ApsaraDB RDS:Use the EAS module of PAI and ApsaraDB RDS for PostgreSQL to deploy a RAG-based LLM chatbot

Background information

Prerequisites

• A virtual private cloud (VPC), vSwitch, and security group are created. For more information, see Create and manage a VPC and Create a security group.

• An Object Storage Service (OSS) bucket or File Storage NAS (NAS) file system is created to store fine-tuned model files. This prerequisite must be met if you use a fine-tuned model to deploy the chatbot. For more information, see Get started by using the OSS console or Create a file system.

  Note

  If you use Faiss to build a vector database, you must prepare an OSS bucket.

Limits

The ApsaraDB RDS for PostgreSQL instance and EAS must reside in the same region.

Usage notes

This practice is subject to the maximum number of tokens of an LLM service and is designed to help you understand the basic retrieval feature of a RAG-based LLM chatbot.

• The chatbot is limited by the server resource size of the LLM service and the default number of tokens. The conversation length supported by the chatbot is also limited.

• If you do not need to perform multiple rounds of conversations, we recommended that you disable the with chat history feature of the chatbot. This effectively reduces the possibility of reaching the limit. For more information, see the How do I disable the with chat history feature of the RAG-based chatbot? section of this topic.

Step 1: Build a vector database in an RDS instance

RAG allows you to build a vector database in an RDS instance. When you build a vector database, you must save the required parameter configurations that are used to connect to the vector database in subsequent operations.

1. Create an RDS instance. For more information, see Create an instance.

   We recommend that the RDS instance resides in the same region as the RAG-based LLM chatbot to enable communications over a VPC.

2. Create an account and a database for the RDS instance. For more information, see Create a database and an account.

   • When you create an account, select Privileged Account for Account Type.

   • When you create a database, select the created privileged account from the Authorized By drop-down list.

3. Configure the database connection.

   1. Go to the Instances page. In the top navigation bar, select the region in which the RDS instance resides. Then, find the RDS instance and click the ID of the instance.

   2. In the left-side navigation pane, click Database Connection.

   3. On the Database Connection page, view the endpoints and port of the database.

4. Add pg_jieba to Running Parameter Value of the shared_preload_libraries parameter. For example, change Running Parameter Value to 'pg_stat_statements,auto_explain,pg_cron'. For more information, see Modify the parameters of an ApsaraDB RDS for PostgreSQL instance.

   Note

   ApsaraDB RDS for PostgreSQL uses the pg_jieba extension to segment Chinese text. You can use this extension for keyword-based retrieval and recall. For more information, see Use the pg_jieba extension.

Step 2: Deploy the RAG-based chatbot

1. Go to the EAS page.

   1. Log on to the Platform for AI (PAI) console.

   2. In the left-side navigation pane, click Workspaces. On the Workspaces page, find the workspace to which you want to deploy the model and click its name to go to the Workspace Details page.

      If no workspaces are available, you must create one. For more information, see Create a workspace.

   3. In the left-side navigation pane, choose Model Deployment > Elastic Algorithm Service (EAS) to go to the Elastic Algorithm Service (EAS) page.

2. On the Elastic Algorithm Service (EAS) page, click Deploy Service. In the Scenario-based Model Deployment section, click RAG-based Smart Dialogue Deployment.

3. On the RAG-based LLM Chatbot Deployment page, configure the parameters. The following tables describe the key parameters in different sections.

   • Basic Information

     Parameter	Description
     Service Name	The name of the service.
     Model Source	The source of the model. Valid values: Open Source Model and Custom Fine-tuned Model.
     Model Type	The model type. You can select a model type based on your business requirements. If you set Model Source to Custom Fine-tuned Model, you must configure the parameter quantity and precision for the model type.
     Model Settings	If you set Model Source to Custom Fine-tuned Model, you must configure the path in which the fine-tuned model file is stored. Valid values: Note Make sure that the model file format is compatible with Hugging Face transformers. Mount OSS: Select the OSS path in which the fine-tuned model file is stored. Mount NAS: Select the NAS file system in which the fine-tuned model file is stored and the source path of the NAS file system.

