MLeap ML 模型导出 MLeap ML Model Export

MLeap 采用序列化格式,是机器学习管道的执行引擎。MLeap is a serialization format and execution engine for machine learning pipelines. 可在 MLeap GitHub 存储库中找到该包。You can find the package at the MLeap GitHub repo. 它是一个开源包,使用 Apache 2.0 许可证It is an open source package and employs the Apache 2.0 license. 有关许可的详细信息,请参阅 MLeap 许可证For licensing details, see the MLeap license.

备注

  • MLeap 适用于 Spark 版本 2.0、2.1、2.2、2.3 和2.4。MLeap works with Spark versions 2.0, 2.1, 2.2, 2.3, and 2.4.
  • 本指南并不是关于 MLeap 的全面指南。This guide is not a comprehensive guide on MLeap. 请参阅 MLeap 文档Refer to the MLeap documentation.

备注

MLeap 包含在用于机器学习的 Databricks Runtime 中,后者是一种为机器学习和数据科学提供现成环境的 Databricks Runtime 运行时。MLeap is included in Databricks Runtime for Machine Learning, a Databricks Runtime runtime that provides a ready-to-go environment for machine learning and data science. 可直接使用 Databricks Runtime ML 来创建群集,而不必按照以下说明安装 MLeap。Instead of installing MLeap using the following instructions, you can simply create a cluster using Databricks Runtime ML. 请参阅用于机器学习的 Databricks RuntimeSee Databricks Runtime for Machine Learning.

下面的笔记本提供了相关指导来讲解如何在 Azure Databricks 上安装 MLeap,还提供了模型导出工作流的示例。The following notebooks provide guidance on installing MLeap on Azure Databricks and give an example of a Model Export workflow.

在 Scala 中导出和导入模型Export and import models in Scala

MLeap 导出 Scala 笔记本MLeap export Scala notebook

获取笔记本Get notebook

在 Python 中导出和导入模型Export and import models in Python

MLeap 导出 Python 笔记本MLeap export Python notebook

获取笔记本Get notebook