from agents import DeepQLearningAgent, RandomAgent, TabQLearningAgent, DeepQLearningExperienceReplayAgent, \
DoubleDeepQLearningAgent
from environments.battle_royale import BattleRoyalGameWorldTerminal, BattleRoyale
from runners import run_for_n_games_and_print_stats, run_step
import tensorflow as tf
if __name__ == "__main__":
tf.compat.v1.disable_eager_execution()
list_agent=[DoubleDeepQLearningAgent(action_space_size=48) if i <7 else TabQLearningAgent() for i in range(2)]
for _ in range(100):
gs = BattleRoyalGameWorldTerminal(0,numberofPlayer=2,list_agent = list_agent)
gs.run()
list_agent[0].epsilon = -1
list_agent[1].epsilon = -1
gs2 = BattleRoyale(numberofPlayer=2,list_agent=list_agent)
gs2.run()
from agents import DeepQLearningAgent, RandomAgent, TabQLearningAgent
from environments.battle_royale import BattleRoyalGameWorldTerminal, BattleRoyale
from runners import run_for_n_games_and_print_stats, run_step
if __name__ == "__main__":
list_agent = [
TabQLearningAgent() if i < 7 else RandomAgent() for i in range(6)
]
for i in range(1000):
gs = BattleRoyalGameWorldTerminal(i, list_agent=list_agent)
gs.run()
#list_agent[0].epsilon =-1
#list_agent[1].epsilon = -1
gs2 = BattleRoyale(list_agent=list_agent)
gs2.run()
from agents import TabQLearningAgent
from environments import GridWorldGameState
from runners import run_for_n_games_and_print_stats, run_step
if __name__ == "__main__":
gs = GridWorldGameState()
agent = TabQLearningAgent()
for _ in range(500):
run_for_n_games_and_print_stats([agent], gs, 100)
agent.epsilon = -1.0
run_for_n_games_and_print_stats([agent], gs, 100)
gs = gs.clone()
while not gs.is_game_over():
run_step([agent], gs)
print(gs)
from agents import DeepQLearningAgent, RandomAgent, TabQLearningAgent, DeepQLearningExperienceReplayAgent, \
DoubleDeepQLearningAgent, DoubleDeepQLearningExprerienceReplayAgent
from environments.battle_royale import BattleRoyalGameWorldTerminal, BattleRoyale
from runners import run_for_n_games_and_print_stats, run_step
import tensorflow as tf
if __name__ == "__main__":
tf.compat.v1.disable_eager_execution()
list_agent = [
DoubleDeepQLearningExprerienceReplayAgent(
action_space_size=48) if i < 7 else TabQLearningAgent()
for i in range(2)
]
for i in range(100):
gs = BattleRoyalGameWorldTerminal(i,
numberofPlayer=2,
list_agent=list_agent)
gs.run()
#list_agent[0].epsilon = -1
#list_agent[1].epsilon = -1
gs2 = BattleRoyale(numberofPlayer=2, list_agent=list_agent)
gs2.run()
from agents import TabQLearningAgent, CommandLineAgent, RandomAgent
from environments.tictactoe import TicTacToeGameState
from runners import run_for_n_games_and_print_stats, run_step
if __name__ == "__main__":
gs = TicTacToeGameState()
agent0 = TabQLearningAgent()
agent1 = TabQLearningAgent()
agent0.alpha = 0.1
agent0.epsilon = 0.005
agent1.alpha = 0.1
agent1.epsilon = 0.005
for _ in range(100):
run_for_n_games_and_print_stats([agent0, agent1], gs, 5000)
agent0.epsilon = -1.0
agent1.epsilon = -1.0
run_for_n_games_and_print_stats([agent0, agent1], gs, 100)
gs_clone = gs.clone()
while not gs_clone.is_game_over():
run_step([agent0, CommandLineAgent()], gs_clone)
print(gs_clone)
gs_clone = gs.clone()
while not gs_clone.is_game_over():
run_step([CommandLineAgent(), agent1], gs_clone)
print(gs_clone)
from agents import RandomAgent, TabQLearningAgent
from environments.battle_royale import BattleRoyale
from runners import run_for_n_games_and_print_stats, run_step
if __name__ == "__main__":
list_agent = list([TabQLearningAgent() for i in range(6)])
gs = BattleRoyale(list_agent=list_agent)
gs.run()
def run_BattleRoyal(i):
list_agent = [DeepQLearningAgent(action_space_size=48) if i < 3 else TabQLearningAgent() for i in range(6)]
Terminalworld = BattleRoyalGameWorldTerminal(i,list_agent=list_agent)
a = Terminalworld.run()
return a