   • Resource Configuration

     Parameter	Description
     Resource Configuration	If you set Model Source to Open Source Model, the system automatically selects an instance type based on the selected model type as the default value. If you set Model Source to Custom Fine-tuned Model, you need to select an instance type that matches the model. For more information, see Deploy LLM applications in EAS.
     Inference Acceleration	Inference acceleration can be enabled for the Qwen, Llama2, ChatGLM, or Baichuan2 model that is deployed on A10 or GU30 instances. The following inference acceleration methods are provided: BladeLLM Inference Acceleration: The BladeLLM inference acceleration engine ensures high concurrency and low latency. You can use BladeLLM to accelerate LLM inference in a cost-effective manner. Open-source vLLM Inference Acceleration

   • Vector Database Settings

     Parameter	Description
     Vector Database Type	The engine type of the vector database. Select RDS PostgreSQL.
     Host Address	The internal endpoint or public endpoint of the RDS instance. Internal endpoint: If the RAG application and the database reside in the same region, you can use the internal endpoint to connect the RAG application to your RDS instance. Public endpoint: If the RAG application and the database reside in the different regions, you must apply for a public endpoint for your RDS instance. For more information, see Apply for or release a public endpoint.
     Port	The port number. Default value: 5432.
     Database	The name of the database that you create in the RDS instance.
     Table Name	The name of a new or an existing table. If you enter the name of an existing table, the table schema must meet the requirements of the RAG-based LLM chatbot. For example, you can enter the name of the table that is automatically created when you deploy the RAG-based LLM chatbot by using EAS.
     Account	The privileged account of the RDS instance
     Password	The password of the privileged account.

   • VPC Configuration

     Parameter	Description
     VPC	If the host address is an internal endpoint, you must select the VPC of the RDS instance for the RAG-based LLM chatbot. If the host address is a public endpoint, you must configure a VPC and vSwitch for the RAG-based LLM chatbot and create a NAT gateway and an elastic IP address (EIP) for the VPC. This enables the RAG-based LLM chatbot to be accessible over the Internet. You must add the EIP to the IP address whitelist of the RDS instance. For more information, see Use the SNAT feature of an Internet NAT gateway to access the Internet and Configure an IP address whitelist.
     vSwitch
     Security Group Name	The security group. Important Do not use the security group named created_by_rds because this security group is used only for system access control.

4. Click Deploy.

   If the value in the Service Status column changes to Running, the RAG-based chatbot is deployed.

Step 3: Perform model inference on the web UI

This section describes how to debug the RAG-based chatbot on the web UI. After you test the Q&A performance of the RAG-based chatbot on the web UI, you can call API operations provided by Platform for AI (PAI) to apply the RAG-based chatbot to your business system. For more information, see Step 4: Call API operations to perform model inference in this topic.

1. Configure the RAG-based chatbot

1. After you deploy the RAG-based chatbot, click View Web App in the Service Type column to enter the web UI.

2. Configure the machine learning model.

   • Embedding Model Name: Four models are available. By default, the optimal model is selected.

   • Embedding Dimension: After you configure the Embedding Model Name parameter, the system automatically configures this parameter.

3. Check whether the vector database is connected.

   The system automatically recognizes and applies the vector database settings that are configured when you deploy the chatbot. The settings cannot be modified. Click Connect PostgreSQL to check whether the vector database on the RDS instance is connected.

2. Upload specified business data files

3. Configure model inference parameters

On the Chat tab, configure Q&A policies.

4. Perform model inference

Step 4: Call API operations to perform model inference

1. Obtain the invocation information of the RAG-based chatbot.

   1. Click the name of the RAG-based chatbot to go to the Service Details page.

   2. In the Basic Information section, click View Endpoint Information.

   3. On the Public Endpoint tab of the Invocation Method dialogue box, obtain the service endpoint and token.

2. Connect to the vector database through the WebUI and upload knowledge base files.

   You can also upload knowledge base to the vector base based on the structure of a generated table, which conforms to the PAI-RAG format.

3. Call the service through APIs.

   PAI allows you to call the RAG-based chatbot by using the following API operations in different query modes: service/query/retrieval in retrieval mode, service/query/llm in LLM mode, and service/query in RAG mode. Sample code:

   cURL command

   • Initiate a single-round conversation request

     • Method 1: Call the service/query/retrieval operation.

       curl -X 'POST'  '<service_url>service/query/retrieval' -H 'Authorization: <service_token>' -H 'accept: application/json' -H 'Content-Type: application/json'  -d '{"question": "What is PAI?"}'
       # Replace <service_url> and <service_token> with the service endpoint and service token that you obtained in Step 1.

     • Method 2: Call the /service/query/llm operation.

       curl -X 'POST'  '<service_url>service/query/llm' -H 'Authorization: <service_token>' -H 'accept: application/json'  -H 'Content-Type: application/json'  -d '{"question": "What is PAI?"}'
       # Replace <service_url> and <service_token> with the service endpoint and service token that you obtained in Step 1.

       You can add other adjustable inference parameters such as {"question":"What is PAI?", "temperature": 0.9}.

     • Method 3: Call the service/query operation.

       curl -X 'POST'  '<service_url>service/query' -H 'Authorization: <service_token>' -H 'accept: application/json'  -H 'Content-Type: application/json'  -d '{"question": "What is PAI?"}'
       # Replace <service_url> and <service_token> with the service endpoint and service token that you obtained in Step 1.

       You can add other adjustable inference parameters such as {"question":"What is PAI?", "temperature": 0.9}.

   • Initiate a multi-round conversational search request

     You can initiate a multi-round conversational search request only in RAG and LLM query modes. The following sample code shows an example on how to initiate a multi-round conversational search request in RAG query mode:

     # Send the request.  
     curl -X 'POST'  '<service_url>service/query' -H 'Authorization: <service_token>' -H 'accept: application/json'  -H 'Content-Type: application/json'  -d '{"question": "What is PAI?"}'
     
     # Provide the session ID returned for the request. This ID uniquely identifies a conversation in the conversation history. After the session ID is provided, the corresponding conversation is stored and is automatically included in subsequent requests to call an LLM. 
     curl -X 'POST'  '<service_url>service/query' -H 'Authorization: <service_token>' -H 'accept: application/json'  -H 'Content-Type: application/json'  -d '{"question": "What are the benefits of PAI?","session_id": "ed7a80e2e20442eab****"}'
     
     # Provide the chat_history parameter, which contains the conversation history between you and the chatbot. The parameter value is a list in which each element indicates a single round of conversation in the {"user":"Inputs","bot":"Outputs"} format. Multiple conversations are sorted in chronological order. 
     curl -X 'POST'  '<service_url>service/query' -H 'Authorization: <service_token>' -H 'accept: application/json'  -H 'Content-Type: application/json'  -d '{"question":"What are the features of PAI?", "chat_history": [{"user":"What is PAI", "bot":"PAI is an AI platform provided by Alibaba Cloud..."}]}'
     
     # If you provide both the session_id and chat_history parameters, the conversation history is appended to the conversation that corresponds to the specified session ID.  
     curl -X 'POST'  '<service_url>service/query' -H 'Authorization: <service_token>' -H 'accept: application/json'  -H 'Content-Type: application/json'  -d '{"question":"What are the features of PAI?", "chat_history": [{"user":"What is PAI", "bot":"PAI is an AI platform provided by Alibaba Cloud..."}], "session_id": "1702ffxxad3xxx6fxxx97daf7c"}'

   Python

   • The following sample code shows an example on how to initiate a single-round conversational search request:

     import requests
     
     EAS_URL = 'http://xxxx.****.cn-beijing.pai-eas.aliyuncs.com'
     headers = {
         'accept': 'application/json',
         'Content-Type': 'application/json',
         'Authorization': 'MDA5NmJkNzkyMGM1Zj****YzM4M2YwMDUzZTdiZmI5YzljYjZmNA==',
     }
     
     
     def test_post_api_query_llm():
         url = EAS_URL + '/service/query/llm'
         data = {
            "question":"What is PAI?"
         }
         response = requests.post(url, headers=headers, json=data)
     
         if response.status_code != 200:
             raise ValueError(f'Error post to {url}, code: {response.status_code}')
         ans = dict(response.json())
         print(f"======= Question =======\n {data['question']}")
         print(f"======= Answer =======\n {ans['answer']} \n\n")
     
     
     def test_post_api_query_retrieval():
         url = EAS_URL + '/service/query/retrieval'
         data = {
            "question":"What is PAI?"
         }
         response = requests.post(url, headers=headers, json=data)
     
         if response.status_code != 200:
             raise ValueError(f'Error post to {url}, code: {response.status_code}')
         ans = dict(response.json())
         print(f"======= Question =======\n {data['question']}")
         print(f"======= Answer =======\n {ans['docs']}\n\n")
     
     
     def test_post_api_query_rag():
         url = EAS_URL + '/service/query'
         data = {
            "question":"What is PAI?"
         }
         response = requests.post(url, headers=headers, json=data)
     
         if response.status_code != 200:
             raise ValueError(f'Error post to {url}, code: {response.status_code}')
         ans = dict(response.json())
         print(f"======= Question =======\n {data['question']}")
         print(f"======= Answer =======\n {ans['answer']}")
         print(f"======= Retrieved Docs =======\n {ans['docs']}\n\n")
     # LLM
     test_post_api_query_llm()
     # Retrieval
     test_post_api_query_retrieval()
     # RAG (Retrieval + LLM)
     test_post_api_query_rag()
     

     Set the EAS_URL parameter to the endpoint of the RAG-based chatbot. Make sure to remove the forward slash (/) at the end of the endpoint. Set the Authorization parameter to the token of the RAG-based chatbot.

