在本 notebook 中，我们将展示如何使用 Vecs 在 LlamaIndex 中执行向量搜索。
有关在 Supabase 上托管数据库的说明，请参阅本指南

如果你在 Colab 中打开这个 Notebook，可能需要先安装 LlamaIndex 🦙。

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

!pip install llama-index

import logging
import sys

# Uncomment to see debug logs
# logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
# logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))

from llama_index.core import SimpleDirectoryReader, Document, StorageContext
from llama_index.core import VectorStoreIndex
from llama_index.vector_stores.supabase import SupabaseVectorStore
import textwrap

设置 OpenAI

第一步是配置 OpenAI 密钥。它将用于为加载到索引中的文档生成嵌入

import os

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

下载数据

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

加载文档

使用 SimpleDirectoryReader 加载存储在 ./data/paul_graham/ 中的文档

documents = SimpleDirectoryReader("./data/paul_graham/").load_data()
print(
    "Document ID:",
    documents[0].doc_id,
    "Document Hash:",
    documents[0].doc_hash,
)

Document ID: fb056993-ee9e-4463-80b4-32cf9509d1d8 Document Hash: 77ae91ab542f3abb308c4d7c77c9bc4c9ad0ccd63144802b7cbe7e1bb3a4094e

创建由 Supabase 向量存储支持的索引。

这适用于所有支持 pgvector 的 Postgres 提供程序。如果集合不存在，我们将尝试创建一个新集合

注意：如果不使用 OpenAI 的 text-embedding-ada-002，则需要传入嵌入维度，例如 vector_store = SupabaseVectorStore(..., dimension=...)

vector_store = SupabaseVectorStore(
    postgres_connection_string=(
        "postgresql://<user>:<password>@<host>:<port>/<db_name>"
    ),
    collection_name="base_demo",
)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = VectorStoreIndex.from_documents(
    documents, storage_context=storage_context
)

查询索引

现在我们可以使用索引来提问了。

query_engine = index.as_query_engine()
response = query_engine.query("Who is the author?")

/Users/suo/miniconda3/envs/llama/lib/python3.9/site-packages/vecs/collection.py:182: UserWarning: Query does not have a covering index for cosine_distance. See Collection.create_index
  warnings.warn(

print(textwrap.fill(str(response), 100))

 The author of this text is Paul Graham.

response = query_engine.query("What did the author do growing up?")

print(textwrap.fill(str(response), 100))

 The author grew up writing essays, learning Italian, exploring Florence, painting people, working
with computers, attending RISD, living in a rent-stabilized apartment, building an online store
builder, editing Lisp expressions, publishing essays online, writing essays, painting still life,
working on spam filters, cooking for groups, and buying a building in Cambridge.

使用元数据过滤器

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",
            },
        }
    ),
]

vector_store = SupabaseVectorStore(
    postgres_connection_string=(
        "postgresql://<user>:<password>@<host>:<port>/<db_name>"
    ),
    collection_name="metadata_filters_demo",
)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = VectorStoreIndex(nodes, storage_context=storage_context)

定义元数据过滤器

from llama_index.core.vector_stores import ExactMatchFilter, MetadataFilters

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

使用过滤器从向量存储中检索

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

[NodeWithScore(node=Node(text='The Godfather', doc_id='f837ed85-aacb-4552-b88a-7c114a5be15d', embedding=None, doc_hash='f8ee912e238a39fe2e620fb232fa27ade1e7f7c819b6d5b9cb26f3dddc75b6c0', extra_info={'theme': 'Mafia', 'director': 'Francis Ford Coppola'}, node_info={'_node_type': '1'}, relationships={}), score=0.20671339734643313)]