This release introduces the evaluate_player function, two baseline players (MaxBaseDamagePlayer and SimpleHeuristicsPlayer) and several helper methods (notably Player.damage_multiplier and Player.battle_against).

This release adds the ladder method to Player objects, which allow them to search and play games directly on a showdown server's ladder.

Battle objects also now include rating and opponent_rating attributes.

This release introduces default options for Player instances:

• By default, the player configuration is automatically populated with a unique username designed to work on authentication-less showdown servers
• By default, the server configuration points to the default localhost server
• By default, the Player plays gen8randombattles
• The documentation is updated accordingly

This update introduces minor performance optimizations (15% to 30% reduced runtime on RandomPlayer battles) and new Effects.

Version 0.3.0 includes support for gen 8 single mechanics, including dynamax and gigantamax. It also fixes some minor errors.

This release introduces two main novelties:

• Support for most gen 7 single formats
• The Teambuilder module, which opens the door for team-building related AI projects. For now, it just incorporates basic utilities (eg., constant team).

Additionally, it also includes a couple of small bug fixes, minor utilities and data / enum updates.

NB: the actual code correspond to the following commit, which updated the internal package version

This release introduce minor changes meant to ensure compatibility with the latest version of pokemon-showdown.

This release adds a missing effect, LIMBER.

This release adds EnvPlayer, a class exposing the Open Ai Gym Environment API to control agents.