   • Initiate a multi-round conversational search request

     You can initiate a multi-round conversational search request only in RAG (Retrieval + LLM) and LLM query modes. Sample code:

     import requests
     
     EAS_URL = 'http://xxxx.****.cn-beijing.pai-eas.aliyuncs.com'
     headers = {
         'accept': 'application/json',
         'Content-Type': 'application/json',
         'Authorization': 'MDA5NmJkN****jNlMDgzYzM4M2YwMDUzZTdiZmI5YzljYjZmNA==',
     }
     
     
     def test_post_api_query_llm_with_chat_history():
         url = EAS_URL + '/service/query/llm'
         # Round 1 query
         data = {
            "question":"What is PAI?"
         }
         response = requests.post(url, headers=headers, json=data)
     
         if response.status_code != 200:
             raise ValueError(f'Error post to {url}, code: {response.status_code}')
         ans = dict(response.json())
         print(f"=======Round 1: Question =======\n {data['question']}")
         print(f"=======Round 1: Answer =======\n {ans['answer']} session_id: {ans['session_id']} \n")
        
         # Round 2 query
         data_2 = {
            "question": "What are the benefits of PAI?",
            "session_id": ans['session_id']
         }
         response_2 = requests.post(url, headers=headers, json=data_2)
     
         if response.status_code != 200:
             raise ValueError(f'Error post to {url}, code: {response.status_code}')
         ans_2 = dict(response_2.json())
         print(f"=======Round 2: Question =======\n {data_2['question']}")
         print(f"=======Round 2: Answer =======\n {ans_2['answer']} session_id: {ans_2['session_id']} \n\n")
     
     
     def test_post_api_query_rag_with_chat_history():
         url = EAS_URL + '/service/query'
        
         # Round 1 query
         data = {
            "question":"What is PAI?"
         }
         response = requests.post(url, headers=headers, json=data)
     
         if response.status_code != 200:
             raise ValueError(f'Error post to {url}, code: {response.status_code}')
         ans = dict(response.json())
     
         print(f"=======Round 1: Question =======\n {data['question']}")
         print(f"=======Round 1: Answer =======\n {ans['answer']} session_id: {ans['session_id']}")
         print(f"=======Round 1: Retrieved Docs =======\n {ans['docs']}\n")
     
         # Round 2 query
         data = {
            "question":"What are the features of PAI?",
            "session_id": ans['session_id']
         }
         response = requests.post(url, headers=headers, json=data)
     
         if response.status_code != 200:
             raise ValueError(f'Error post to {url}, code: {response.status_code}')
         ans = dict(response.json())
     
         print(f"=======Round 2: Question =======\n {data['question']}")
         print(f"=======Round 2: Answer =======\n {ans['answer']} session_id: {ans['session_id']}")
         print(f"=======Round 2: Retrieved Docs =======\n {ans['docs']}")
     # LLM
     test_post_api_query_llm_with_chat_history()
     # RAG (Retrieval + LLM)
     test_post_api_query_rag_with_chat_history()
     

     Set the EAS_URL parameter to the endpoint of the RAG-based chatbot. Make sure to remove the forward slash (/) at the end of the endpoint. Set the Authorization parameter to the token of the RAG-based chatbot.

Other features

After you connect to a database that is used as a vector database on the RDS instance, you can view the imported knowledge base content. For more information about how to connect to a database, see Connect to an ApsaraDB RDS for PostgreSQL instance.

References

You can also use EAS to deploy the following items:

• You can deploy an LLM application that can be called by using the web UI or API operations. After the LLM application is deployed, use the LangChain framework to integrate enterprise knowledge bases into the LLM application and implement intelligent Q&A and automation features. For more information, see Quickly deploy LLMs in EAS.

• You can deploy an AI video generation model service by using ComfyUI and Stable Video Diffusion models. This helps you complete tasks such as short video generation and animation on social media platforms. For more information, see Use ComfyUI to deploy an AI video generation model service.

FAQ

How do I disable the with chat history feature of the RAG-based chatbot?

On the web UI page of the RAG-based chatbot, do not select Chat history.