrewardName

rewardName?: string

A unique reward name used to identify each single reinforcement tuning reward.

OptionalstringMatchRewardScorer

stringMatchRewardScorer?: ReinforcementTuningStringMatchRewardScorer

Scores parsed responses for simple string matching use cases against reference answer without writing python code.

Settings

On This Page

Properties