The getML engine

The getML engine is a standalone program written in C++ that does the actual work of feature engineering and prediction.

Starting the engine

The engine can be started using the dedicated launcher icon or by using the getML command line interface (CLI). For more information check out the installation instructions for your operating system.

Shutting down the engine

There are several ways to shut down the getML engine:

  • Click the ‘icon_shutdown Shutdown’ tab in the sidebar of the monitor

  • Press Ctrl-C (if started via the command line)

  • Run the getML command-line interface (CLI) (see Installation) using the -stop option

  • macOS only: Right-click the getML icon icon_getml in the status bar and click ‘Quit’ (if started via the launchpad)

Logging

The engine keeps a log about what it is currently doing.

The easiest way to view the log is to click the ‘icon_log Log’ tab in the sidebar of the getML monitor. The engine will also output its log to the command line when it is started using the command-line interface.

Managing data using projects

All data sets you use and all models you fit belong to a project.

Every project has its own folder in ~/.getML/getml-VERSION/projects (for Linux and macOS) in which all of its data and models are stored. On Windows, the projects folder is in the same location as getML.exe. These folders can be easily shared between different instances of getML; even between different operating systems. However, individual models or data frames cannot be simply copied to another project folder - they are tied to the project.

To load an existing project or create a new one, you can use the ‘icon_projects Projects’ view in the monitor or API getml.engine.set_project().

When you change the project, the memory of the engine is flushed and all unsaved changes to the data frames are lost (see Lifecycles and synchronization between engine and API for details). All models of the new project are automatically loaded into memory.

Projects can be deleted by clicking the trash can icon in the ‘icon_projects Projects’ tab of the getML monitor or by calling delete_project().