Get Started#

The following information is designed to get users up and running with skbase quickly. For more detailed information, see the links in each of the subsections.

Installation#

skbase currently supports:

  • environments with python version 3.7, 3.8, 3.9 or 3.10

  • operating systems Mac OS X, Unix-like OS, Windows 8.1 and higher

  • installation via PyPi

To install skbase with its core dependencies via pip use:

pip install skbase

To install skbase with maximum dependencies, including soft dependencies, install with the all_extras modifier:

pip install skbase[all_extras]

For additional details see our full installation guide.

Key Concepts#

skbase seeks to provide a general framework for creating and working with classes that follow scikit-learn, sktime style design patterns.

Primary functionality is provided through base classes that provide interfaces for:

  • scikit-learn style parameter getting and setting

  • using tags to record characteristics of the class that can be used to alter the classes code or how it interacts with other functionality

  • generating test instances

To make it easy to build toolboxes and applications that use skbase’s, interfaces are also provided for:

  • retrieving information on BaseObject-s

  • testing BaseObject-s

Warning

The skbase package is new and its interfaces are still experimental.

Quickstart#

The code snippets below are designed to introduce skbase’s functionality. For more detailed information see the Tutorials, User Guide and API Reference in skbase’s Documentation.