milvus-logo
LFAI
Home
  • Integraciones

Retrieval-Augmented Generation (RAG) con Milvus y LangChain

Open In Colab GitHub Repository

Esta guía muestra cómo construir un sistema de Generación Aumentada por Recuperación (RAG) utilizando LangChain y Milvus.

El sistema RAG combina un sistema de recuperación con un modelo generativo para generar nuevo texto basado en una petición dada. En primer lugar, el sistema recupera documentos relevantes de un corpus utilizando Milvus y, a continuación, utiliza un modelo generativo para generar un nuevo texto basado en los documentos recuperados.

LangChain es un marco para el desarrollo de aplicaciones basadas en grandes modelos lingüísticos (LLM). Milvus es la base de datos vectorial de código abierto más avanzada del mundo, creada para potenciar la búsqueda de similitudes de incrustación y las aplicaciones de IA.

Requisitos previos

Antes de ejecutar este cuaderno, asegúrate de tener instaladas las siguientes dependencias:

$ pip install --upgrade --quiet  langchain langchain-core langchain-community langchain-text-splitters langchain-milvus langchain-openai bs4

Si utilizas Google Colab, para habilitar las dependencias que acabas de instalar, es posible que tengas que reiniciar el tiempo de ejecución. (Haz clic en el menú "Tiempo de ejecución" en la parte superior de la pantalla y selecciona "Reiniciar sesión" en el menú desplegable).

Utilizaremos los modelos de OpenAI. Debes preparar la clave api OPENAI_API_KEY como variable de entorno.

import os

os.environ["OPENAI_API_KEY"] = "sk-***********"

Preparar los datos

Usamos el Langchain WebBaseLoader para cargar documentos desde fuentes web y dividirlos en trozos usando el RecursiveCharacterTextSplitter.

import bs4
from langchain_community.document_loaders import WebBaseLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter

# Create a WebBaseLoader instance to load documents from web sources
loader = WebBaseLoader(
    web_paths=(
        "https://lilianweng.github.io/posts/2023-06-23-agent/",
        "https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/",
    ),
    bs_kwargs=dict(
        parse_only=bs4.SoupStrainer(
            class_=("post-content", "post-title", "post-header")
        )
    ),
)
# Load documents from web sources using the loader
documents = loader.load()
# Initialize a RecursiveCharacterTextSplitter for splitting text into chunks
text_splitter = RecursiveCharacterTextSplitter(chunk_size=2000, chunk_overlap=200)

# Split the documents into chunks using the text_splitter
docs = text_splitter.split_documents(documents)

# Let's take a look at the first document
docs[1]
Document(page_content='Fig. 1. Overview of a LLM-powered autonomous agent system.\nComponent One: Planning#\nA complicated task usually involves many steps. An agent needs to know what they are and plan ahead.\nTask Decomposition#\nChain of thought (CoT; Wei et al. 2022) has become a standard prompting technique for enhancing model performance on complex tasks. The model is instructed to “think step by step” to utilize more test-time computation to decompose hard tasks into smaller and simpler steps. CoT transforms big tasks into multiple manageable tasks and shed lights into an interpretation of the model’s thinking process.\nTree of Thoughts (Yao et al. 2023) extends CoT by exploring multiple reasoning possibilities at each step. It first decomposes the problem into multiple thought steps and generates multiple thoughts per step, creating a tree structure. The search process can be BFS (breadth-first search) or DFS (depth-first search) with each state evaluated by a classifier (via a prompt) or majority vote.\nTask decomposition can be done (1) by LLM with simple prompting like "Steps for XYZ.\\n1.", "What are the subgoals for achieving XYZ?", (2) by using task-specific instructions; e.g. "Write a story outline." for writing a novel, or (3) with human inputs.\nAnother quite distinct approach, LLM+P (Liu et al. 2023), involves relying on an external classical planner to do long-horizon planning. This approach utilizes the Planning Domain Definition Language (PDDL) as an intermediate interface to describe the planning problem. In this process, LLM (1) translates the problem into “Problem PDDL”, then (2) requests a classical planner to generate a PDDL plan based on an existing “Domain PDDL”, and finally (3) translates the PDDL plan back into natural language. Essentially, the planning step is outsourced to an external tool, assuming the availability of domain-specific PDDL and a suitable planner which is common in certain robotic setups but not in many other domains.\nSelf-Reflection#', metadata={'source': 'https://lilianweng.github.io/posts/2023-06-23-agent/'})

Como podemos ver, el documento ya está dividido en trozos. Y el contenido de los datos es sobre el agente de IA.

