def main(_):
# Parses whether to run a single bsuite_id, or multiprocess sweep.
bsuite_id = FLAGS.bsuite_id
if bsuite_id in sweep.SWEEP:
print(f'Running single experiment: bsuite_id={bsuite_id}.')
run(bsuite_id)
elif hasattr(sweep, bsuite_id):
bsuite_sweep = getattr(sweep, bsuite_id)
print(f'Running sweep over bsuite_id in sweep.{bsuite_sweep}')
FLAGS.verbose = False
pool.map_mpi(run, bsuite_sweep)
else:
raise ValueError(f'Invalid flag: bsuite_id={bsuite_id}.')
def main(_):
"""Parses whether to run a single bsuite_id, or multiprocess sweep."""
bsuite_id = FLAGS.bsuite_id
if bsuite_id in sweep.SWEEP:
print('Running a single bsuite_id={}'.format(bsuite_id))
run(bsuite_id)
elif hasattr(sweep, bsuite_id):
bsuite_sweep = getattr(sweep, bsuite_id)
print(
'Running a sweep over bsuite_id in sweep.{}'.format(bsuite_sweep))
FLAGS.verbose = False
pool.map_mpi(run, bsuite_sweep)
else:
raise ValueError('Invalid flag bsuite_id={}'.format(bsuite_id))
'qnet_settings': qnet_settings,
'start_optimization': 64,
'update_qnet_every': 2,
'update_target_every': 50,
'ddqn': True,
'n_steps': 4,
'duelling_dqn': True,
'prioritized_buffer': True,
'alpha': 0.6,
'beta0': 0.4,
'beta_increment': 1e-6
}
agent = Agent(action_spec=env.action_spec(),
observation_spec=env.observation_spec(),
device=device,
settings=settings)
experiment.run(agent=agent,
environment=env,
num_episodes=env.bsuite_num_episodes,
verbose=False)
return bsuite_id
bsuite_sweep = getattr(sweep, 'CARTPOLE_SCALE')
pool.map_mpi(run, bsuite_sweep, 6)
#bsuite_sweep = getattr(sweep, 'SWEEP')
#pool.map_mpi(run, bsuite_sweep, 6)
