适用于 Snowflake ML 的 Python APIs¶

The snowflake-ml-python Python package provides Python APIs that connect to the various Snowflake ML workflow components and also includes APIs for building and training your own models. You can use these APIs in your favorite Python IDE on your own workstation, in Snowsight worksheets, or in Snowflake notebooks.

在 Snowflake 笔记本中使用 Snowflake ML¶

Snowflake Notebooks provide an easy-to-use notebook interface for your data work, blending Python, SQL, and Markdown. To use Snowflake ML features in notebooks, choose the Anaconda package snowflake-ml-python using the Packages menu at the top of the notebook.

笔记本同时支持 CPU 和 GPU 运行时选项。许多类型的模型都需要或受益于可用的 GPU。

在 Snowsight 工作表中使用 Snowflake ML¶

Snowsight Worksheets provide a powerful and versatile method for running Python code. To use Snowflake ML features in worksheets, choose the Anaconda package snowflake-ml-python using the Packages menu at the top of the worksheet.

在本地使用 Snowflake ML¶

You must install the snowflake-ml-python package to develop on your own workstation or elsewhere outside Snowflake. All Snowflake ML features are available in a single package, snowflake-ml-python. You can install the package from the Snowflake conda channel using the conda command or from the Python Package Index (PyPI) using pip. Conda is preferred.

• 从 Snowflake Conda 通道安装
• 从 PyPI 安装

从 Snowflake Conda 通道安装¶

1. 创建将安装 Snowflake ML 的 Conda 环境。如果您希望使用现有环境，请跳过此步骤。

   Copy codeExpand code block

   conda create --name snowflake-ml

   Show lessSee more

   Scroll to top

2. 激活 Conda 环境：

   Copy codeExpand code block

   conda activate snowflake-ml

   Show lessSee more

   Scroll to top

3. Install snowflake-ml-python from the Snowflake conda channel

   Copy codeExpand code block

   conda install --override-channels --channel https://repo.anaconda.com/pkgs/snowflake/ snowflake-ml-python

   Show lessSee more

   Scroll to top

从 PyPI 安装¶

You can install snowflake-ml-python from the Python Package Index (PyPI) by using the standard Python package manager, pip.

1. 创建并激活 Python 虚拟环境：

   Copy codeExpand code block

   python3 -m virtualenv venv
   source venv/bin/activate

   Show lessSee more

   Scroll to top

2. Install the snowflake-ml-python package:

   Copy codeExpand code block

   python -m pip install snowflake-ml-python

   Show lessSee more

   Scroll to top

安装可选依赖项

Some APIs require dependencies that are not installed as dependencies of snowflake-ml-python. By default, scikit-learn is installed. Other packages such as lightgbm, xgboost, keras, pytorch, and others are optional dependencies.

If you plan to use the snowflake.ml.modeling.lightgbm module, install lightgbm. Use the following commands to activate your conda environment and install lightgbm from the Snowflake conda channel.

conda activate snowflake-ml
conda install --override-channels --channel https://repo.anaconda.com/pkgs/snowflake/ lightgbm

Use the following commands to activate your virtual environment and install lightgbm using pip.

source venv/bin/activate
python -m pip install 'snowflake-ml-python[lightgbm]'

设置 Snowpark Python¶

Snowpark Python is a dependency of snowflake-ml-python and is installed automatically with it. If Snowpark Python is not already set up on your system, you might need to perform additional configuration steps. See Setting up your development environment for Snowpark Python for Snowpark Python setup instructions.

连接到 Snowflake¶

Before using Snowflake ML features in Python, connect to Snowflake using a Snowpark Session object. Use the SnowflakeLoginOptions function in the snowflake.ml.utils.connection_params module to get the configuration settings to create the session. The function can read the connection settings from a named connection in your SnowSQL configuration file or from environment variables that you set. It returns a dictionary containing these parameters, which can be used to create a connection.

The following examples read the connection parameters from the named connection myaccount in the SnowSQL configuration file. To create a Snowpark Python session, create a builder for the Session class, and pass the connection information to the builder’s configs method:

from snowflake.snowpark import Session
from snowflake.ml.utils import connection_params

params = connection_params.SnowflakeLoginOptions("myaccount")
sp_session = Session.builder.configs(params).create()

现在，您可以将会话传递给任何需要它的内容。

session = Session.builder.configs({"connection": connection}).create()

指定仓库

Many Snowflake ML features, for example model training or inference, run code in a Snowflake warehouse. These operations run in the warehouse specified by the session you use to connect. For example, if you create a session from a named connection in your SnowSQL configuration file, you can specify a warehouse using the warehousename parameter in the named configuration.

You can add the warehouse setting when creating the Session object, as shown here, if it does not already exist in the configuration.

from snowflake.snowpark import Session
from snowflake.ml.utils import connection_params

# Get named connection from SnowSQL configuration file
params = connection_params.SnowflakeLoginOptions(connection_name="my_connection")

# Add warehouse name for model method calls if it's not already present
if "warehouse" not in params:
    params["warehouse"] = "mlwarehouse"
sp_session = Session.builder.configs(params).create()

If no warehouse is specified in the session, or if you want to use a different warehouse, call the session’s use_warehouse method to specify a warehouse.

sp_session.use_warehouse("mlwarehouse")

API 参考¶

The Snowflake ML API reference includes documentation on all publicly-released functionality. You can also obtain detailed API documentation for any API by using Python’s help function in an interactive Python session. For example:

from snowflake.ml.modeling.preprocessing import OneHotEncoder

help(OneHotEncoder)