def main():
"""Run PPO until the environment throws an exception."""
config = tf.ConfigProto()
#os.environ["CUDA_VISIBLE_DEVICES"]="-1"
config.gpu_options.allow_growth = True # pylint: disable=E1101
with tf.Session(config=config):
# Take more timesteps than we need to be sure that
# we stop due to an exception.
ppo2ttifrutti.learn(policy=policies.CnnPolicy,
env=SubprocVecEnv([
env.make_train_5, env.make_train_18,
env.make_train_43
]),
nsteps=2048,
nminibatches=16,
lam=0.95,
gamma=0.99,
noptepochs=4,
log_interval=1,
ent_coef=0.01,
lr=lambda _: 2e-4,
cliprange=lambda _: 0.1,
total_timesteps=int(1e9),
save_interval=25)
def main():
"""Run PPO until the environment throws an exception."""
config = tf.ConfigProto()
#os.environ["CUDA_VISIBLE_DEVICES"]="-1"
config.gpu_options.allow_growth = True # pylint: disable=E1101
# Creating directory if necessary
train_path = args.log_path
if not os.path.isdir(train_path):
os.mkdir(train_path)
if not os.path.isdir(train_path + 'log/'):
os.mkdir(train_path + 'log/')
if not os.path.isdir(train_path + 'model/'):
os.mkdir(train_path + 'model/')
if not os.path.isdir(train_path + 'log/'):
os.mkdir(train_path + 'log/')
if not os.path.isdir(train_path + 'model/'):
os.mkdir(train_path + 'model/')
with tf.Session(config=config):
# Take more timesteps than we need to be sure that
# we stop due to an exception.
ppo2ttifrutti.learn(
policy=policies.CnnPolicy,
env=SubprocVecEnv([
env.make_train_0, env.make_train_1, env.make_train_2,
env.make_train_3, env.make_train_4, env.make_train_5,
env.make_train_6, env.make_train_7, env.make_train_8,
env.make_train_9, env.make_train_10, env.make_train_11,
env.make_train_12, env.make_train_13, env.make_train_14,
env.make_train_15, env.make_train_16, env.make_train_17,
env.make_train_18, env.make_train_19, env.make_train_20,
env.make_train_21, env.make_train_22, env.make_train_23,
env.make_train_24, env.make_train_25, env.make_train_26,
env.make_train_27, env.make_train_28, env.make_train_29,
env.make_train_30, env.make_train_31, env.make_train_32,
env.make_train_33, env.make_train_34, env.make_train_35,
env.make_train_36, env.make_train_37, env.make_train_38,
env.make_train_39, env.make_train_40, env.make_train_41,
env.make_train_42, env.make_train_43, env.make_train_44,
env.make_train_45, env.make_train_46
]),
# env=SubprocVecEnv([env.make_train_0, env.make_train_1, env.make_train_2, env.make_train_3, env.make_train_4, env.make_train_5, env.make_train_6]),
nsteps=args.nsteps,
nminibatches=args.nminibatches,
lam=0.95,
gamma=0.99,
noptepochs=4,
log_interval=1,
ent_coef=0.01,
lr=lambda _: args.lr,
cliprange=lambda _: 0.1,
total_timesteps=args.total_timesteps,
save_interval=25,
log_path=args.log_path,
train=args.train)
def main():
"""Run PPO until the environment throws an exception."""
config = tf.ConfigProto()
#os.environ["CUDA_VISIBLE_DEVICES"]="-1"
config.gpu_options.allow_growth = True # pylint: disable=E1101
with tf.Session(config=config):
