ronaldosvieira / gym-locm

OpenAI Gym environments for Legends of Code and Magic, a collectible card game designed for AI research

ronaldosvieira/gym-locm Issues

`LOCMEnv` sometimes returns `None` instead of a numerical state representation
Updated 8 months ago1
Include action number in the `Action` class for Offline RL use cases
Updated 10 months ago
[Question] Obtaining numerical state for an agent
Updated 10 months ago2
NativeAgent class doesn't work with agents that use stderr
Closed a year ago
Make runner support LOCM 1.5
Closed a year ago
Create and automatize consistency checks
Closed a year ago7
Implement saving models as JSON
Closed a year ago
Use stable-baseline3 instead of stable-baseline for training draft agents
Closed a year ago
Remove player_hp, enemy_hp, card_draw, and area from creatures on the board
Closed a year ago1
Create battle envs for LOCM 1.5
Updated 2 years ago
Add support for Gymnasium
Updated 2 years ago
Update the READMEs
Updated 2 years ago
Add support for LOCM 1.5
Updated 2 years ago2
Make runner seed agents
Updated 2 years ago
Cannot fully reproduce the Coac vs Chad winrate as reported in CEC2020 using NativeAgent
Updated 2 years ago1
Add Black formatting as a GitHub Action
Updated 2 years ago
Publish to PyPI
Updated 2 years ago
Reformat the code with Black
Updated 2 years ago
Update dependencies and better manage them
Updated 2 years ago
Allow hand/board randomizing on state representations
Closed 2 years ago1
Battle training: self-play opponents use deterministic actions while agents don't
Closed 2 years ago1
Training script: draft needs tensorflow==1.14.0 but battle needs tensorflow>2
Closed 2 years ago
Coac Agent Implementation
Closed 2 years ago4
The engine code and its consistency with the original nim code
Closed 3 years ago5
Player features on battle envs are not normalized
Closed 3 years ago1
The "n-switches" parameter on training.py actually means "amount of times to switch throughout training" and not "amount of episodes to run before switching"
Closed 4 years ago1