Class SingleReinforcementTuningRewardConfig.Builder
- Enclosing class:
- SingleReinforcementTuningRewardConfig
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionautoraterScorer(ReinforcementTuningAutoraterScorer autoraterScorer) Setter for autoraterScorer.autoraterScorer(ReinforcementTuningAutoraterScorer.Builder autoraterScorerBuilder) Setter for autoraterScorer builder.build()Clears the value of autoraterScorer field.Clears the value of cloudRunRewardScorer field.Clears the value of codeExecutionRewardScorer field.Clears the value of parseResponseConfig field.Clears the value of rewardName field.Clears the value of stringMatchRewardScorer field.cloudRunRewardScorer(ReinforcementTuningCloudRunRewardScorer cloudRunRewardScorer) Setter for cloudRunRewardScorer.cloudRunRewardScorer(ReinforcementTuningCloudRunRewardScorer.Builder cloudRunRewardScorerBuilder) Setter for cloudRunRewardScorer builder.codeExecutionRewardScorer(ReinforcementTuningCodeExecutionRewardScorer codeExecutionRewardScorer) Setter for codeExecutionRewardScorer.codeExecutionRewardScorer(ReinforcementTuningCodeExecutionRewardScorer.Builder codeExecutionRewardScorerBuilder) Setter for codeExecutionRewardScorer builder.parseResponseConfig(ReinforcementTuningParseResponseConfig parseResponseConfig) Setter for parseResponseConfig.parseResponseConfig(ReinforcementTuningParseResponseConfig.Builder parseResponseConfigBuilder) Setter for parseResponseConfig builder.rewardName(String rewardName) Setter for rewardName.stringMatchRewardScorer(ReinforcementTuningStringMatchRewardScorer stringMatchRewardScorer) Setter for stringMatchRewardScorer.stringMatchRewardScorer(ReinforcementTuningStringMatchRewardScorer.Builder stringMatchRewardScorerBuilder) Setter for stringMatchRewardScorer builder.
-
Constructor Details
-
Builder
public Builder()
-
-
Method Details
-
autoraterScorer
public abstract SingleReinforcementTuningRewardConfig.Builder autoraterScorer(ReinforcementTuningAutoraterScorer autoraterScorer) Setter for autoraterScorer.autoraterScorer: Scores parsed responses for autorater use cases by using a model to compute the reward.
-
autoraterScorer
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder autoraterScorer(ReinforcementTuningAutoraterScorer.Builder autoraterScorerBuilder) Setter for autoraterScorer builder.autoraterScorer: Scores parsed responses for autorater use cases by using a model to compute the reward.
-
clearAutoraterScorer
Clears the value of autoraterScorer field. -
rewardName
Setter for rewardName.rewardName: A unique reward name for identifying each single reinforcement tuning reward.
-
clearRewardName
Clears the value of rewardName field. -
parseResponseConfig
public abstract SingleReinforcementTuningRewardConfig.Builder parseResponseConfig(ReinforcementTuningParseResponseConfig parseResponseConfig) Setter for parseResponseConfig.parseResponseConfig: Defines how to parse sample response. For example, given a sample response for evaluating the reward, users might want to extract the text only between `` and `` in the sample response, and keeps only the last one in case there are multiple such tags. To achieve such a purpose, they can define a regex `".*(.*?)"` using the ReinforcementTuningParseResponseConfig.ResponseParseType.REGEX_EXTRACT parse type.
-
parseResponseConfig
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder parseResponseConfig(ReinforcementTuningParseResponseConfig.Builder parseResponseConfigBuilder) Setter for parseResponseConfig builder.parseResponseConfig: Defines how to parse sample response. For example, given a sample response for evaluating the reward, users might want to extract the text only between `` and `` in the sample response, and keeps only the last one in case there are multiple such tags. To achieve such a purpose, they can define a regex `".*(.*?)"` using the ReinforcementTuningParseResponseConfig.ResponseParseType.REGEX_EXTRACT parse type.
-
clearParseResponseConfig
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder clearParseResponseConfig()Clears the value of parseResponseConfig field. -
codeExecutionRewardScorer
public abstract SingleReinforcementTuningRewardConfig.Builder codeExecutionRewardScorer(ReinforcementTuningCodeExecutionRewardScorer codeExecutionRewardScorer) Setter for codeExecutionRewardScorer.codeExecutionRewardScorer: ReinforcementTuningCodeExecutionRewardScorer is used to score parsed responses for code execution use cases.
-
codeExecutionRewardScorer
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder codeExecutionRewardScorer(ReinforcementTuningCodeExecutionRewardScorer.Builder codeExecutionRewardScorerBuilder) Setter for codeExecutionRewardScorer builder.codeExecutionRewardScorer: ReinforcementTuningCodeExecutionRewardScorer is used to score parsed responses for code execution use cases.
-
clearCodeExecutionRewardScorer
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder clearCodeExecutionRewardScorer()Clears the value of codeExecutionRewardScorer field. -
stringMatchRewardScorer
public abstract SingleReinforcementTuningRewardConfig.Builder stringMatchRewardScorer(ReinforcementTuningStringMatchRewardScorer stringMatchRewardScorer) Setter for stringMatchRewardScorer.stringMatchRewardScorer: ReinforcementTuningStringMatchRewardScorer is used to score parsed responses for simple string matching use cases against reference answers.
-
stringMatchRewardScorer
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder stringMatchRewardScorer(ReinforcementTuningStringMatchRewardScorer.Builder stringMatchRewardScorerBuilder) Setter for stringMatchRewardScorer builder.stringMatchRewardScorer: ReinforcementTuningStringMatchRewardScorer is used to score parsed responses for simple string matching use cases against reference answers.
-
clearStringMatchRewardScorer
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder clearStringMatchRewardScorer()Clears the value of stringMatchRewardScorer field. -
cloudRunRewardScorer
public abstract SingleReinforcementTuningRewardConfig.Builder cloudRunRewardScorer(ReinforcementTuningCloudRunRewardScorer cloudRunRewardScorer) Setter for cloudRunRewardScorer.cloudRunRewardScorer: ReinforcementTuningCloudRunRewardScorer is used to score parsed responses by calling a Cloud Run service.
-
cloudRunRewardScorer
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder cloudRunRewardScorer(ReinforcementTuningCloudRunRewardScorer.Builder cloudRunRewardScorerBuilder) Setter for cloudRunRewardScorer builder.cloudRunRewardScorer: ReinforcementTuningCloudRunRewardScorer is used to score parsed responses by calling a Cloud Run service.
-
clearCloudRunRewardScorer
@CanIgnoreReturnValue public SingleReinforcementTuningRewardConfig.Builder clearCloudRunRewardScorer()Clears the value of cloudRunRewardScorer field. -
build
-