# Take more timesteps than we need to be sure that
# we stop due to an exception.
ppo2ttifrutti.learn(
policy=policies.CnnPolicy,
env=SubprocVecEnv([
env.make_train_0, env.make_train_1, env.make_train_2,
env.make_train_3, env.make_train_4, env.make_train_5,
env.make_train_6, env.make_train_7, env.make_train_8,
env.make_train_9, env.make_train_10, env.make_train_11,
env.make_train_12, env.make_train_13, env.make_train_14,
env.make_train_15, env.make_train_16, env.make_train_17,
env.make_train_18, env.make_train_19, env.make_train_20,
env.make_train_21, env.make_train_22, env.make_train_23,
env.make_train_24, env.make_train_25, env.make_train_26,
env.make_train_27, env.make_train_28, env.make_train_29,
env.make_train_30, env.make_train_31, env.make_train_32,
env.make_train_33, env.make_train_34, env.make_train_35,
env.make_train_36, env.make_train_37, env.make_train_38,
env.make_train_39, env.make_train_40, env.make_train_41,
env.make_train_42, env.make_train_43, env.make_train_44,
env.make_train_45, env.make_train_46, env.make_val_0,
env.make_val_1, env.make_val_2, env.make_val_3, env.make_val_4,
env.make_val_5, env.make_val_6, env.make_val_7, env.make_val_8,
env.make_val_9, env.make_val_10, env.make_extra_0,
env.make_extra_1, env.make_extra_2, env.make_extra_3,
env.make_extra_4, env.make_extra_5, env.make_extra_6,
env.make_extra_7, env.make_extra_8, env.make_extra_9,
env.make_extra_10, env.make_extra_11, env.make_extra_12,
env.make_extra_13, env.make_extra_14, env.make_extra_15,
env.make_extra_16, env.make_extra_17, env.make_extra_18,
env.make_extra_19, env.make_extra_20, env.make_extra_21,
env.make_extra_22, env.make_extra_23, env.make_extra_24,
env.make_extra_25, env.make_extra_26, env.make_extra_27,
env.make_extra_28, env.make_extra_29, env.make_extra_30,
env.make_extra_31, env.make_extra_32, env.make_extra_33,
env.make_extra_34, env.make_extra_35, env.make_extra_36,
env.make_extra_37, env.make_extra_38, env.make_extra_39
]),
nsteps=2048,
nminibatches=16,
lam=0.95,
gamma=0.99,
noptepochs=4,
log_interval=1,
ent_coef=0.01,
lr=lambda _: 2e-4,
cliprange=lambda _: 0.1,
total_timesteps=int(1e9),
save_interval=25)
def main():
"""Run PPO until the environment throws an exception."""
config = tf.ConfigProto()
#os.environ["CUDA_VISIBLE_DEVICES"]="-1"
config.gpu_options.allow_growth = True # pylint: disable=E1101
lis = [
env.make_val_0, env.make_val_1, env.make_val_2, env.make_val_3,
env.make_val_4, env.make_val_5, env.make_val_6, env.make_val_7,
env.make_val_8, env.make_val_9, env.make_val_10
]
i = args.val_idx
# Creating directory if necessary
val_path = args.log_path + 'val_' + str(i) + '/'
if not os.path.isdir(val_path):
os.mkdir(val_path)
if not os.path.isdir(val_path + 'log/'):
os.mkdir(val_path + 'log/')
if not os.path.isdir(val_path + 'model/'):
os.mkdir(val_path + 'model/')
if not os.path.isdir(val_path + 'log/'):
os.mkdir(val_path + 'log/')
if not os.path.isdir(val_path + 'model/'):
os.mkdir(val_path + 'model/')
# Create session
with tf.Session(config=config):
# Take more timesteps than we need to be sure that
# we stop due to an exception.
ppo2ttifrutti.learn(policy=policies.CnnPolicy,
env=DummyVecEnv([lis[i]]),
nsteps=args.nsteps,
nminibatches=args.nminibatches,
lam=0.95,
gamma=0.99,
noptepochs=4,
log_interval=1,
ent_coef=0.01,
lr=lambda _: args.lr,
cliprange=lambda _: 0.1,
total_timesteps=args.total_timesteps,
load_path=tf.train.latest_checkpoint(
args.load_path)[:-3],
log_path=val_path)
def main():
"""Run PPO until the environment throws an exception."""
config = tf.ConfigProto()
#os.environ["CUDA_VISIBLE_DEVICES"]="-1"
config.gpu_options.allow_growth = True # pylint: disable=E1101
with tf.Session(config=config):
# Take more timesteps than we need to be sure that
# we stop due to an exception.
ppo2ttifrutti.learn(policy=policies.CnnPolicy,
env=DummyVecEnv([env.make_custom]),
nsteps=4096,
nminibatches=4,
lam=0.95,
gamma=0.99,
noptepochs=4,
log_interval=1,
ent_coef=0.01,
lr=lambda _: 7.5e-5,
cliprange=lambda _: 0.1,
total_timesteps=int(1e7),
load_path='model')