Release to keep compatibility for current users before change in environment dynamics endpoint.
Done and reward functions
The environment delivers on the first implementation of reward and done functions, which are executed for each agent. The documentation provides examples. We recommend not using this functionality and stick to environment dynamics, as those endpoints will be deprecated in the future.
Environment dynamics
The environment delivers on the first implementation of environment dynamic classes, which are executed for each agent. The documentation provides examples. In the next release, those need to return done and info as well.
First Wiki Documentation
Release v0.0.1 focuses on enhancing user experience through improved documentation. Clearer explanations and examples have been added to help users understand the functionality of the EnvironmentDynamic and Language classes. Explore the documentation for a smoother onboarding experience.
Bug Fixes and Compatibility
Addressing user feedback, release v0.0.1 includes bug fixes and improved compatibility. The dynamic method in both classes now handles scenarios with a different number of agents, ensuring a more robust and adaptable environment for users. Update to benefit from a more stable simulation experience.