def run_mstar(args, num_trials, render_len):
env = Independent_NavigationV8_0(args)
TIME_LIMIT = 10*60
INFLATION = 1.1
render_frames = []
results = {"agents_on_goal": [],
"all_done": [],
"episode_length": [],
"execution_time": [],
"obstacle_collisions": [],
"agent_collisions": []}
def signal_handler(signum, frame):
raise Exception("Time expired")
handle = signal.signal(signal.SIGALRM, signal_handler)
for i in range(num_trials):
render_cntr = render_len
_ = env.reset()
start_pos = make_start_postion_list(env)
end_pos = make_end_postion_list(env)
all_actions, time_taken = None, None
signal.alarm(TIME_LIMIT)
try:
all_actions, time_taken = env.graph.mstar_search4_OD(start_pos, end_pos, inflation=INFLATION, return_time_taken=True)
except:
print("ODMstar time ran out of time")
finally:
signal.signal(signal.SIGALRM, handle)
signal.alarm(0)
if all_actions is not None:
results['agents_on_goal'].append(1)
results["all_done"].append(1)
results["episode_length"].append(len(all_actions))
else:
results['agents_on_goal'].append(0)
results["all_done"].append(0)
results["episode_length"].append(None)
if time_taken is not None:
results["execution_time"].append(float(time_taken))
else:
results["execution_time"].append(None)
if render_cntr > 0 and all_actions is not None:
render_cntr -= 1
for a in all_actions:
_, _, _, _ = env.step(a)
frame = env.render()
render_frames.append(frame)
return render_frames, results
def run_bc(args, num_trials, render_len = 5):
CHECKPOINT_PATH = "/home/james/Desktop/Gridworld/EXPERIMENTS/FINAL_COMPARISON/Checkpoint_Policies/bc/checkpoint_20"
env_name = "independent_navigation-v8_0"
args.env_name = env_name
env = Independent_NavigationV8_0(args)
obs_space = env.observation_space[-1]
ppo = PPO(5, obs_space, "primal7", env.n_agents, True, True, 8, 512,recurrent=False)
model_info = torch.load(CHECKPOINT_PATH)
ppo.load(model_info)
DEVICE = 'cpu'
render_frames, results = benchmark_func(args, True, ppo, num_trials, render_len, DEVICE)
return render_frames, results
#Environment:
parser.add_argument("--map_shape", default = 32, type=int)
parser.add_argument("--n_agents", default = 20, type=int)
parser.add_argument("--view_d", default = 3, type=int)
parser.add_argument("--env_name", default = name, type= str)
parser.add_argument("--use_default_rewards", default=True, type=bool)
parser.add_argument("--obj_density", default = 0.2, type=float)
parser.add_argument("--use_custom_rewards", default = False, action='store_true')
parser.add_argument("--ppo_heur_block", default= False, type=bool)
parser.add_argument("--inflation", default = 1.1, type=float)
parser.add_argument("--step_time", default = 0.25, type=float, help="Time taken between rendered env steps.")
args = parser.parse_args()
args.map_shape = (args.map_shape, args.map_shape)
t0 = time.time()
env = Independent_NavigationV8_0(args)
t01 = time.time()
obs = env.reset()
obs = env.reset()
def make_start_postion_list(env_hldr):
'''Assumes agent keys in evn.agents is the same as agent id's '''
start_positions = []
for i in range(len(env_hldr.agents.values())):
start_positions.append(env.agents[i].pos)
return start_positions
def make_end_postion_list(env_hldr):
'''Assumes agent keys in evn.agents is the same as agent id's '''
end_positions = []
def mstar50():
from Env.env import Independent_NavigationV8_0
FILE_PATH = "/home/james/Desktop/Gridworld/EXPERIMENTS/odmstarFinal/"
#FILE_PATH = "/home/jellis/workspace5/gridworld2/odmstarFinal/"
NUM_TRIALS = 20
TIME_LIMIT = 5*60
name ="test_mstar-v0"
mode = "human"
parser = argparse.ArgumentParser("Testing")
#Environment:
parser.add_argument("--map_shape", default = (50,50), type=object)
parser.add_argument("--n_agents", default = 10, type=int)
parser.add_argument("--view_d", default = 3, type=int)
parser.add_argument("--env_name", default = name, type= str)
parser.add_argument("--use_default_rewards", default=True, type=bool)
parser.add_argument("--obj_density", default = 0.2, type=int)
parser.add_argument("--use_custom_rewards", default = False, action='store_true')
parser.add_argument("--custom_env_ind", default= 1, type=int)
parser.add_argument("--ppo_heur_block", default=False, type=bool)
args = parser.parse_args()
def make_start_postion_list(env_hldr):
'''Assumes agent keys in evn.agents is the same as agent id's '''
start_positions = []
for i in range(len(env_hldr.agents.values())):
start_positions.append(env.agents[i].pos)
return start_positions
def make_end_postion_list(env_hldr):
'''Assumes agent keys in evn.agents is the same as agent id's '''
end_positions = []
for i in range(len(env_hldr.goals.values())):
end_positions.append(env.goals[i].pos)
return end_positions
def signal_handler(signum, frame):
raise Exception("Time expired")
signal.signal(signal.SIGALRM, signal_handler)
for inf in [1.1,2.0,10.0]:
for ob_dens in [0.0,0.1,0.2,0.3]:
for n in [120,80,20]:
args.n_agents = n
args.obj_density = ob_dens
name = "map_shape_" + str(args.map_shape[0]) + "_density_" + str(ob_dens) + "_nagents_" + str(n) + "_inf_" + str(inf)
env = Independent_NavigationV8_0(args)
print("Running inflation: {} nagents: {} Density: {}".format(inf, n, ob_dens))
# data = {
# "name": name,
# "od_mstar_time": [],
# "od_mstar_length": [],
# }
results = {"agents_on_goal": [],
"all_done": [],
"episode_length": [],
"execution_time": [],
"obstacle_collisions": [],
"agent_collisions": []}
#name_hldr = data["name"]
#data["name"] = data["name"] + "_nagents_" + str(n) + "_inflation_" + str(inf)
for i in range(NUM_TRIALS):
_ = env.reset()
start_pos = make_start_postion_list(env)
end_pos = make_end_postion_list(env)
all_actions = None
time_taken = None
signal.alarm(TIME_LIMIT)
try:
all_actions, time_taken = env.graph.mstar_search4_OD(start_pos, end_pos, inflation=inf, return_time_taken=True)
except:
print("ODMstar time ran out of time")
if all_actions is not None:
results['agents_on_goal'].append(1)
results["all_done"].append(1)
results["episode_length"].append(len(all_actions))
else:
results['agents_on_goal'].append(0)
results["all_done"].append(0)
results["episode_length"].append(None)
if time_taken is not None:
results["execution_time"].append(float(time_taken))
else:
results["execution_time"].append(None)
torch.save(results, os.path.join(FILE_PATH, name+".pkl"))
def mstar_comparison_1_0_0():
from Env.env import Independent_NavigationV8_0
FILE_PATH = "/home/james/Desktop/Gridworld/EXPERIMENTS/odmstar_and_odrmstar_comparison2/"
NUM_TRIALS = 100
TIME_LIMIT = 5*60 + 1
name ="test_mstar-v0"
mode = "human"
parser = argparse.ArgumentParser("Testing")
#Environment:
parser.add_argument("--map_shape", default = (32,32), type=object)
parser.add_argument("--n_agents", default = 10, type=int)
parser.add_argument("--view_d", default = 3, type=int)
parser.add_argument("--env_name", default = name, type= str)
parser.add_argument("--use_default_rewards", default=True, type=bool)
parser.add_argument("--obj_density", default = 0.2, type=int)
parser.add_argument("--use_custom_rewards", default = False, action='store_true')
parser.add_argument("--custom_env_ind", default= 1, type=int)
parser.add_argument("--ppo_heur_block", default=False, type=bool)
args = parser.parse_args()
env = Independent_NavigationV8_0(args)
print(env.summary())
obs = env.reset()
def make_start_postion_list(env_hldr):
'''Assumes agent keys in evn.agents is the same as agent id's '''
start_positions = []
for i in range(len(env_hldr.agents.values())):
start_positions.append(env.agents[i].pos)
return start_positions
def make_end_postion_list(env_hldr):
'''Assumes agent keys in evn.agents is the same as agent id's '''
end_positions = []
for i in range(len(env_hldr.goals.values())):
end_positions.append(env.goals[i].pos)
return end_positions
def signal_handler(signum, frame):
raise Exception("Time expired")
signal.signal(signal.SIGALRM, signal_handler)
for inf in [1.1]:
for n in [40,50,60]:
args.n_agents = n
env = Independent_NavigationV8_0(args)
print("Running inflation: {} nagents: {}".format(inf, n))
data = {
"name": "32x32_density_0.2",
"od_mstar_time": [],
"od_mstar_length": [],
"odr_mstar_time": [],
"odr_mstar_length": []
}
#name_hldr = data["name"]
data["name"] = data["name"] + "_nagents_" + str(n) + "_inflation_" + str(inf)
for i in range(NUM_TRIALS):
obs = env.reset()
start_pos = make_start_postion_list(env)
end_pos = make_end_postion_list(env)
print("Start: {} End {}".format(start_pos, end_pos))
#Mstar_OD
all_actions = None
time_taken = None
signal.alarm(TIME_LIMIT)
try:
all_actions, time_taken = env.graph.mstar_search4_OD(start_pos, end_pos, inflation=inf, return_time_taken=True)
except:
print("ODMstar time ran out of time")
if time_taken is not None:
data["od_mstar_time"].append(float(time_taken))
else:
data["od_mstar_time"].append(None)
if all_actions is not None:
data["od_mstar_length"].append(len(all_actions))
else:
data["od_mstar_length"].append(None)
# Mstar_ODr
all_actions = None
time_taken = None
signal.alarm(TIME_LIMIT)
try:
all_actions, time_taken = env.graph.mstar_search_ODrMstar(start_pos, end_pos, inflation=inf, return_time_taken=True)
except:
print("ODrMstar time ran out of time")
if time_taken is not None:
data["odr_mstar_time"].append(float(time_taken))
else:
data["odr_mstar_time"].append(time_taken)
if all_actions is not None:
data["odr_mstar_length"].append(len(all_actions))
else:
data["odr_mstar_length"].append(None)
torch.save(data, os.path.join(FILE_PATH, data["name"]+".pkl"))