def sample_paths(policy_params,
max_samples,
max_path_length=np.inf,
env_params=None,
scope=None):
"""
:param policy_params: parameters for the policy. This will be updated on
each worker process
:param max_samples: desired maximum number of samples to be collected. The
actual number of collected samples might be greater since all trajectories
will be rolled out either until termination or until max_path_length is
reached
:param max_path_length: horizon / maximum length of a single trajectory
:return: a list of collected paths
"""
singleton_pool.run_each(_worker_set_policy_params,
[(policy_params, scope)] *
singleton_pool.n_parallel)
if env_params is not None:
singleton_pool.run_each(_worker_set_env_params, [(env_params, scope)] *
singleton_pool.n_parallel)
return singleton_pool.run_collect(_worker_collect_one_path,
threshold=max_samples,
args=(max_path_length, scope),
show_prog_bar=True)
def sample_paths(self,
policy_params,
max_samples,
max_path_length,
env_params=None,
inference_params=None,
scope=None):
singleton_pool.run_each(
parallel_sampler._worker_set_policy_params,
[(policy_params, scope)] * singleton_pool.n_parallel,
)
singleton_pool.run_each(
_worker_set_inference_params,
[(inference_params, scope)] * singleton_pool.n_parallel,
)
# if env_params:
# singleton_pool.run_each(
# parallel_sampler._worker_set_env_params,
# [(env_params, scope)] * singleton_pool.n_parallel,
# )
return singleton_pool.run_collect(
_worker_collect_one_path,
threshold=max_samples,
args=(max_path_length, scope),
show_prog_bar=True,
)
def sample_paths(policy_params,
max_samples,
max_path_length=np.inf,
env_params=None,
scope=None):
"""Sample paths from each worker.
Parameters
----------
policy_params :
parameters for the policy. This will be updated on each worker process
max_samples : int
desired maximum number of samples to be collected. The
actual number of collected samples might be greater since all trajectories
will be rolled out either until termination or until max_path_length is
reached
max_path_length : int, optional
horizon / maximum length of a single trajectory
scope : str
Scope for identifying the algorithm.
Must be specified if running multiple algorithms
simultaneously, each using different environments
and policies.
"""
singleton_pool.run_each(_worker_set_policy_params,
[(policy_params, scope)] *
singleton_pool.n_parallel)
if env_params is not None:
singleton_pool.run_each(_worker_set_env_params, [(env_params, scope)] *
singleton_pool.n_parallel)
return singleton_pool.run_collect(_worker_collect_one_path,
threshold=max_samples,
args=(max_path_length, scope),
show_prog_bar=True)