This package allow you to turn any gameboy memory event into a reinforcement learning environment rule.
Install the dependencies:
pip install -r requirements.txt
Install the package with pip
:
pip install ./
import gym
import pyboyenv
# Create the environment
env = gym.make('Pyboy-v0', game=<FILE>)
# Add rules
env.set_reward_rule(ADDRESS, TYPE, VALUE, LABEL)
# Available rules:
# - increase: add VALUE if the memory at the address ADDRESS increases
# - decrease: add VALUE if the memory at the address ADDRESS decreases
# - smaller X: add VALUE if the memory at the address ADDRESS is smaller than X
# - bigger X: add VALUE if the memory at the address ADDRESS is bigger than X
# - equals X: add VALUE if the memory at the address ADDRESS equals X
# - in X1,..,XN: add VALUE if the memory at the address ADDRESS is equal to X1 or ... or XN
# Useful values
cumul = 0 # Sum of rewards
done = False # Is the game done ?
action = 16 # Initial action (action list available below, 16=nothing)
# Game loop
while not done:
# State = Window screen
state, reward, done, info = env.step(action)
cumul += reward
action = Agent.get_action(state, cumul, reward) # Next action
# Print infos
for i in info:
print(f"{i[0]}: {i[1]}")
- 0:
WindowEvent.PRESS_ARROW_UP
- 1:
WindowEvent.PRESS_ARROW_DOWN
- 2:
WindowEvent.PRESS_ARROW_LEFT
- 3:
WindowEvent.PRESS_ARROW_RIGHT
- 4:
WindowEvent.PRESS_BUTTON_A
- 5:
WindowEvent.PRESS_BUTTON_B
- 6:
WindowEvent.PRESS_BUTTON_SELECT
- 7:
WindowEvent.PRESS_BUTTON_START
- 8:
WindowEvent.RELEASE_ARROW_UP
- 9:
WindowEvent.RELEASE_ARROW_DOWN
- 10:
WindowEvent.RELEASE_ARROW_LEFT
- 11:
WindowEvent.RELEASE_ARROW_RIGHT
- 12:
WindowEvent.RELEASE_BUTTON_A
- 13:
WindowEvent.RELEASE_BUTTON_B
- 14:
WindowEvent.RELEASE_BUTTON_SELECT
- 15:
WindowEvent.RELEASE_BUTTON_START
- 16:
WindowEvent.PASS