谷歌 Colab

Google Colab 是一个托管的 Jupyter Notebook 服务。它提供免费的计算资源，包括 GPU 和 TPU，非常适合机器学习、数据科学和教育。我们可以使用 Colab 来管理使用 Supabase Vecs 的集合。

🌐 Google Colab is a hosted Jupyter Notebook service. It provides free access to computing resources, including GPUs and TPUs, and is well-suited to machine learning, data science, and education. We can use Colab to manage collections using Supabase Vecs.

在本教程中，我们将连接到运行在 Supabase 平台上的数据库。如果你还没有数据库，可以在这里创建一个：database.new。

🌐 In this tutorial we'll connect to a database running on the Supabase platform. If you don't already have a database, you can create one here: database.new.

创建一个新注意本 #

🌐 Create a new notebook

首先访问 colab.research.google.com。在那里你可以创建一个新的注意本。

🌐 Start by visiting colab.research.google.com. There you can create a new notebook.

安装 Vecs #

🌐 Install Vecs

我们将使用 Supabase Vector 客户端 Vecs 来管理我们的集合。

🌐 We'll use the Supabase Vector client, Vecs, to manage our collections.

在注意本顶部，粘贴以下代码并点击“执行”（ctrl+enter）:

🌐 At the top of the notebook, paste the following code and click "Execute" (ctrl+enter):

1
pip install vecs

连接到你的数据库 #

🌐 Connect to your database

在你的项目仪表板上，点击 Connect。连接字符串应该看起来像 postgres://postgres.xxxx:password@xxxx.pooler.supabase.com:6543/postgres

🌐 On your project dashboard, click Connect. The connection string should look like postgres://postgres.xxxx:password@xxxx.pooler.supabase.com:6543/postgres

在安装模块（ctrl+m b）下面创建一个新的代码块，并使用你上面复制的 Postgres URI 添加以下代码：

🌐 Create a new code block below the install block (ctrl+m b) and add the following code using the Postgres URI you copied above:

1
import vecs
2
3
DB_CONNECTION = "postgres://postgres.xxxx:password@xxxx.pooler.supabase.com:6543/postgres"
4
5
# create vector store client
6
vx = vecs.create_client(DB_CONNECTION)

执行代码块（ctrl+enter）。如果没有返回错误，那说明你的连接成功了。

🌐 Execute the code block (ctrl+enter). If no errors were returned then your connection was successful.

创建一个收藏 #

🌐 Create a collection

现在我们要创建一个新集合并插入一些文档。

🌐 Now we're going to create a new collection and insert some documents.

在安装块（ctrl+m b）下面创建一个新的代码块。将以下代码添加到代码块中并执行它（ctrl+enter）:

🌐 Create a new code block below the install block (ctrl+m b). Add the following code to the code block and execute it (ctrl+enter):

1
collection = vx.get_or_create_collection(name="colab_collection", dimension=3)
2
3
collection.upsert(
4
    vectors=[
5
        (
6
         "vec0",           # the vector's identifier
7
         [0.1, 0.2, 0.3],  # the vector. list or np.array
8
         {"year": 1973}    # associated  metadata
9
        ),
10
        (
11
         "vec1",
12
         [0.7, 0.8, 0.9],
13
         {"year": 2012}
14
        )
15
    ]
16
)

这将在你的数据库的 vecs 架构中创建一个名为 colab_collection 的表。你可以在 表格编辑器 中查看插入的项目，只需从架构下拉菜单中选择 vecs 架构即可。

🌐 This will create a table inside your database within the vecs schema, called colab_collection. You can view the inserted items in the Table Editor, by selecting the vecs schema from the schema dropdown.

查询你的文档 #

🌐 Query your documents

现在我们可以根据文档的相似性来搜索文档。创建一个新的代码块并执行以下代码：

🌐 Now we can search for documents based on their similarity. Create a new code block and execute the following code:

1
collection.query(
2
    query_vector=[0.4,0.5,0.6],  # required
3
    limit=5,                     # number of records to return
4
    filters={},                  # metadata filters
5
    measure="cosine_distance",   # distance measure to use
6
    include_value=False,         # should distance measure values be returned?
7
    include_metadata=False,      # should record metadata be returned?
8
)

你会看到这会返回一个包含两个文档的数组 ['vec1', 'vec0']:

🌐 You will see that this returns two documents in an array ['vec1', 'vec0']:

它还会返回一个警告：

🌐 It also returns a warning:

1
Query does not have a covering index for cosine_distance.

你可以在 Vecs 文档 中了解更多关于创建索引的内容。

🌐 You can lean more about creating indexes in the Vecs documentation.

资源 #

🌐 Resources

• Vecs API：supabase.github.io/vecs/api