package documentation
The rlgt.environments package defines the GraphEnvironment abstract base class, which
formalizes a reinforcement learning environment for graph theory applications. It also provides
seven concrete environment classes that inherit from GraphEnvironment and implement specific
graph building dynamics. In addition, the package defines the EpisodeStatus enumeration, which
describes the execution status of a batch of episodes, and includes several utility functions for
constructing graph generator functions used to initialize batches of underlying graphs.
| Module | global |
This Python module defines two global reinforcement learning environments that inherit from the GraphEnvironment class. These environments model graph building games in which the edges (resp. arcs) are initially fully colored in some predetermined manner, and at each step, any edge (resp... |
| Module | graph |
This Python module defines the GraphEnvironment abstract class, which encapsulates the concept of a reinforcement learning (RL) environment for graph theory applications, together with an associated enumeration describing the possible episode statuses. |
| Module | graph |
This Python module defines several factory functions for constructing graph generator functions. These generators implement various mechanisms for producing batches of fully colored k-edge-colored looped complete graphs of a specified batch size. |
| Module | linear |
This Python module defines three linear reinforcement learning environments that inherit from the GraphEnvironment class and model graph building games where the edges (resp. arcs) are all initially either uncolored, or fully colored in some predetermined manner, and are then properly (re)colored one by one, either in the row-major order or the clockwise order. |
| Module | local |
This Python module defines two local reinforcement learning environments that inherit from the GraphEnvironment class and model graph building games in which the edges (resp. arcs) are initially fully colored in some predetermined manner... |