Share via

PromptExecutionSettings.Builder Class

Package:
com.microsoft.semantickernel.orchestration
Maven Artifact:
com.microsoft.semantic-kernel:semantickernel-api:1.4.0
  • java.lang.Object
    • com.microsoft.semantickernel.orchestration.PromptExecutionSettings.Builder

public static class PromptExecutionSettings.Builder

Builder for PromptExecutionSettings.

Constructor Summary

Constructor Description
Builder()

Method Summary

Modifier and Type Method and Description
PromptExecutionSettings build()

Build the PromptExecutionSettings.
Builder withBestOf(int bestOf)

Set the best of setting for prompt execution.
Builder withFrequencyPenalty(double frequencyPenalty)

Set the frequency penalty setting for prompt execution.
Builder withJsonSchemaResponseFormat(Class<?> responseFormat)

Set the response format to use a json schema generated for the given class.
Builder withMaxTokens(int maxTokens)

Set the maximum number of tokens to generate in the output.
Builder withModelId(String modelId)

Set the id of the model to use for prompt execution.
Builder withPresencePenalty(double presencePenalty)

Set the presence penalty setting for prompt execution.
Builder withResponseFormat(ResponseFormat responseFormat)

Set the response format to use for prompt execution.
Builder withResponseFormat(ResponseFormat.Type responseFormat)

Set the response format to use for prompt execution.
Builder withResultsPerPrompt(int resultsPerPrompt)

Set the number of results to generate for each prompt.
Builder withServiceId(String serviceId)

Set the id of the AI service to use for prompt execution.
Builder withStopSequences(List<String> stopSequences)

Set the stop sequences to use for prompt execution.
Builder withTemperature(double temperature)

Set the temperature setting for prompt execution.
Builder withTokenSelectionBiases(Map<Integer,Integer> tokenSelectionBiases)

Set the token selection biases to use for prompt execution.
Builder withTopP(double topP)

Set the topP setting for prompt execution.
Builder withUser(String user)

Set the user to associate with the prompt execution.

Methods inherited from java.lang.Object

java.lang.Object.clone java.lang.Object.equals java.lang.Object.finalize java.lang.Object.getClass java.lang.Object.hashCode java.lang.Object.notify java.lang.Object.notifyAll java.lang.Object.toString java.lang.Object.wait java.lang.Object.wait java.lang.Object.wait

Constructor Details

Builder

public Builder()

Method Details

build

public PromptExecutionSettings build()

Build the PromptExecutionSettings.

Returns:

A new PromptExecutionSettings from this builder.

withBestOf

public PromptExecutionSettings.Builder withBestOf(int bestOf)

Set the best of setting for prompt execution. The value is clamped to the range [1, Integer.MAX_VALUE], and the default is 1.

Parameters:

bestOf - The best of setting for prompt execution.

Returns:

This builder.

withFrequencyPenalty

public PromptExecutionSettings.Builder withFrequencyPenalty(double frequencyPenalty)

Set the frequency penalty setting for prompt execution. The value is clamped to the range [-2.0, 2.0], and the default is 0.0.

Parameters:

frequencyPenalty - The frequency penalty setting for prompt execution.

Returns:

This builder.

withJsonSchemaResponseFormat

public PromptExecutionSettings.Builder withJsonSchemaResponseFormat(Class responseFormat)

Set the response format to use a json schema generated for the given class. The name of the response format will be the name of the class.

Parameters:

responseFormat - The response format type.

Returns:

This builder.

withMaxTokens

public PromptExecutionSettings.Builder withMaxTokens(int maxTokens)

Set the maximum number of tokens to generate in the output. The value is clamped to the range [1, Integer.MAX_VALUE], and the default is 256.

Parameters:

maxTokens - The maximum number of tokens to generate in the output.

Returns:

This builder.

withModelId

public PromptExecutionSettings.Builder withModelId(String modelId)

Set the id of the model to use for prompt execution.

Parameters:

modelId - The id of the model to use for prompt execution.

Returns:

This builder.

withPresencePenalty

public PromptExecutionSettings.Builder withPresencePenalty(double presencePenalty)

Set the presence penalty setting for prompt execution. The value is clamped to the range [-2.0, 2.0], and the default is 0.0.

Parameters:

presencePenalty - The presence penalty setting for prompt execution.

Returns:

This builder.

withResponseFormat

public PromptExecutionSettings.Builder withResponseFormat(ResponseFormat responseFormat)

Set the response format to use for prompt execution.

Parameters:

responseFormat - The response format to use for prompt execution.

Returns:

This builder.

withResponseFormat

public PromptExecutionSettings.Builder withResponseFormat(ResponseFormat.Type responseFormat)

Set the response format to use for prompt execution.

Parameters:

responseFormat - The response format to use for prompt execution.

Returns:

This builder.

withResultsPerPrompt

public PromptExecutionSettings.Builder withResultsPerPrompt(int resultsPerPrompt)

Set the number of results to generate for each prompt. The value is clamped to the range [1, Integer.MAX_VALUE], and the default is 1.

Parameters:

resultsPerPrompt - The number of results to generate for each prompt.

Returns:

This builder.

withServiceId

public PromptExecutionSettings.Builder withServiceId(String serviceId)

Set the id of the AI service to use for prompt execution.

Parameters:

serviceId - The id of the AI service to use for prompt execution.

Returns:

This builder.

withStopSequences

public PromptExecutionSettings.Builder withStopSequences(List stopSequences)

Set the stop sequences to use for prompt execution.

Parameters:

stopSequences - The stop sequences to use for prompt execution.

Returns:

This builder.

withTemperature

public PromptExecutionSettings.Builder withTemperature(double temperature)

Set the temperature setting for prompt execution. The value is clamped to the range [0.0, 2.0], and the default is 1.0.

Parameters:

temperature - The temperature setting for prompt execution.

Returns:

This builder.

withTokenSelectionBiases

public PromptExecutionSettings.Builder withTokenSelectionBiases(Map tokenSelectionBiases)

Set the token selection biases to use for prompt execution. The bias values are clamped to the range [-100, 100].

Parameters:

tokenSelectionBiases - The token selection biases to use for prompt execution.

Returns:

This builder.

withTopP

public PromptExecutionSettings.Builder withTopP(double topP)

Set the topP setting for prompt execution. The value is clamped to the range [0.0, 1.0], and the default is 1.0.

Parameters:

topP - The topP setting for prompt execution.

Returns:

This builder.

withUser

public PromptExecutionSettings.Builder withUser(String user)

Set the user to associate with the prompt execution.

Parameters:

user - The user to associate with the prompt execution.

Returns:

This builder.

Applies to