args = prs.parse_args()
experiment_time = str(datetime.now()).split('.')[0]
out_csv = 'outputs/2way-single-intersection/sarsa_lambda{}'.format(args.lamb)
write_route_file('nets/2way-single-intersection/single-intersection-gen.rou.xml', 400000, 100000)
env = SumoEnvironment(net_file='nets/2way-single-intersection/single-intersection.net.xml',
single_agent=True,
route_file=args.route,
out_csv_name=out_csv,
use_gui=args.gui,
num_seconds=args.seconds,
min_green=args.min_green,
max_green=args.max_green,
max_depart_delay=0,
time_to_load_vehicles=120,
phases=[
traci.trafficlight.Phase(32, "GGrrrrGGrrrr"),
traci.trafficlight.Phase(2, "yyrrrryyrrrr"),
traci.trafficlight.Phase(32, "rrGrrrrrGrrr"),
traci.trafficlight.Phase(2, "rryrrrrryrrr"),
traci.trafficlight.Phase(32, "rrrGGrrrrGGr"),
traci.trafficlight.Phase(2, "rrryyrrrryyr"),
traci.trafficlight.Phase(32, "rrrrrGrrrrrG"),
traci.trafficlight.Phase(2, "rrrrryrrrrry")
])
for run in range(1, args.runs+1):
obs = env.reset()
agent = TrueOnlineSarsaLambda(env.observation_space, env.action_space, alpha=args.alpha, gamma=args.gamma, epsilon=args.epsilon, fourier_order=21)
done = False
import numpy as np
from sumo_rl.environment.env import SumoEnvironment
import traci
if __name__ == '__main__':
ray.init()
register_env(
"4x4grid",
lambda _: SumoEnvironment(
net_file='nets/4x4-Lucas/4x4.net.xml',
route_file='nets/4x4-Lucas/4x4c1c2c1c2.rou.xml',
out_csv_name='outputs/4x4grid/a3c-4x4grid',
use_gui=False,
num_seconds=80000,
time_to_load_vehicles=120,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(35, "GGGrrr"), # north-south
traci.trafficlight.Phase(2, "yyyrrr"),
traci.trafficlight.Phase(35, "rrrGGG"), # west-east
traci.trafficlight.Phase(2, "rrryyy")
]))
trainer = A3CTrainer(
env="4x4grid",
config={
"multiagent": {
"policy_graphs": {
'0':
(A3CTFPolicy, spaces.Box(low=np.zeros(10),
high=np.ones(10)),
args = prs.parse_args()
experiment_time = str(datetime.now()).split('.')[0]
out_csv = 'outputs/2way-single-intersection/{}_alpha{}_gamma{}_eps{}_decay{}_reward{}'.format(
experiment_time, args.alpha, args.gamma, args.epsilon, args.decay,
args.reward)
env = SumoEnvironment(
net_file='nets/2way-single-intersection/single-intersection.net.xml',
route_file=args.route,
out_csv_name=out_csv,
use_gui=args.gui,
num_seconds=args.seconds,
min_green=args.min_green,
max_green=args.max_green,
max_depart_delay=0,
time_to_load_vehicles=120,
phases=[
traci.trafficlight.Phase(32, "GGrrrrGGrrrr"),
traci.trafficlight.Phase(2, "yyrrrryyrrrr"),
traci.trafficlight.Phase(32, "rrGrrrrrGrrr"),
traci.trafficlight.Phase(2, "rryrrrrryrrr"),
traci.trafficlight.Phase(32, "rrrGGrrrrGGr"),
traci.trafficlight.Phase(2, "rrryyrrrryyr"),
traci.trafficlight.Phase(32, "rrrrrGrrrrrG"),
traci.trafficlight.Phase(2, "rrrrryrrrrry")
])
if args.reward == 'queue':
env._compute_rewards = env._queue_average_reward
else:
env._compute_rewards = env._waiting_time_reward
for run in range(1, args.runs + 1):
if __name__ == '__main__':
experiment_time = str(datetime.now()).split('.')[0]
out_csv = 'outputs/big-intersection/sarsa-f7' #+ experiment_time
env = SumoEnvironment(
net_file='nets/big-intersection/big-intersection.net.xml',
single_agent=True,
route_file='nets/big-intersection/routes.rou.xml',
out_csv_name=out_csv,
use_gui=False,
num_seconds=5400,
yellow_time=4,
min_green=5,
max_green=60,
max_depart_delay=300,
time_to_load_vehicles=0,
phases=[
traci.trafficlight.Phase(30, "GGGGrrrrrrGGGGrrrrrr"),
traci.trafficlight.Phase(4, "yyyyrrrrrryyyyrrrrrr"),
traci.trafficlight.Phase(15, "rrrrGrrrrrrrrrGrrrrr"),
traci.trafficlight.Phase(4, "rrrryrrrrrrrrryrrrrr"),
traci.trafficlight.Phase(30, "rrrrrGGGGrrrrrrGGGGr"),
traci.trafficlight.Phase(4, "rrrrryyyyrrrrrryyyyr"),
traci.trafficlight.Phase(15, "rrrrrrrrrGrrrrrrrrrG"),
traci.trafficlight.Phase(4, "rrrrrrrrryrrrrrrrrry")
])
fixed_tl = False
agent = TrueOnlineSarsaLambda(env.observation_space,
env.action_space,
alpha=0.000000001,
# if id == 'gneJ00':
# return 'gneJ00'
# else:
return 'offset_agent'
if __name__ == '__main__':
ray.init()
register_env(
"2TLS", lambda _: SumoEnvironment(
net_file=
'/home/sonic/Desktop/sumo-rl-research-offset/sumo-rl-research/experiments/nets/Research/case04/intersection.net.xml',
route_file=
'/home/sonic/Desktop/sumo-rl-research-offset/sumo-rl-research/experiments/nets/Research/case04/intersection.rou.xml',
out_csv_path='outputs/case04/',
out_csv_name='DDPG',
use_gui=False,
num_seconds=12240612,
time_to_load_vehicles=612,
max_depart_delay=0))
trainer = DDPGTrainer(
env="2TLS",
config={
"multiagent": {
"policy_graphs": {
'offset_agent': (DDPGTFPolicy,
spaces.Box(low=np.zeros(2),
high=np.array(['inf'] * 2)),
spaces.Box(low=np.array([0, 0]),
import numpy as np
from sumo_rl.environment.env import SumoEnvironment
import traci
if __name__ == '__main__':
env = SumoEnvironment(net_file='nets/2way-single-intersection/single-intersection.net.xml',
route_file='nets/2way-single-intersection/single-intersection-vhvh.rou.xml',
out_csv_name='outputs/2way-single-intersection/dqn-vhvh2-stable-mlp-bs',
single_agent=True,
use_gui=True,
num_seconds=100000,
time_to_load_vehicles=120,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(32000, 32000, 32000, "GGrrrrGGrrrr"),
traci.trafficlight.Phase(2000, 2000, 2000, "yyrrrryyrrrr"),
traci.trafficlight.Phase(32000, 32000, 32000, "rrGrrrrrGrrr"),
traci.trafficlight.Phase(2000, 2000, 2000, "rryrrrrryrrr"),
traci.trafficlight.Phase(32000, 32000, 32000, "rrrGGrrrrGGr"),
traci.trafficlight.Phase(2000, 2000, 2000, "rrryyrrrryyr"),
traci.trafficlight.Phase(32000, 32000, 32000, "rrrrrGrrrrrG"),
traci.trafficlight.Phase(2000, 2000, 2000, "rrrrryrrrrry")
])
model = DQN(
env=env,
policy=MlpPolicy,
learning_rate=1e-3,
buffer_size=50000,
exploration_fraction=0.1,
if __name__ == '__main__':
ray.init()
register_env(
"2way-single-intersection", lambda _: SumoEnvironment(
net_file=
'nets/2way-single-intersection/single-intersection.net.xml',
route_file=
'nets/2way-single-intersection/single-intersection-gen.rou.xml',
out_csv_name='outputs/2way-single-intersection/a3c-contexts',
use_gui=False,
num_seconds=100000,
time_to_load_vehicles=120,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(32000, "GGrrrrGGrrrr"),
traci.trafficlight.Phase(2000, "yyrrrryyrrrr"),
traci.trafficlight.Phase(32000, "rrGrrrrrGrrr"),
traci.trafficlight.Phase(2000, "rryrrrrryrrr"),
traci.trafficlight.Phase(32000, "rrrGGrrrrGGr"),
traci.trafficlight.Phase(2000, "rrryyrrrryyr"),
traci.trafficlight.Phase(32000, "rrrrrGrrrrrG"),
traci.trafficlight.Phase(2000, "rrrrryrrrrry")
]))
trainer = A3CAgent(
env="2way-single-intersection",
config={
"multiagent": {
"policy_graphs": {
return '0'
if __name__ == '__main__':
ray.init()
register_env(
"2x2grid", lambda _: SumoEnvironment(
net_file='nets/2x2grid/2x2.net.xml',
route_file='nets/2x2grid/2x2.rou.xml',
out_csv_name='outputs/2x2grid/a3c-contexts',
use_gui=False,
num_seconds=100000,
time_to_load_vehicles=120,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(32, "GGrrrrGGrrrr"),
traci.trafficlight.Phase(2, "yyrrrryyrrrr"),
traci.trafficlight.Phase(32, "rrGrrrrrGrrr"),
traci.trafficlight.Phase(2, "rryrrrrryrrr"),
traci.trafficlight.Phase(32, "rrrGGrrrrGGr"),
traci.trafficlight.Phase(2, "rrryyrrrryyr"),
traci.trafficlight.Phase(32, "rrrrrGrrrrrG"),
traci.trafficlight.Phase(2, "rrrrryrrrrry")
]))
trainer = A3CTrainer(
env="2x2grid",
config={
"multiagent": {
"policy_graphs": {
'0':
if __name__ == '__main__':
alpha = 0.1
gamma = 0.99
decay = 1
runs = 1
env = SumoEnvironment(
net_file='nets/4x4-Lucas/4x4.net.xml',
route_file='nets/4x4-Lucas/4x4c2.rou.xml',
use_gui=True,
num_seconds=40000,
time_to_load_vehicles=300,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(35000, 35000, 35000,
"GGGrrr"), # north-south
traci.trafficlight.Phase(2000, 2000, 2000, "yyyrrr"),
traci.trafficlight.Phase(35000, 35000, 35000,
"rrrGGG"), # west-east
traci.trafficlight.Phase(2000, 2000, 2000, "rrryyy")
])
for run in range(1, runs + 1):
initial_states = env.reset()
ql_agents = {
ts:
QLAgent(starting_state=env.encode(initial_states[ts]),
state_space=env.observation_space,
action_space=env.action_space,
100000)
# multiprocess environment
n_cpu = 2
env = SubprocVecEnv([
lambda: SumoEnvironment(
net_file='nets/2way-single-intersection/single-intersection.net.xml',
route_file=
'nets/2way-single-intersection/single-intersection-gen.rou.xml',
out_csv_name=
'outputs/2way-single-intersection/a2c-contexts-5s-vmvm-400k',
single_agent=True,
use_gui=False,
num_seconds=400000,
min_green=5,
time_to_load_vehicles=120,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(32000, "GGrrrrGGrrrr"),
traci.trafficlight.Phase(2000, "yyrrrryyrrrr"),
traci.trafficlight.Phase(32000, "rrGrrrrrGrrr"),
traci.trafficlight.Phase(2000, "rryrrrrryrrr"),
traci.trafficlight.Phase(32000, "rrrGGrrrrGGr"),
traci.trafficlight.Phase(2000, "rrryyrrrryyr"),
traci.trafficlight.Phase(32000, "rrrrrGrrrrrG"),
traci.trafficlight.Phase(2000, "rrrrryrrrrry")
]) for i in range(n_cpu)
])
model = A2C(MlpPolicy,
env,
verbose=1,
return "4708662059"
