Skip to content

NevarokML: UNevarokMLSpace

The UNevarokMLSpace class represents a space in the NevarokML plugin. It provides various methods for creating, modifying, and inspecting different types of spaces used in machine learning tasks.

Overview

The UNevarokMLSpace class is a base class for different types of spaces, including discrete spaces, multi-discrete spaces, multi-binary spaces, continuous spaces, and box spaces. Spaces define the possible states or actions that an agent can take in a reinforcement learning environment.

The NevarokML plugin provides several factory methods to create different types of spaces. These methods include:

  • Discrete: Creates a discrete space with a specified size.
  • MultiDiscrete: Creates a multi-discrete space with an array specifying the number of actions for each discrete action space.
  • MultiBinary: Creates a multi-binary space with a specified size.
  • Continuous: Creates a continuous space with a specified size and bounds.
  • Box: Creates a box space with a specified shape and bounds.
  • MultiDiscreteStack: Creates a stack of multi-discrete spaces.
  • MultiBinaryStack: Creates a stack of multi-binary spaces.
  • ContinuousStack: Creates a stack of continuous spaces.
  • BoxDXStack: Creates a stack of box spaces along the x-axis.
  • BoxDYStack: Creates a stack of box spaces along the y-axis.

The UNevarokMLSpace class provides methods to convert the space to different types, such as converting to discrete, multi-discrete, multi-binary, continuous, and box spaces. These conversion methods allow you to modify the properties of the space and update it in-place.

You can also access various properties of the space, such as its shape, stack dimension, size, bounds, and more. The class provides methods to check the validity of the space and retrieve its type.

API Reference

For detailed information on the API of the UNevarokMLSpace class, refer to the following documentation pages:

Conclusion

The UNevarokMLSpace class provides a powerful set of tools for creating and manipulating different types of spaces used in machine learning tasks. With the NevarokML plugin, you can easily define and customize spaces to suit your specific requirements.