We’re excited to announce that the Oracle Machine Learning for Python (OML4Py) 2.1 client is now available for Windows users. This enables data scientists and developers to build, train, and deploy machine learning models directly from their Windows laptops and workstations, eliminating a need for a local Linux client or virtual machine.

OML4Py brings Oracle’s in-database machine learning algorithms to Python, allowing data scientists to build and score models in their preferred language while maintaining scalability and performance directly where the data resides. In addition, OML4Py supports data transformations and statistical, machine learning, and graphical analysis on data stored in or accessible through the database. You can run user-defined Python functions through database-spawned and controlled Python engines, with optional built-in data-parallelism and task-parallelism. This embedded execution functionality enables calling user-defined functions from SQL, and on Autonomous AI Database, from REST. OML4Py also supports Automated Machine Learning (AutoML) for algorithm and feature selection, and model tuning and selection. You can further augment the built-in functionality with third-party packages from the Python ecosystem. For more details about OML4Py, refer to Oracle Machine Learning for Python documentation. For more details about OML4Py, refer to Oracle Machine Learning for Python documentation.

Earlier this year, we announced the OML4Py 2.1 release. The OML4Py 2.1 release added support for the VECTOR datatype, enabling data scientists to work with vectors for AI applications. With Windows client support, these capabilities are now accessible to a broader audience of users.

Supported Platforms

OML4Py 2.1 for Windows is certified for:

  • Windows Server 2022 x64 Standard Edition
  • Windows 11 x64 Pro Edition
  • Oracle AI Database 26ai
  • Oracle Autonomous AI Database 26ai

Installing the OML4Py Client

To get started with the OML4Py Windows client, follow these installation steps:

  1. Install the Oracle Instant Client
  2. Download and configure the Autonomous AI Database Wallet (for Autonomous AI Database only)
  3. Install Python 3.13.2
  4. Install the required supporting packages
  5. Install OML4Py client package

For detailed installation instructions and download links, refer to the OML4Py client installation instructions.

Platform specific notes

The Windows client supports core OML4Py functionality with the following exceptions:

For workflows requiring these features, use a Linux client instead.

Resources

For more information and to get started, refer to the following resources.