Package com.microsoft.semantickernel.textcompletion

Class CompletionRequestSettings

java.lang.Object

com.microsoft.semantickernel.textcompletion.CompletionRequestSettings

Direct Known Subclasses:

ChatRequestSettings

public class CompletionRequestSettings
extends Object

Settings for a text completion request

Constructor Summary

Constructors

Constructor	Description
CompletionRequestSettings()	Create a new settings object with default values.
CompletionRequestSettings(double temperature, double topP, double presencePenalty, double frequencyPenalty, int maxTokens)	Create a new settings object with the given values.
CompletionRequestSettings(double temperature, double topP, double presencePenalty, double frequencyPenalty, int maxTokens, int bestOf, String user, List<String> stopSequences)	Create a new settings object with the given values.

Method Summary

Modifier and Type	Method	Description
static CompletionRequestSettings	fromCompletionConfig(PromptTemplateConfig.CompletionConfig config)	Create a new settings object with the values from another settings object.
Integer	getBestOf()	The maximum number of completions to generate for each prompt.
double	getFrequencyPenalty()	Number between -2.0 and 2.0.
int	getMaxTokens()	The maximum number of tokens to generate in the completion.
double	getPresencePenalty()	Number between -2.0 and 2.0.
List<String>	getStopSequences()	Sequences where the completion will stop generating further tokens.
double	getTemperature()	Temperature controls the randomness of the completion.
double	getTopP()	TopP controls the diversity of the completion.
String	getUser()	A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

CompletionRequestSettings

public CompletionRequestSettings(double temperature,
                                 double topP,
                                 double presencePenalty,
                                 double frequencyPenalty,
                                 int maxTokens)

Create a new settings object with the given values.

Parameters:

temperature - Temperature controls the randomness of the completion.

topP - TopP controls the diversity of the completion.

presencePenalty - Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.

frequencyPenalty - Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.

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

CompletionRequestSettings

public CompletionRequestSettings(double temperature,
                                 double topP,
                                 double presencePenalty,
                                 double frequencyPenalty,
                                 int maxTokens,
                                 int bestOf,
                                 String user,
                                 List<String> stopSequences)

Create a new settings object with the given values.

Parameters:

temperature - Temperature controls the randomness of the completion.

topP - TopP controls the diversity of the completion.

presencePenalty - Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.

frequencyPenalty - Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.

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

bestOf - The maximum number of completions to generate for each prompt. This is used by the CompletionService to generate multiple completions for a single prompt.

user - A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse

stopSequences - Sequences where the completion will stop generating further tokens.

CompletionRequestSettings

public CompletionRequestSettings()

Create a new settings object with default values.

Method Detail

fromCompletionConfig

public static CompletionRequestSettings fromCompletionConfig(PromptTemplateConfig.CompletionConfig config)

Create a new settings object with the values from another settings object.

Parameters:

config - The config to copy values from

getTemperature

public double getTemperature()

Temperature controls the randomness of the completion. The higher the temperature, the more random the completion

getTopP

public double getTopP()

TopP controls the diversity of the completion. The higher the TopP, the more diverse the completion.

getPresencePenalty

public double getPresencePenalty()

Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.

getFrequencyPenalty

public double getFrequencyPenalty()

Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.

getMaxTokens

public int getMaxTokens()

The maximum number of tokens to generate in the completion.

getStopSequences

public List<String> getStopSequences()

Sequences where the completion will stop generating further tokens.

getBestOf

public Integer getBestOf()

The maximum number of completions to generate for each prompt. This is used by the CompletionService to generate multiple completions for a single prompt.

getUser

public String getUser()

A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse