def setUpClass(cls):
super().setUpClass()
cls.agent = TD3(
state_shape=cls.continuous_env.observation_space.shape,
action_dim=cls.continuous_env.action_space.low.size,
batch_size=cls.batch_size,
gpu=-1)
def policy_fn(env, name, memory_capacity=int(1e6), gpu=-1):
return TD3(
state_shape=env.observation_space.shape,
action_dim=env.action_space.high.size,
gpu=gpu,
name=name,
batch_size=100,
memory_capacity=memory_capacity)
def main():
parser = Trainer.get_argument()
parser = OnPolicyTrainer.get_argument(parser)
parser = SAC.get_argument(parser)
parser = PPO.get_argument(parser)
parser = TD3.get_argument(parser)
parser.add_argument('--SAC', action='store_true')
parser.add_argument('--PPO', action='store_true')
parser.add_argument('--TD3', action='store_true')
parser.add_argument('--DEBUG', action='store_true')
parser.add_argument('--env', type=int, default=0)
parser.set_defaults(batch_size=32) #100
parser.set_defaults(n_warmup=10000) #10000
parser.set_defaults(max_steps=2e6)
parser.set_defaults(gpu=0)
parser.set_defaults(test_interval=200 * 100)
parser.set_defaults(test_episodes=3)
args = parser.parse_args()
print(vars(args))
run(parser)
from tf2rl.algos.td3 import TD3
from tf2rl.experiments.trainer import Trainer
from arm_env import ArmEnvironment
import rospy
import numpy as np
normalise_obs = False
static_goal = False
testing = False
slow_step = True
# conda activate RL_arm_noetic && source devel/setup.bash && tensorboard --logdir /home/devon/RL_Arm_noetic/results
parser = Trainer.get_argument()
parser = TD3.get_argument(parser)
parser.set_defaults(batch_size=100)
slow_suffix = "slow" if(slow_step) else ""
if(static_goal and normalise_obs):
parser.set_defaults(model_dir='model_TD3_static_normed'+slow_suffix)
parser.set_defaults(logdir='results/TD3_static_normed'+slow_suffix)
elif(static_goal):
parser.set_defaults(model_dir='model_TD3_static'+slow_suffix)
parser.set_defaults(logdir='results/TD3_static'+slow_suffix)
elif(normalise_obs):
parser.set_defaults(model_dir='model_TD3_normed'+slow_suffix)
parser.set_defaults(logdir='results/TD3_normed'+slow_suffix)
else:
parser.set_defaults(model_dir='model_TD3'+slow_suffix)
parser.set_defaults(logdir='results/TD3'+slow_suffix)
parser.set_defaults(normalise_obs=normalise_obs)
import gym
from tf2rl.algos.td3 import TD3
from tf2rl.experiments.trainer import Trainer
if __name__ == '__main__':
parser = Trainer.get_argument()
parser = TD3.get_argument(parser)
parser.add_argument('--env-name', type=str, default="Pendulum-v0")
parser.set_defaults(batch_size=100)
parser.set_defaults(n_warmup=10000)
args = parser.parse_args()
env = gym.make(args.env_name)
test_env = gym.make(args.env_name)
policy = TD3(
state_shape=env.observation_space.shape,
action_dim=env.action_space.high.size,
gpu=args.gpu,
memory_capacity=args.memory_capacity,
max_action=env.action_space.high[0],
batch_size=args.batch_size,
n_warmup=args.n_warmup)
trainer = Trainer(policy, env, args, test_env=test_env)
if args.evaluate:
trainer.evaluate_policy_continuously()
else:
trainer()
def run(parser):
args = parser.parse_args()
if args.gpu < 0:
tf.config.experimental.set_visible_devices([], 'GPU')
else:
physical_devices = tf.config.list_physical_devices('GPU')
tf.config.set_visible_devices(physical_devices[args.gpu], 'GPU')
tf.config.experimental.set_virtual_device_configuration(
physical_devices[args.gpu], [
tf.config.experimental.VirtualDeviceConfiguration(
memory_limit=1024 * 3)
])
if args.env == 200:
envname = 'ScratchItchPR2X'
elif args.env == 201:
envname = 'DressingPR2X'
elif args.env == 202:
envname = 'BedBathingPR2X'
logdir = f'MFBox_Assistive'
if args.SAC:
wandb.init(config=vars(args),
project="Assistive Gym",
name=f'SAC on {envname}')
elif args.PPO:
wandb.init(config=vars(args),
project="Assistive Gym",
name=f'PPO on {envname}')
elif args.TD3:
wandb.init(config=vars(args),
project="Assistive Gym",
name=f'TD3 on {envname}')
elif args.DEBUG:
logdir = f'DEBUG_Assistive'
wandb.init(config=vars(args),
project="Assistive Gym",
name=f'DEBUG on {envname}')
else:
print('PLEASE INDICATE THE ALGORITHM !!')
