node和rasa交互之可通信

合集 - rasa(5)

1.rasa笔记-rasa core等2024-12-092.rasa 最简单对话实现2024-12-12

3.node和rasa交互之可通信2024-12-12

4.rasa 实现简易的多轮对话2024-12-185.rasa nlu 传递信息及 REST API 替代 rasa nlu2024-12-29

收起

1.2 通过node和rasa交互

1.2.1 配置

更改 config.yml

# The config recipe.
# https://rasa.com/docs/rasa/model-configuration/
recipe: default.v1

# The assistant project unique identifier
# This default value must be replaced with a unique assistant name within your deployment
assistant_id: 20241120-170106-khaki-margarine

# Configuration for Rasa NLU.
# https://rasa.com/docs/rasa/nlu/components/
language: zh # 改为中文

pipeline: 
  - name: JiebaTokenizer  # 使用支持中文的 Tokenizer
  - name: CountVectorsFeaturizer
    analyzer: "word"  # 基于分词结果生成特征
    min_ngram: 1
    max_ngram: 2
  - name: DIETClassifier
    epochs: 100
  - name: EntitySynonymMapper
#  - name: ResponseSelector
#    epochs: 100
# # No configuration for the NLU pipeline was provided. The following default pipeline was used to train your model.
# # If you'd like to customize it, uncomment and adjust the pipeline.
# # See https://rasa.com/docs/rasa/tuning-your-model for more information.

#   - name: RegexFeaturizer
#   - name: LexicalSyntacticFeaturizer

#   - name: CountVectorsFeaturizer
#     analyzer: char_wb
#     min_ngram: 1
#     max_ngram: 4

#     constrain_similarities: true

#     constrain_similarities: true
#   - name: FallbackClassifier
#     threshold: 0.3
#     ambiguity_threshold: 0.1

# Configuration for Rasa Core.
# https://rasa.com/docs/rasa/core/policies/
# policies: 
#  - name: TEDPolicy
#    max_history: 5
#    epochs: 100
#    constrain_similarities: true

# # No configuration for policies was provided. The following default policies were used to train your model.
# # If you'd like to customize them, uncomment and adjust the policies.
# # See https://rasa.com/docs/rasa/policies for more information.
#   - name: MemoizationPolicy
#   - name: RulePolicy
#   - name: UnexpecTEDIntentPolicy
#     max_history: 5
#     epochs: 100
policies:
# # No configuration for policies was provided. The following default policies were used to train your model.
# # If you'd like to customize them, uncomment and adjust the policies.
# # See https://rasa.com/docs/rasa/policies for more information.
#   - name: MemoizationPolicy
#   - name: RulePolicy
#   - name: UnexpecTEDIntentPolicy
#     max_history: 5
#     epochs: 100
#   - name: TEDPolicy
#     max_history: 5
#     epochs: 100
#     constrain_similarities: true

注意：别忘了先进虚拟环境，因为 jieba-0.42.1 的兼容性问题，我还把 setuptools 降级了

pip install setuptools==67.6.1

1.2.2 训练模型

然后重新训练模型

rasa train

1.2.3 启动 rasa 服务器

rasa run -m models --enable-api --cors "*" --debug

1.2.4 apipost 测试及可用路由

启动成功后即可使用接口访问

可用的 Web 服务器路由有以下这些：

/conversations/<conversation_id:path>/messages       POST                           add_message
/conversations/<conversation_id:path>/tracker/events POST                           append_events
/webhooks/rasa                                       GET                            custom_webhook_RasaChatInput.health
/webhooks/rasa/webhook                               POST                           custom_webhook_RasaChatInput.receive
/webhooks/rest                                       GET                            custom_webhook_RestInput.health
/webhooks/rest/webhook                               POST                           custom_webhook_RestInput.receive
/model/test/intents                                  POST                           evaluate_intents
/model/test/stories                                  POST                           evaluate_stories
/conversations/<conversation_id:path>/execute        POST                           execute_action
/domain                                              GET                            get_domain
/                                                    GET                            hello
/model                                               PUT                            load_model
/model/parse                                         POST                           parse
/conversations/<conversation_id:path>/predict        POST                           predict
/conversations/<conversation_id:path>/tracker/events PUT                            replace_events
/conversations/<conversation_id:path>/story          GET                            retrieve_story
/conversations/<conversation_id:path>/tracker        GET                            retrieve_tracker
/status                                              GET                            status
/model/predict                                       POST                           tracker_predict
/model/train                                         POST                           train
/conversations/<conversation_id:path>/trigger_intent POST                           trigger_intent
/model                                               DELETE                         unload_model
/version                                             GET                            version

