Package com.hw.langchain.llms.base

Class BaseLLM

java.lang.Object

com.hw.langchain.llms.base.BaseLLM

All Implemented Interfaces:

BaseLanguageModel

Direct Known Subclasses:

BaseOpenAI, LLM, Ollama, OpenAIChat

public abstract class BaseLLM
extends Object
implements BaseLanguageModel

LLM wrapper should take in a prompt and return a string.

Author:

HamaWhite

Constructor Summary

Constructors

Constructor	Description
BaseLLM()

Method Summary

Modifier and Type	Method	Description
List<reactor.core.publisher.Flux<AsyncLLMResult>>	asyncGeneratePrompt(List<PromptValue> prompts, List<String> stop)	Take in a list of prompt values and return an Flux for every PromptValue.
protected abstract reactor.core.publisher.Flux<AsyncLLMResult>	asyncInnerGenerate(List<String> prompts, List<String> stop)	Run the LLM on the given prompts async.
reactor.core.publisher.Flux<String>	asyncPredict(String text, List<String> stop)	Predict text from text async.
String	call(String prompt)
String	call(String prompt, List<String> stop)	Check Cache and run the LLM on the given prompt and input.
LLMResult	generate(List<String> prompts, List<String> stop)	Run the LLM on the given prompt and input.
LLMResult	generatePrompt(List<PromptValue> prompts, List<String> stop)	Take in a list of prompt values and return an LLMResult.
protected abstract LLMResult	innerGenerate(List<String> prompts, List<String> stop)	Run the LLM on the given prompts.
abstract String	llmType()	Return type of llm.
String	predict(String text, List<String> stop)	Predict text from text.
BaseMessage	predictMessages(List<BaseMessage> messages, List<String> stop)	Predict message from messages.

Methods inherited from class java.lang.Object

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

Methods inherited from interface com.hw.langchain.base.language.BaseLanguageModel

asyncGeneratePrompt, asyncPredict, asyncPredictMessages, predict, predictMessages

Constructor Details

BaseLLM

public BaseLLM()

Method Details

llmType

public abstract String llmType()

Return type of llm.

innerGenerate

protected abstract LLMResult innerGenerate(List<String> prompts,
 List<String> stop)

Run the LLM on the given prompts.

asyncInnerGenerate

protected abstract reactor.core.publisher.Flux<AsyncLLMResult> asyncInnerGenerate(List<String> prompts,
 List<String> stop)

Run the LLM on the given prompts async.

call

public String call(String prompt,
 List<String> stop)

Check Cache and run the LLM on the given prompt and input.

call

public String call(String prompt)

generate

public LLMResult generate(List<String> prompts,
 List<String> stop)

Run the LLM on the given prompt and input.

generatePrompt

public LLMResult generatePrompt(List<PromptValue> prompts,
 List<String> stop)

Description copied from interface: BaseLanguageModel

Take in a list of prompt values and return an LLMResult.

Specified by:

generatePrompt in interface BaseLanguageModel

asyncGeneratePrompt

public List<reactor.core.publisher.Flux<AsyncLLMResult>> asyncGeneratePrompt(List<PromptValue> prompts,
 List<String> stop)

Description copied from interface: BaseLanguageModel

Take in a list of prompt values and return an Flux for every PromptValue.

Specified by:

asyncGeneratePrompt in interface BaseLanguageModel

predict

public String predict(String text,
 List<String> stop)

Description copied from interface: BaseLanguageModel

Predict text from text.

Specified by:

predict in interface BaseLanguageModel

asyncPredict

public reactor.core.publisher.Flux<String> asyncPredict(String text,
 List<String> stop)

Description copied from interface: BaseLanguageModel

Predict text from text async.

Specified by:

asyncPredict in interface BaseLanguageModel

predictMessages

public BaseMessage predictMessages(List<BaseMessage> messages,
 List<String> stop)

Description copied from interface: BaseLanguageModel

Predict message from messages.

Specified by:

predictMessages in interface BaseLanguageModel