if not os.path.exists(logdir):
os.makedirs(logdir)
parser.set_defaults(logdir=logdir)
args = parser.parse_args()
env = gym.make(f'{envname}-v0')
#test_env = Monitor(env,logdir,force=True)
test_env = gym.make(f'{envname}-v0')
if args.SAC:
policy = SAC(state_shape=env.observation_space.shape,
action_dim=env.action_space.high.size,
gpu=args.gpu,
memory_capacity=args.memory_capacity,
max_action=env.action_space.high[0],
batch_size=args.batch_size,
n_warmup=args.n_warmup,
alpha=args.alpha,
auto_alpha=args.auto_alpha)
trainer = Trainer(policy, env, args, test_env=test_env)
elif args.PPO:
policy = PPO(state_shape=env.observation_space.shape,
action_dim=get_act_dim(env.action_space),
is_discrete=is_discrete(env.action_space),
max_action=None if is_discrete(env.action_space) else
env.action_space.high[0],
batch_size=args.batch_size,
actor_units=(64, 64),
critic_units=(64, 64),
n_epoch=10,
lr_actor=3e-4,
lr_critic=3e-4,
hidden_activation_actor="tanh",
hidden_activation_critic="tanh",
discount=0.99,
lam=0.95,
entropy_coef=0.,
horizon=args.horizon,
normalize_adv=args.normalize_adv,
enable_gae=args.enable_gae,
gpu=args.gpu)
trainer = OnPolicyTrainer(policy, env, args, test_env=test_env)
elif args.TD3:
policy = TD3(state_shape=env.observation_space.shape,
action_dim=env.action_space.high.size,
gpu=args.gpu,
memory_capacity=args.memory_capacity,
max_action=env.action_space.high[0],
batch_size=args.batch_size,
n_warmup=args.n_warmup)
trainer = Trainer(policy, env, args, test_env=test_env)
elif args.DEBUG:
policy = SAC(state_shape=env.observation_space.shape,
action_dim=env.action_space.high.size,
gpu=args.gpu,
memory_capacity=args.memory_capacity,
max_action=env.action_space.high[0],
batch_size=args.batch_size,
n_warmup=100,
alpha=args.alpha,
auto_alpha=args.auto_alpha)
parser.set_defaults(test_interval=200)
args = parser.parse_args()
trainer = Trainer(policy, env, args, test_env=None)
trainer()
import roboschool, gym
from tf2rl.algos.td3 import TD3
from tf2rl.trainer.trainer import Trainer
if __name__ == '__main__':
parser = Trainer.get_argument()
parser.add_argument('--env-name', type=str, default="RoboschoolAnt-v1")
args = parser.parse_args()
env = gym.make(args.env_name)
test_env = gym.make(args.env_name)
policy = TD3(state_shape=env.observation_space.shape,
action_dim=env.action_space.high.size,
gpu=args.gpu,
batch_size=100)
trainer = Trainer(policy, env, args, test_env=test_env)
trainer()