JavaAI：LangChain4j学习（一） 集成SpringBoot和阿里通义千问DashScope

LangChain4j集成SpringBoot和阿里通义千问DashScope

• 使用资源
• 记录
• 不使用Spring Boot ，仅集成
• 集成Spring Boot 启动器
• • pom.xml ：
  • 配置
  • 案例使用

    使用资源

    Langchain4j官方文档

    LangChain4j社区DashScope

    Langchain4jMaven仓库

    阿里通义千问API

    记录

    本人使用版本： 1.0.0-beta2

    不使用Spring Boot ，仅集成

    额外：如果不使用SpringBoot，仅集成LangChain4j - DashScope

    如果使用的langChain4j-dashScope版本小于等于1.0.0-alpha1

        dev.langchain4j
        langchain4j-dashscope
        ${previous version here}
    
    

    如果使用的langChain4j-dashScope版本大于等于1.0.0-alpha1

        dev.langchain4j
        langchain4j-community-dashscope
        ${langchain4j.version}
    
    

    集成Spring Boot 启动器

    Spring Boot 启动器有助于创建和配置语言模型、嵌入模型、嵌入存储、 和其他核心 LangChain4j 组件。如果要使用Spring Boot 相关启动器， 需要在pom.xml中引入相应依赖项

    pom.xml ：

    如果使用的langChain4j-dashScope版本小于等于1.0.0-alpha1

        dev.langchain4j
        langchain4j-community-dashscope-spring-boot-starter
        ${langchain4j.version}
    
    

    如果使用的langChain4j-dashScope版本大于等于1.0.0-alpha1

        dev.langchain4j
        langchain4j-community-dashscope-spring-boot-starter
        ${latest version here}
    
    

    或者使用 BOM 管理依赖项：

        
            dev.langchain4j
            langchain4j-community-bom
            ${latest version here}
            pom
            import
        
    
    

    配置

    引入依赖后，在propertity中配置注册

    langchain4j.community.dashscope.api-key=
    langchain4j.community.dashscope.model-name==
    #是否输出日志
    langchain4j.community.dashscope.log-requests=true
    langchain4j.community.dashscope.log-responses=true
    #日志等级
    logging.level.dev.langchain4j=DEBUG
    

    其他相关参数也可添加，例如 QwenChatModel参数

    langchain4j.community.dashscope.temperature=0.7
    langchain4j.community.dashscope.max-tokens=4096
    

    官方参数如下：

    
    （图片来源网络，侵删）

    langchain4j-community-dashscope 有四个model可使用:

    QwenChatModel

    
    （图片来源网络，侵删）

    QwenStreamingChatModel

    QwenLanguageModel

    
    （图片来源网络，侵删）

    QwenStreamingLanguageModel

    QwenChatModel 参数如下，其他也相同

    Property	Description	Default Value
    baseUrl	The URL to connect to. You can use HTTP or websocket to connect to DashScope	Text Inference and Multi-Modal
    apiKey	The API Key
    modelName	The model to use.	qwen-plus
    topP	The probability threshold for kernel sampling controls the diversity of texts generated by the model. the higher the , the more diverse the generated texts, and vice versa. Value range: (0, 1.0]. We generally recommend altering this or temperature but not both.top_p
    topK	The size of the sampled candidate set during the generation process.
    enableSearch	Whether the model uses Internet search results for reference when generating text or not.
    seed	Setting the seed parameter will make the text generation process more deterministic, and is typically used to make the results consistent.
    repetitionPenalty	Repetition in a continuous sequence during model generation. Increasing reduces the repetition in model generation, 1.0 means no penalty. Value range: (0, +inf)repetition_penalty
    temperature	Sampling temperature that controls the diversity of the text generated by the model. the higher the temperature, the more diverse the generated text, and vice versa. Value range: [0, 2)
    stops	With the stop parameter, the model will automatically stop generating text when it is about to contain the specified string or token_id.
    maxTokens	The maximum number of tokens returned by this request.
    listeners	Listeners that listen for request, response and errors.

    案例使用

    例如 QwenChatModel

    ChatLanguageModel qwenModel = QwenChatModel.builder()
                        .apiKey("You API key here")
                        .modelName("qwen-max")
                        .build();