在本 notebook 中，我们将快速演示如何使用 MyScaleVectorStore。

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-myscale

!pip install llama-index

创建 MyScale 客户端

import logging
import sys

logging.basicConfig(stream=sys.stdout, level=logging.INFO)
logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))

from os import environ
import clickhouse_connect

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

# initialize client
client = clickhouse_connect.get_client(
    host="YOUR_CLUSTER_HOST",
    port=8443,
    username="YOUR_USERNAME",
    password="YOUR_CLUSTER_PASSWORD",
)

加载文档，使用 MyScaleVectorStore 构建并存储 VectorStoreIndex

在这里，我们将使用一组 Paul Graham 的文章作为文本，将其转换为嵌入，存储到 MyScaleVectorStore 中，并进行查询，以为我们的 LLM 问答循环查找上下文。

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
from llama_index.vector_stores.myscale import MyScaleVectorStore
from IPython.display import Markdown, display

# load documents
documents = SimpleDirectoryReader("../data/paul_graham").load_data()
print("Document ID:", documents[0].doc_id)
print("Number of Documents: ", len(documents))

Document ID: a5f2737c-ed18-4e5d-ab9a-75955edb816d
Number of Documents:  1

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'

你可以使用 SimpleDirectoryReader 单独处理文件：

loader = SimpleDirectoryReader("./data/paul_graham/")
documents = loader.load_data()
for file in loader.input_files:
    print(file)
    # Here is where you would do any preprocessing

../data/paul_graham/paul_graham_essay.txt

# initialize with metadata filter and store indexes
from llama_index.core import StorageContext

for document in documents:
    document.metadata = {"user_id": "123", "favorite_color": "blue"}
vector_store = MyScaleVectorStore(myscale_client=client)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = VectorStoreIndex.from_documents(
    documents, storage_context=storage_context
)

查询索引

现在 MyScale 向量存储支持过滤搜索和混合搜索

你可以详细了解查询引擎和检索器。

import textwrap

from llama_index.core.vector_stores import ExactMatchFilter, MetadataFilters

# set Logging to DEBUG for more detailed outputs
query_engine = index.as_query_engine(
    filters=MetadataFilters(
        filters=[
            ExactMatchFilter(key="user_id", value="123"),
        ]
    ),
    similarity_top_k=2,
    vector_store_query_mode="hybrid",
)
response = query_engine.query("What did the author learn?")
print(textwrap.fill(str(response), 100))

清除所有索引

for document in documents:
    index.delete_ref_doc(document.doc_id)