htrbao / ppo_m3_simulator

🎲 Match 3 Agent using PPO

PPO with M3 Simulator

Thanks for kamildar/gym-match3 publish the baseline of building gymnasium match-3 games

Configurations

M3 Simulator Setup

1. You need to clone this repo into local first:

   git clone https://github.com/htrbao/PPO_M3_Simulator

2. Install the requirements

   • Using conda:

     conda create -n m3_simu python=3.9
     conda activate m3_simu
     conda install pytorch pytorch-cuda=11.8 -c pytorch -c nvidia
     cd gym-match3
     pip install -e .
     pip install -r requirements.txt
     

   • Using venv:

     python -m venv ./venv
     path/to/venv/Scripts/activate
     pip3 install torch --index-url https://download.pytorch.org/whl/cu118
     cd gym-match3
     pip install -e .
     pip install -r requirements.txt

M3 Simulator Usage

1. Please refer to this notebook to know how to use gym-match3 environment.

M3 Simulator Levels Configs

1. You can also contribute your custom levels by append it into LEVELS list within gym_match3.envs.levels

   For example:

   Level(h=10, w=9, n_shape=6, board=[
       [-1, -1, -1, -1,  0, -1, -1, -1, -1],
       [-1, -1, -1,  0,  0,  0, -1, -1, -1],
       [-1, -1,  0,  0,  0,  0,  0, -1, -1],
       [-1,  0,  0,  0,  0,  0,  0,  0, -1],
       [ 0,  0,  0,  0,  0,  0,  0,  0,  0],
       [-1,  0,  0,  0,  0,  0,  0,  0, -1],
       [-1, -1,  0,  0,  0,  0,  0, -1, -1],
       [-1, -1, -1,  0,  0,  0, -1, -1, -1],
       [-1, -1, -1, -1,  0, -1, -1, -1, -1],
       [-1, -1, -1, -1, -1, -1, -1, -1, -1],
   ])
   

M3 Training

• To understand board on wandb, you can follow stackexchange.