def train_tank_1_player_machine_lstm():
game_engine = TankBattle(render=False,
player1_human_control=False,
player2_human_control=False,
two_players=False,
speed=1000,
frame_skip=5,
debug=False
)
env = FruitEnvironment(game_engine,
max_episode_steps=10000,
state_processor=AtariProcessor(),
multi_objective=False)
network_config = AtariA3CLSTMConfig(env,
initial_learning_rate=0.004)
network = PolicyNetwork(network_config,
num_of_checkpoints=40,
using_gpu=True)
agent = A3CLSTMAgent(network, env,
num_of_epochs=10,
steps_per_epoch=1e6,
save_frequency=5e5,
update_network_frequency=4,
log_dir='./train/nips/TankBattle/a3c_gpu_8_threads_tank_time_based_10_lstm_lr_0004',
num_of_threads=8)
agent.train()
def evaluate_tank_1_player_machine():
game_engine = TankBattle(render=True,
player1_human_control=False,
player2_human_control=False,
two_players=False,
speed=60,
frame_skip=5,
debug=False
)
env = FruitEnvironment(game_engine,
max_episode_steps=10000,
state_processor=AtariProcessor(),
multi_objective=False)
network_config = AtariA3CConfig(env)
network = PolicyNetwork(network_config,
using_gpu=True,
load_model_path='./train/nips/TankBattle/a3c_gpu_8_threads_tank_time_based_10_lr_0004_04-10-2018-16-27/model-9500578'
)
agent = A3CAgent(network, env,
num_of_epochs=1,
steps_per_epoch=100000,
report_frequency=1,
log_dir='./thi_test/nips/TankBattle/a3c_gpu_8_threads_tank_time_based_30_49_lr_0004',
num_of_threads=1)
agent.evaluate()
def train_tank_2_player_machine_with_a3c():
game_engine = TankBattle(render=False,
player1_human_control=False,
player2_human_control=False,
two_players=True,
speed=1000,
frame_skip=5,
debug=False,
enable_terminal_one_player_die=True
)
env = MAFruitEnvironment(game_engine,
max_episode_steps=10000,
state_processor=AtariProcessor(),
multi_objective=True)
network_config = AtariA3C2Config(env, initial_learning_rate=0.004)
network = MAPolicyNetwork(network_config,
num_of_checkpoints=80,
using_gpu=True)
agent = MAA3CAgent(network, env,
num_of_epochs=20,
steps_per_epoch=1e6,
save_frequency=5e5,
update_network_frequency=4,
log_dir='./train/nips/TankBattle/a3c_gpu_8_threads_tank_time_based_20_2_players_lr_0004_epoches_20_terminal',
num_of_threads=8)
agent.train()
def evaluate_tank_1_player_machine_with_map():
game_engine = TankBattle(render=True,
player1_human_control=False,
player2_human_control=False,
two_players=False,
speed=200,
frame_skip=5,
debug=False,
num_of_enemies=5,
using_map=True,
multi_target=True,
strategy=3
)
env = FruitEnvironment(game_engine,
max_episode_steps=10000,
state_processor=AtariProcessor(),
multi_objective=True)
network_config = NIPSA3CConfig(env)
network = NIPSPolicyNetwork(network_config,
using_gpu=True,
load_model_path='./train/nips/TankBattle/a3c_gpu_8_threads_tank_with_map_time_based_10_lr_0004_5_enemy_s3_multi_target_04-18-2018-21-16/model-9500310'
)
agent = NIPSA3CAgent(network, env,
num_of_epochs=1,
steps_per_epoch=10000,
report_frequency=1,
log_dir='./thi_test/nips/tankbattle/a3c_gpu_8_threads_tank_with_10_lr_0004',
num_of_threads=1)
a3c_reward = agent.evaluate()
print(a3c_reward)
mr_a3c = np.mean([x[0] for x in a3c_reward])
s_a3c = np.mean([x[2] for x in a3c_reward])
print("Mean", mr_a3c, s_a3c)
def evaluate_tank_2_player_machine_with_a3c():
game_engine = TankBattle(render=True,
player1_human_control=False,
player2_human_control=False,
two_players=True,
speed=90,
frame_skip=5,
debug=False
)
env = MAFruitEnvironment(game_engine,
max_episode_steps=10000,
state_processor=AtariProcessor(),
multi_objective=True)
network_config = AtariA3C2Config(env, initial_learning_rate=0.004)
network = MAPolicyNetwork(network_config,
num_of_checkpoints=40,
load_model_path="./train/nips/TankBattle/a3c_gpu_8_threads_tank_time_based_10_2_players_lr_0004_epoches_10_terminal_05-05-2018-13-02/model-9500513",
using_gpu=True)
agent = MAA3CAgent(network, env,
num_of_epochs=1,
steps_per_epoch=50000,
report_frequency=1,
save_frequency=5e5,
update_network_frequency=4,
log_dir='./thi_test/nips/TankBattle/a3c_gpu_8_threads_tank_time_based_10_2_players_lr_0004',
num_of_threads=1)
a3c_reward = agent.evaluate()
print(a3c_reward)
mr_a3c = np.mean([x[0] for x in a3c_reward])
s_a3c = np.mean([x[2] for x in a3c_reward])
print("Mean", mr_a3c, s_a3c)