def main(**kargs):
initial_weights_file, initial_i_frame = latest(kargs['weights_dir'])
print("Continuing using weights from file: ", initial_weights_file, "from", initial_i_frame)
if kargs['theano_verbose']:
theano.config.compute_test_value = 'warn'
theano.config.exception_verbosity = 'high'
theano.config.optimizer = 'fast_compile'
ale = ag.init(display_screen=(kargs['visualize'] == 'ale'), record_dir=kargs['record_dir'])
game = ag.SpaceInvadersGame(ale)
def new_game():
game.ale.reset_game()
game.finished = False
game.cum_reward = 0
game.lives = 4
return game
replay_memory = dqn.ReplayMemory(size=kargs['dqn.replay_memory_size']) if not kargs['dqn.no_replay'] else None
# dqn_algo = q.ConstAlgo([3])
dqn_algo = dqn.DQNAlgo(game.n_actions(),
replay_memory=replay_memory,
initial_weights_file=initial_weights_file,
build_network=kargs['dqn.network'],
updates=kargs['dqn.updates'])
dqn_algo.replay_start_size = kargs['dqn.replay_start_size']
dqn_algo.final_epsilon = kargs['dqn.final_epsilon']
dqn_algo.initial_epsilon = kargs['dqn.initial_epsilon']
dqn_algo.i_frames = initial_i_frame
dqn_algo.log_frequency=kargs['dqn.log_frequency']
import Queue
dqn_algo.mood_q = Queue.Queue() if kargs['show_mood'] else None
if kargs['show_mood'] is not None:
plot = kargs['show_mood']()
def worker():
while True:
item = dqn_algo.mood_q.get()
plot.show(item)
dqn_algo.mood_q.task_done()
import threading
t = threading.Thread(target=worker)
t.daemon = True
t.start()
print(str(dqn_algo))
visualizer = ag.SpaceInvadersGameCombined2Visualizer() if kargs['visualize'] == 'q' else q.GameNoVisualizer()
teacher = q.Teacher(new_game, dqn_algo, visualizer,
ag.Phi(skip_every=4), repeat_action=4, sleep_seconds=0)
teacher.teach(500000)
def const_on_space_invaders():
import teacher as q
import ale_game as ag
import dqn
reload(q)
reload(ag)
reload(dqn)
ale = ag.init()
game = ag.SpaceInvadersGame(ale)
def new_game():
game.ale.reset_game()
game.finished = False
game.cum_reward = 0
return game
const_algo = q.ConstAlgo([2, 2, 2, 2, 2, 0, 0, 0, 0])
teacher = q.Teacher(new_game, const_algo, ag.SpaceInvadersGameCombined2Visualizer(),
ag.Phi(skip_every=6), repeat_action=6)
teacher.teach(1)