DocArray内存向量存储¶

DocArrayInMemoryVectorStore是由Docarray提供的文档索引，它将文档存储在内存中。对于小型数据集来说，这是一个很好的起点，您可能不想启动一个数据库服务器。

如果您在Colab上打开这个笔记本，您可能需要安装LlamaIndex 🦙。

In [ ]:

%pip install llama-index-vector-stores-docarray

%pip install llama-index-vector-stores-docarray

In [ ]:

!pip install llama-index

!pip install llama-index

In [ ]:

import osimport sysimport loggingimport textwrapimport warningswarnings.filterwarnings("ignore")# 停止huggingface警告os.environ["TOKENIZERS_PARALLELISM"] = "false"# 取消注释以查看调试日志# logging.basicConfig(stream=sys.stdout, level=logging.INFO)# logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))from llama_index.core import (    GPTVectorStoreIndex,    SimpleDirectoryReader,    Document,)from llama_index.vector_stores.docarray import DocArrayInMemoryVectorStorefrom IPython.display import Markdown, display

import osimport sysimport loggingimport textwrapimport warningswarnings.filterwarnings("ignore")# 停止huggingface警告os.environ["TOKENIZERS_PARALLELISM"] = "false"# 取消注释以查看调试日志# logging.basicConfig(stream=sys.stdout, level=logging.INFO)# logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))from llama_index.core import ( GPTVectorStoreIndex, SimpleDirectoryReader, Document,)from llama_index.vector_stores.docarray import DocArrayInMemoryVectorStorefrom IPython.display import Markdown, display

In [ ]:

import os

os.environ["OPENAI_API_KEY"] = "<your openai key>"

import os os.environ["OPENAI_API_KEY"] = ""

下载数据

In [ ]:

!mkdir -p 'data/paul_graham/'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay.txt'

!mkdir -p 'data/paul_graham/' !wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay.txt'

In [ ]:

# 加载文档documents = SimpleDirectoryReader("./data/paul_graham/").load_data()print(    "文档ID:",    documents[0].doc_id,    "文档哈希值:",    documents[0].doc_hash,)

# 加载文档documents = SimpleDirectoryReader("./data/paul_graham/").load_data()print( "文档ID:", documents[0].doc_id, "文档哈希值:", documents[0].doc_hash,)

Document ID: 1c21062a-50a3-4133-a0b1-75f837a953e5 Document Hash: 77ae91ab542f3abb308c4d7c77c9bc4c9ad0ccd63144802b7cbe7e1bb3a4094e

初始化和索引化¶

In [ ]:

from llama_index.core import StorageContext


vector_store = DocArrayInMemoryVectorStore()
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = GPTVectorStoreIndex.from_documents(
    documents, storage_context=storage_context
)

from llama_index.core import StorageContext vector_store = DocArrayInMemoryVectorStore() storage_context = StorageContext.from_defaults(vector_store=vector_store) index = GPTVectorStoreIndex.from_documents( documents, storage_context=storage_context )

查询¶

In [ ]:

# 将日志级别设置为DEBUG，以获得更详细的输出query_engine = index.as_query_engine()response = query_engine.query("作者在成长过程中做了什么？")print(textwrap.fill(str(response), 100))

# 将日志级别设置为DEBUG，以获得更详细的输出query_engine = index.as_query_engine()response = query_engine.query("作者在成长过程中做了什么？")print(textwrap.fill(str(response), 100))

Token indices sequence length is longer than the specified maximum sequence length for this model (1830 > 1024). Running this sequence through the model will result in indexing errors

 Growing up, the author wrote short stories, programmed on an IBM 1401, and nagged his father to buy
him a TRS-80 microcomputer. He wrote simple games, a program to predict how high his model rockets
would fly, and a word processor. He also studied philosophy in college, but switched to AI after
becoming bored with it. He then took art classes at Harvard and applied to art schools, eventually
attending RISD.

In [ ]:

response = query_engine.query("What was a hard moment for the author?")
print(textwrap.fill(str(response), 100))

response = query_engine.query("What was a hard moment for the author?") print(textwrap.fill(str(response), 100))

 A hard moment for the author was when he realized that the AI programs of the time were a hoax and
that there was an unbridgeable gap between what they could do and actually understanding natural
language. He had invested a lot of time and energy into learning about AI and was disappointed to
find out that it was not going to get him the results he had hoped for.

使用过滤器进行查询¶

In [ ]:

from llama_index.core.schema import TextNode

nodes = [
    TextNode(
        text="The Shawshank Redemption",
        metadata={
            "author": "Stephen King",
            "theme": "Friendship",
        },
    ),
    TextNode(
        text="The Godfather",
        metadata={
            "director": "Francis Ford Coppola",
            "theme": "Mafia",
        },
    ),
    TextNode(
        text="Inception",
        metadata={
            "director": "Christopher Nolan",
        },
    ),
]

from llama_index.core.schema import TextNode nodes = [ TextNode( text="The Shawshank Redemption", metadata={ "author": "Stephen King", "theme": "Friendship", }, ), TextNode( text="The Godfather", metadata={ "director": "Francis Ford Coppola", "theme": "Mafia", }, ), TextNode( text="Inception", metadata={ "director": "Christopher Nolan", }, ), ]

In [ ]:

from llama_index.core import StorageContext


vector_store = DocArrayInMemoryVectorStore()
storage_context = StorageContext.from_defaults(vector_store=vector_store)

index = GPTVectorStoreIndex(nodes, storage_context=storage_context)

from llama_index.core import StorageContext vector_store = DocArrayInMemoryVectorStore() storage_context = StorageContext.from_defaults(vector_store=vector_store) index = GPTVectorStoreIndex(nodes, storage_context=storage_context)

In [ ]:

from llama_index.core.vector_stores import ExactMatchFilter, MetadataFilters


filters = MetadataFilters(
    filters=[ExactMatchFilter(key="theme", value="Mafia")]
)

retriever = index.as_retriever(filters=filters)
retriever.retrieve("What is inception about?")

from llama_index.core.vector_stores import ExactMatchFilter, MetadataFilters filters = MetadataFilters( filters=[ExactMatchFilter(key="theme", value="Mafia")] ) retriever = index.as_retriever(filters=filters) retriever.retrieve("What is inception about?")

Out[ ]:

[NodeWithScore(node=Node(text='director: Francis Ford Coppola\ntheme: Mafia\n\nThe Godfather', doc_id='41c99963-b200-4ce6-a9c4-d06ffeabdbc5', embedding=None, doc_hash='b770e43e6a94854a22dc01421d3d9ef6a94931c2b8dbbadf4fdb6eb6fbe41010', extra_info=None, node_info=None, relationships={<DocumentRelationship.SOURCE: '1'>: 'None'}), score=0.7681788983417586)]