Construir la cadena RAG con Milvus Vector Store

Vamos a inicializar un Milvus vector store con los documentos, que cargar los documentos en el Milvus vector store y construir un índice bajo el capó.

from langchain_milvus import Milvus, Zilliz
from langchain_openai import OpenAIEmbeddings

embeddings = OpenAIEmbeddings()

vectorstore = Milvus.from_documents(  # or Zilliz.from_documents
    documents=docs,
    embedding=embeddings,
    connection_args={
        "uri": "./milvus_demo.db",
    },
    drop_old=True,  # Drop the old Milvus collection if it exists
)

Para el connection_args:

  • Establecer el uri como un archivo local, por ejemplo./milvus.db, es el método más conveniente, ya que utiliza automáticamente Milvus Lite para almacenar todos los datos en este archivo.
  • Si tiene una gran escala de datos, puede configurar un servidor Milvus más eficiente en docker o kubernetes. En esta configuración, por favor utilice la uri del servidor, por ejemplohttp://localhost:19530, como su uri.
  • Si desea utilizar Zilliz Cloud, el servicio en la nube totalmente gestionado para Milvus, sustituya Milvus.from_documents por Zilliz.from_documents, y ajuste uri y token, que corresponden al punto final público y a la clave Api en Zilliz Cloud.

Busque los documentos en el almacén vectorial de Milvus utilizando una pregunta de consulta de prueba. Echemos un vistazo al primer documento.

query = "What is self-reflection of an AI Agent?"
vectorstore.similarity_search(query, k=1)
[Document(page_content='Self-Reflection#\nSelf-reflection is a vital aspect that allows autonomous agents to improve iteratively by refining past action decisions and correcting previous mistakes. It plays a crucial role in real-world tasks where trial and error are inevitable.\nReAct (Yao et al. 2023) integrates reasoning and acting within LLM by extending the action space to be a combination of task-specific discrete actions and the language space. The former enables LLM to interact with the environment (e.g. use Wikipedia search API), while the latter prompting LLM to generate reasoning traces in natural language.\nThe ReAct prompt template incorporates explicit steps for LLM to think, roughly formatted as:\nThought: ...\nAction: ...\nObservation: ...\n... (Repeated many times)', metadata={'source': 'https://lilianweng.github.io/posts/2023-06-23-agent/', 'pk': 449281835035555859})]
from langchain_core.runnables import RunnablePassthrough
from langchain_core.prompts import PromptTemplate
from langchain_core.output_parsers import StrOutputParser
from langchain_openai import ChatOpenAI

# Initialize the OpenAI language model for response generation
llm = ChatOpenAI(model_name="gpt-3.5-turbo", temperature=0)

# Define the prompt template for generating AI responses
PROMPT_TEMPLATE = """
Human: You are an AI assistant, and provides answers to questions by using fact based and statistical information when possible.
Use the following pieces of information to provide a concise answer to the question enclosed in <question> tags.
If you don't know the answer, just say that you don't know, don't try to make up an answer.
<context>
{context}
</context>

<question>
{question}
</question>

The response should be specific and use statistics or numbers when possible.

Assistant:"""

# Create a PromptTemplate instance with the defined template and input variables
prompt = PromptTemplate(
    template=PROMPT_TEMPLATE, input_variables=["context", "question"]
)
# Convert the vector store to a retriever
retriever = vectorstore.as_retriever()


# Define a function to format the retrieved documents
def format_docs(docs):
    return "\n\n".join(doc.page_content for doc in docs)

Utiliza el LCEL(LangChain Expression Language) para construir una cadena RAG.

# Define the RAG (Retrieval-Augmented Generation) chain for AI response generation
rag_chain = (
    {"context": retriever | format_docs, "question": RunnablePassthrough()}
    | prompt
    | llm
    | StrOutputParser()
)

# rag_chain.get_graph().print_ascii()

# Invoke the RAG chain with a specific question and retrieve the response
res = rag_chain.invoke(query)
res
"Self-reflection of an AI agent involves the process of synthesizing memories into higher-level inferences over time to guide the agent's future behavior. It serves as a mechanism to create higher-level summaries of past events. One approach to self-reflection involves prompting the language model with the 100 most recent observations and asking it to generate the 3 most salient high-level questions based on those observations. This process helps the AI agent optimize believability in the current moment and over time."

¡Enhorabuena! Ha construido una cadena RAG básica con Milvus y LangChain.

Filtrado de metadatos

Podemos utilizar las reglas de filtrado escalar de Milvus para filtrar los documentos basándonos en los metadatos. Hemos cargado los documentos de dos fuentes diferentes, y podemos filtrar los documentos por los metadatos source.

vectorstore.similarity_search(
    "What is CoT?",
    k=1,
    expr="source == 'https://lilianweng.github.io/posts/2023-06-23-agent/'",
)