else:
return "5870232715"
if __name__ == '__main__':
ray.init()
register_env(
"2TLS", lambda _: SumoEnvironment(
net_file=
'/home/sonic/Desktop/sumo-rl-suwon/sumo-rl-suwon/experiments/nets/suwon/osm_rl.net.xml',
route_file=
'/home/sonic/Desktop/sumo-rl-suwon/sumo-rl-suwon/experiments/nets/suwon/osm.rou.xml',
out_csv_path='outputs/Test4/',
out_csv_name='DQN',
use_gui=True,
num_seconds=44000,
time_to_load_vehicles=300,
max_depart_delay=0))
trainer = DQNTrainer(
env="2TLS",
config={
"multiagent": {
"policy_graphs": {
'3210041371': (DQNTFPolicy,
spaces.Box(low=np.zeros(16),
high=np.array(['inf'] * 16)),
spaces.Discrete(2), {}),
args = prs.parse_args()
ns = args.ns * 1000
we = args.we * 1000
experiment_time = str(datetime.now()).split('.')[0]
out_csv = 'outputs/single-intersection/{}_alpha{}_gamma{}_eps{}_decay{}_reward{}'.format(
experiment_time, args.alpha, args.gamma, args.epsilon, args.decay,
args.reward)
env = SumoEnvironment(
net_file='nets/single-intersection/single-intersection.net.xml',
route_file=args.route,
out_csv_name=out_csv,
use_gui=args.gui,
num_seconds=args.seconds,
min_green=args.min_green,
max_green=args.max_green,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(ns, ns, ns, "GGrr"), # north-south
traci.trafficlight.Phase(2000, 2000, 2000, "yyrr"),
traci.trafficlight.Phase(we, we, we, "rrGG"), # west-east
traci.trafficlight.Phase(2000, 2000, 2000, "rryy")
])
if args.reward == 'queue':
env._compute_rewards = env._queue_average_reward
else:
env._compute_rewards = env._waiting_time_reward
for run in range(1, args.runs + 1):
initial_states = env.reset()
ql_agents = {
prs.add_argument("-s", dest="seconds", type=int, default=100000, required=False, help="Number of simulation seconds.\n")
prs.add_argument("-r", dest="reward", type=str, default='wait', required=False, help="Reward function: [-r queue] for average queue reward or [-r wait] for waiting time reward.\n")
prs.add_argument("-v", action="store_true", default=False, help="Print experience tuple.\n")
prs.add_argument("-runs", dest="runs", type=int, default=1, help="Number of runs.\n")
args = prs.parse_args()
experiment_time = str(datetime.now()).split('.')[0]
out_csv = '{}_alpha{}_gamma{}_eps{}_decay{}_reward{}'.format(experiment_time, args.alpha, args.gamma, args.epsilon, args.decay, args.reward)
env = SumoEnvironment(net_file='nets/single-intersection/single-intersection.net.xml',
single_agent=True,
route_file=args.route,
out_csv_name=out_csv,
use_gui=args.gui,
num_seconds=args.seconds,
min_green=args.min_green,
max_green=args.max_green,
max_depart_delay=0,
phases=[
traci.trafficlight.Phase(args.ns, "GGrr"), # north-south
traci.trafficlight.Phase(2, "yyrr"),
traci.trafficlight.Phase(args.we, "rrGG"), # west-east
traci.trafficlight.Phase(2, "rryy")
])
env = VisualizationEnv(
env=env,
episodic=False,
features_names=['Phase 0', 'Phase 1', 'Elapsed time'] + ['Density lane ' + str(i) for i in range(4)] + ['Queue lane ' + str(i) for i in range(4)],
actions_names=['Phase 0', 'Phase 1']
)
