def main(args, unknown_args):
args, config = parse_args_uargs(args, unknown_args)
set_global_seed(args.seed)
prepare_cudnn(args.deterministic, args.benchmark)
if args.logdir is not None:
os.makedirs(args.logdir, exist_ok=True)
dump_environment(config, args.logdir, args.configs)
if args.expdir is not None:
module = import_module(expdir=args.expdir) # noqa: F841
if args.logdir is not None:
dump_code(args.expdir, args.logdir)
env = ENVIRONMENTS.get_from_params(**config["environment"])
algorithm_name = config["algorithm"].pop("algorithm")
if algorithm_name in OFFPOLICY_ALGORITHMS_NAMES:
ALGORITHMS = OFFPOLICY_ALGORITHMS
trainer_fn = OffpolicyTrainer
sync_epoch = False
elif algorithm_name in ONPOLICY_ALGORITHMS_NAMES:
ALGORITHMS = ONPOLICY_ALGORITHMS
trainer_fn = OnpolicyTrainer
sync_epoch = True
else:
# @TODO: add registry for algorithms, trainers, samplers
raise NotImplementedError()
db_server = DATABASES.get_from_params(
**config.get("db", {}), sync_epoch=sync_epoch
)
algorithm_fn = ALGORITHMS.get(algorithm_name)
algorithm = algorithm_fn.prepare_for_trainer(env_spec=env, config=config)
if args.resume is not None:
checkpoint = utils.load_checkpoint(filepath=args.resume)
checkpoint = utils.any2device(checkpoint, utils.get_device())
algorithm.unpack_checkpoint(
checkpoint=checkpoint,
with_optimizer=False
)
monitoring_params = config.get("monitoring_params", None)
trainer = trainer_fn(
algorithm=algorithm,
env_spec=env,
db_server=db_server,
logdir=args.logdir,
monitoring_params=monitoring_params,
**config["trainer"],
)
trainer.run()
def main(args, unknown_args):
args, config = parse_args_uargs(args, unknown_args)
set_global_seed(args.seed)
if args.logdir is not None:
os.makedirs(args.logdir, exist_ok=True)
dump_config(config, args.logdir, args.configs)
if args.expdir is not None:
module = import_module(expdir=args.expdir) # noqa: F841
env = ENVIRONMENTS.get_from_params(**config["environment"])
algorithm_name = config["algorithm"].pop("algorithm")
if algorithm_name in OFFPOLICY_ALGORITHMS_NAMES:
ALGORITHMS = OFFPOLICY_ALGORITHMS
trainer_fn = OffpolicyTrainer
sync_epoch = False
weights_sync_mode = "critic" if env.discrete_actions else "actor"
elif algorithm_name in ONPOLICY_ALGORITHMS_NAMES:
ALGORITHMS = ONPOLICY_ALGORITHMS
trainer_fn = OnpolicyTrainer
sync_epoch = True
weights_sync_mode = "actor"
else:
# @TODO: add registry for algorithms, trainers, samplers
raise NotImplementedError()
db_server = DATABASES.get_from_params(
**config.get("db", {}), sync_epoch=sync_epoch
)
algorithm_fn = ALGORITHMS.get(algorithm_name)
algorithm = algorithm_fn.prepare_for_trainer(env_spec=env, config=config)
if args.resume is not None:
algorithm.load_checkpoint(filepath=args.resume)
trainer = trainer_fn(
algorithm=algorithm,
env_spec=env,
db_server=db_server,
logdir=args.logdir,
weights_sync_mode=weights_sync_mode,
**config["trainer"],
)
trainer.run()
def main(args, unknown_args):
args, config = parse_args_uargs(args, unknown_args)
set_global_seed(args.seed)
if args.logdir is not None:
os.makedirs(args.logdir, exist_ok=True)
dump_config(args.configs, args.logdir)
if args.expdir is not None:
module = import_module(expdir=args.expdir) # noqa: F841
algorithm_name = config["algorithm"].pop("algorithm")
if algorithm_name in OFFPOLICY_ALGORITHMS_NAMES:
ALGORITHMS = OFFPOLICY_ALGORITHMS
trainer_fn = OffpolicyTrainer
sync_epoch = False
else:
ALGORITHMS = ONPOLICY_ALGORITHMS
trainer_fn = OnpolicyTrainer
sync_epoch = True
db_server = DATABASES.get_from_params(**config.get("db", {}),
sync_epoch=sync_epoch)
env = ENVIRONMENTS.get_from_params(**config["environment"])
algorithm_fn = ALGORITHMS.get(algorithm_name)
algorithm = algorithm_fn.prepare_for_trainer(env_spec=env, config=config)
if args.resume is not None:
algorithm.load_checkpoint(filepath=args.resume)
trainer = trainer_fn(
algorithm=algorithm,
env_spec=env,
db_server=db_server,
**config["trainer"],
logdir=args.logdir,
)
def on_exit():
for p in trainer.get_processes():
p.terminate()
atexit.register(on_exit)
trainer.run()