# The same as:
# vectorstore.as_retriever(search_kwargs=dict(
#     k=1,
#     expr="source == 'https://lilianweng.github.io/posts/2023-06-23-agent/'",
# )).invoke("What is CoT?")
[Document(page_content='Fig. 1. Overview of a LLM-powered autonomous agent system.\nComponent One: Planning#\nA complicated task usually involves many steps. An agent needs to know what they are and plan ahead.\nTask Decomposition#\nChain of thought (CoT; Wei et al. 2022) has become a standard prompting technique for enhancing model performance on complex tasks. The model is instructed to “think step by step” to utilize more test-time computation to decompose hard tasks into smaller and simpler steps. CoT transforms big tasks into multiple manageable tasks and shed lights into an interpretation of the model’s thinking process.\nTree of Thoughts (Yao et al. 2023) extends CoT by exploring multiple reasoning possibilities at each step. It first decomposes the problem into multiple thought steps and generates multiple thoughts per step, creating a tree structure. The search process can be BFS (breadth-first search) or DFS (depth-first search) with each state evaluated by a classifier (via a prompt) or majority vote.\nTask decomposition can be done (1) by LLM with simple prompting like "Steps for XYZ.\\n1.", "What are the subgoals for achieving XYZ?", (2) by using task-specific instructions; e.g. "Write a story outline." for writing a novel, or (3) with human inputs.\nAnother quite distinct approach, LLM+P (Liu et al. 2023), involves relying on an external classical planner to do long-horizon planning. This approach utilizes the Planning Domain Definition Language (PDDL) as an intermediate interface to describe the planning problem. In this process, LLM (1) translates the problem into “Problem PDDL”, then (2) requests a classical planner to generate a PDDL plan based on an existing “Domain PDDL”, and finally (3) translates the PDDL plan back into natural language. Essentially, the planning step is outsourced to an external tool, assuming the availability of domain-specific PDDL and a suitable planner which is common in certain robotic setups but not in many other domains.\nSelf-Reflection#', metadata={'source': 'https://lilianweng.github.io/posts/2023-06-23-agent/', 'pk': 449281835035555858})]

Si queremos cambiar dinámicamente los parámetros de búsqueda sin reconstruir la cadena, podemos configurar los internos de la cadena en tiempo de ejecución. Definamos un nuevo recuperador con esta configuración dinámica y utilicémoslo para construir una nueva cadena RAG.

from langchain_core.runnables import ConfigurableField

# Define a new retriever with a configurable field for search_kwargs
retriever2 = vectorstore.as_retriever().configurable_fields(
    search_kwargs=ConfigurableField(
        id="retriever_search_kwargs",
    )
)

# Invoke the retriever with a specific search_kwargs which filter the documents by source
retriever2.with_config(
    configurable={
        "retriever_search_kwargs": dict(
            expr="source == 'https://lilianweng.github.io/posts/2023-06-23-agent/'",
            k=1,
        )
    }
).invoke(query)
[Document(page_content='Self-Reflection#\nSelf-reflection is a vital aspect that allows autonomous agents to improve iteratively by refining past action decisions and correcting previous mistakes. It plays a crucial role in real-world tasks where trial and error are inevitable.\nReAct (Yao et al. 2023) integrates reasoning and acting within LLM by extending the action space to be a combination of task-specific discrete actions and the language space. The former enables LLM to interact with the environment (e.g. use Wikipedia search API), while the latter prompting LLM to generate reasoning traces in natural language.\nThe ReAct prompt template incorporates explicit steps for LLM to think, roughly formatted as:\nThought: ...\nAction: ...\nObservation: ...\n... (Repeated many times)', metadata={'source': 'https://lilianweng.github.io/posts/2023-06-23-agent/', 'pk': 449281835035555859})]
# Define a new RAG chain with this dynamically configurable retriever
rag_chain2 = (
    {"context": retriever2 | format_docs, "question": RunnablePassthrough()}
    | prompt
    | llm
    | StrOutputParser()
)

Probemos esta cadena RAG configurable dinámicamente con diferentes condiciones de filtro.

# Invoke this RAG chain with a specific question and config
rag_chain2.with_config(
    configurable={
        "retriever_search_kwargs": dict(
            expr="source == 'https://lilianweng.github.io/posts/2023-06-23-agent/'",
        )
    }
).invoke(query)
"Self-reflection of an AI agent involves the process of synthesizing memories into higher-level inferences over time to guide the agent's future behavior. It serves as a mechanism to create higher-level summaries of past events. One approach to self-reflection involves prompting the language model with the 100 most recent observations and asking it to generate the 3 most salient high-level questions based on those observations. This process helps the AI agent optimize believability in the current moment and over time."

Cuando cambiamos la condición de búsqueda para filtrar los documentos por la segunda fuente, como el contenido de esta fuente de blog no tiene nada que ver con la pregunta de la consulta, obtenemos una respuesta sin información relevante.

rag_chain2.with_config(
    configurable={
        "retriever_search_kwargs": dict(
            expr="source == 'https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/'",
        )
    }
).invoke(query)
"I'm sorry, but based on the provided context, there is no specific information or statistical data available regarding the self-reflection of an AI agent."

Este tutorial se centra en el uso básico de la integración de Milvus LangChain y el enfoque simple de RAG. Para técnicas más avanzadas de RAG, por favor consulte el curso avanzado de RAG.

Traducido porDeepLogo

Try Managed Milvus for Free

Zilliz Cloud is hassle-free, powered by Milvus and 10x faster.

Get Started
Feedback

¿Fue útil esta página?