具体来说：

1./conversations/<conversation_id:path>/messages

• 方法：POST

• 功能：向指定的对话会话发送一条消息。

• 用途：用户与机器人交互的主要端点，用于发送用户消息并触发 Rasa 的响应。

• 参数：

  • conversation_id：对话会话的唯一标识符。
  • 请求体中需要提供用户消息。

• 示例：

  {
    "sender": "user_1",
    "message": "你好"
  }
  

2./conversations/<conversation_id:path>/tracker/events

• 方法：POST

• 功能：手动向会话的 tracker（追踪器）中添加事件。

• 用途：用于模拟事件流，或在测试和调试时添加自定义事件。

• 示例：

  {
    "event": "user",
    "text": "Hello",
    "timestamp": 1625146462.123
  }
  

3./webhooks/rasa 和 /webhooks/rasa/webhook

• 方法：

  • GET（/webhooks/rasa）：用于检查 Rasa webhook 的健康状态。
  • POST（/webhooks/rasa/webhook）：用于通过 Rasa 自定义输入通道发送消息。

• 用途：这些端点与自定义的输入通道集成，例如，来自外部平台的消息。

• 示例：

  {
    "sender": "test_user",
    "message": "Hello Rasa"
  }
  

4./webhooks/rest 和 /webhooks/rest/webhook

• 方法：

  • GET（/webhooks/rest）：检查 REST 输入通道的健康状态。
  • POST（/webhooks/rest/webhook）：通过 REST API 与 Rasa 交互，用于标准化的 REST 输入。

• 用途：适用于通过 REST 与 Rasa 进行消息传递的场景。

• 示例：

  {
    "sender": "test_user",
    "message": "你好"
  }
  

1.2.5 Node.js与rasa响应的交互

创建一个 index.js

node.js 代码如下：

需要注意的是，这里使用的是 127.0.0.1，笔者试过 localhost 会失败，127.0.0.1 可以强制使用 ipv4

const express = require('express');
const axios = require('axios');
const bodyParser = require('body-parser');

const app = express();
const port = 6000;

// 中间件
app.use(bodyParser.json());

// 定义与 Rasa 交互的端点
app.post('/webhook', async (req, res) => {
    const userMessage = req.body.message;
    const senderId = req.body.sender || 'default';

    if (!userMessage) {
        return res.status(400).json({ error: "Message is required" });
    }

    try {
        // 发送用户消息到 Rasa
        const response = await axios.post('http://127.0.0.1:5005/webhooks/rest/webhook', {
            sender: senderId,
            message: userMessage
        });

        // 收集 Rasa 的响应
        const rasaResponses = response.data.map(r => r.text).filter(text => text);

        // 返回响应到客户端
        res.json({ replies: rasaResponses });
    } catch (error) {
        console.error('Error communicating with Rasa:', error.message);

        // 检查 Axios 错误并处理
        if (error.response) {
            // Rasa 服务返回了非 2xx 状态
            res.status(error.response.status).json({ error: error.response.data });
        } else if (error.request) {
            // 没有收到 Rasa 的响应
            res.status(500).json({ error: 'No response from Rasa server' });
        } else {
            // 请求配置错误
            res.status(500).json({ error: 'Request error: ' + error.message });
        }
    }
});

// 启动服务器
app.listen(port, () => {
    console.log(`Node.js server is running on http://localhost:${port}`);
});

good，good

END

本文主要实现node和rasa的交互，这样为后面node介入rasa奠定基础，目前实现的是node获取rasa响应后的结果，从接口访问rasa然后返回rasa的结果。涉及如何配置，训练模型，启动rasa服务器，打通路由，交互。