def get_experiment_environment(**args):
from misc_util import set_global_seeds
process_seed = args["seed"]
set_global_seeds(process_seed)
logger_context = logger.scoped_configure(
dir='tmp', format_strs=['stdout', 'log', 'csv'])
return logger_context
def get_experiment_environment(**args):
from utils import setup_mpi_gpus, setup_tensorflow_session
from baselines.common import set_global_seeds
from gym.utils.seeding import hash_seed
process_seed = args["seed"] + 1000 * MPI.COMM_WORLD.Get_rank()
process_seed = hash_seed(process_seed, max_bytes=4)
set_global_seeds(process_seed)
setup_mpi_gpus()
logger_context = logger.scoped_configure(dir=None,
format_strs=['stdout', 'log',
'csv'] if MPI.COMM_WORLD.Get_rank() == 0 else ['log'])
tf_context = setup_tensorflow_session()
return logger_context, tf_context
def get_experiment_environment(**args):
from utils import setup_mpi_gpus, setup_tensorflow_session
from baselines.common import set_global_seeds
from gym.utils.seeding import hash_seed
process_seed = args["seed"] + 1000 * MPI.COMM_WORLD.Get_rank()
process_seed = hash_seed(process_seed, max_bytes=4)
set_global_seeds(process_seed)
setup_mpi_gpus()
logdir = './' + args["logdir"] + '/' + datetime.datetime.now().strftime(
args["expID"] + "-openai-%Y-%m-%d-%H-%M-%S-%f")
logger_context = logger.scoped_configure(
dir=logdir,
format_strs=['stdout', 'log', 'csv', 'tensorboard']
if MPI.COMM_WORLD.Get_rank() == 0 else ['log'])
tf_context = setup_tensorflow_session()
return logger_context, tf_context, logdir
def __init__(self, env_fns, spaces=None):
"""
If you don't specify observation_space, we'll have to create a dummy
environment to get it.
"""
if spaces:
observation_space, action_space = spaces
else:
logger.log('Creating dummy env object to get spaces')
with logger.scoped_configure(format_strs=[]):
dummy = env_fns[0]()
observation_space, action_space = dummy.observation_space, dummy.action_space
dummy.close()
del dummy
VecEnv.__init__(self, len(env_fns), observation_space, action_space)
self.envs = [env_fn() for env_fn in env_fns]
def __init__(self, env_fns, spaces=None):
"""
If you don't specify observation_space, we'll have to create a dummy
environment to get it.
"""
#multiprocessing.set_start_method('spawn')
if spaces:
observation_space, action_space = spaces
else:
logger.log('Creating dummy env object to get spaces')
with logger.scoped_configure(format_strs=[]):
dummy = env_fns[0]()
observation_space, action_space = dummy.observation_space, dummy.action_space
dummy.close()
del dummy
VecEnv.__init__(self, len(env_fns), observation_space, action_space)
obs_spaces = observation_space.spaces if isinstance(
self.observation_space,
gym.spaces.Tuple) else (self.observation_space, )
self.obs_bufs = [
tuple(
Array(_NP_TO_CT[s.dtype.type], int(np.prod(s.shape)))
for s in obs_spaces) for _ in env_fns
]
self.obs_shapes = [s.shape for s in obs_spaces]
self.obs_dtypes = [s.dtype for s in obs_spaces]
self.parent_pipes = []
self.procs = []
for env_fn, obs_buf in zip(env_fns, self.obs_bufs):
wrapped_fn = CloudpickleWrapper(env_fn)
parent_pipe, child_pipe = Pipe()
proc = Process(target=_subproc_worker,
args=(child_pipe, parent_pipe, wrapped_fn, obs_buf,
self.obs_shapes))
proc.daemon = True
self.procs.append(proc)
self.parent_pipes.append(parent_pipe)
proc.start()
child_pipe.close()
self.waiting_step = False
def __init__(self, env_fns, spaces=None, context='spawn'):
"""
If you don't specify observation_space, we'll have to create a dummy
environment to get it.
"""
ctx = mp.get_context(context)
if spaces:
observation_space, action_space = spaces
else:
logger.log('Creating dummy env object to get spaces')
with logger.scoped_configure(format_strs=[]):
dummy = env_fns[0]()
observation_space, action_space = dummy.observation_space, dummy.action_space
dummy.close()
del dummy
VecEnv.__init__(self, len(env_fns), observation_space, action_space)
self.obs_keys, self.obs_shapes, self.obs_dtypes = obs_space_info(
observation_space)
self.obs_bufs = [{
k: ctx.Array(_NP_TO_CT[self.obs_dtypes[k].type],
int(np.prod(self.obs_shapes[k])))
for k in self.obs_keys
} for _ in env_fns]
self.parent_pipes = []
self.procs = []
with clear_mpi_env_vars():
for env_fn, obs_buf in zip(env_fns, self.obs_bufs):
wrapped_fn = CloudpickleWrapper(env_fn)
parent_pipe, child_pipe = ctx.Pipe()
proc = ctx.Process(target=_subproc_worker,
args=(child_pipe, parent_pipe, wrapped_fn,
obs_buf, self.obs_shapes,
self.obs_dtypes, self.obs_keys))
proc.daemon = True
self.procs.append(proc)
self.parent_pipes.append(parent_pipe)
proc.start()
child_pipe.close()
self.waiting_step = False
self.viewer = None
def test_mpi_weighted_mean():
from mpi4py import MPI
comm = MPI.COMM_WORLD
with logger.scoped_configure(comm=comm):
if comm.rank == 0:
name2valcount = {'a': (10, 2), 'b': (20, 3)}
elif comm.rank == 1:
name2valcount = {'a': (19, 1), 'c': (42, 3)}
else:
raise NotImplementedError
d = mpi_util.mpi_weighted_mean(comm, name2valcount)
correctval = {'a': (10 * 2 + 19) / 3.0, 'b': 20, 'c': 42}
if comm.rank == 0:
assert d == correctval, f'{d} != {correctval}'
for name, (val, count) in name2valcount.items():
for _ in range(count):
logger.logkv_mean(name, val)
d2 = logger.dumpkvs()
if comm.rank == 0:
assert d2 == correctval
