def main():
if not torch.cuda.is_available():
logging.info('no gpu device available')
sys.exit(1)
if args.auxiliary and args.net_type == 'macro':
logging.info('auxiliary head classifier not supported for macro search space models')
sys.exit(1)
logging.info("args = %s", args)
cudnn.enabled = True
cudnn.benchmark = True
np.random.seed(args.seed)
torch.manual_seed(args.seed)
torch.cuda.manual_seed(args.seed)
# Data
_, valid_transform = utils._data_transforms_cifar10(args)
valid_data = torchvision.datasets.CIFAR10(root=args.data, train=False, download=True, transform=valid_transform)
valid_queue = torch.utils.data.DataLoader(
valid_data, batch_size=args.batch_size, shuffle=False, pin_memory=True, num_workers=1)
# classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')
# Model
if args.net_type == 'micro':
logging.info("==> Building micro search space encoded architectures")
genotype = eval("genotypes.%s" % args.arch)
net = PyrmNASNet(args.init_channels, num_classes=10, layers=args.layers,
auxiliary=args.auxiliary, genotype=genotype,
increment=args.filter_increment, SE=args.SE)
elif args.net_type == 'macro':
genome = eval("macro_genotypes.%s" % args.arch)
channels = [(3, 128), (128, 128), (128, 128)]
net = EvoNetwork(genome, channels, 10, (32, 32), decoder='dense')
else:
raise NameError('Unknown network type, please only use supported network type')
# logging.info("{}".format(net))
logging.info("param size = %fMB", utils.count_parameters_in_MB(net))
net = net.to(device)
# no drop path during inference
net.droprate = 0.0
utils.load(net, args.model_path)
criterion = nn.CrossEntropyLoss()
criterion.to(device)
# inference on original CIFAR-10 test images
infer(valid_queue, net, criterion)
def main():
args = parser.parse_args()
spec = cluster_spec(args.n_tasks, 1)
cluster = tf.train.ClusterSpec(spec).as_cluster_def()
cls = load("agents.actorcritic.a3c_worker:" + args.cls)
config = json_to_dict(args.config)
task = cls(args.env_id, args.task_id, cluster, args.monitor_path, config, video=args.video)
task.learn()
def main():
comm = MPI.Comm.Get_parent()
task_id = comm.Get_rank()
args = parser.parse_args()
cls = load("agents.ppo.dppo_worker:" + args.cls)
config = json_to_dict(args.config)
task = cls(args.env_id, task_id, comm, args.monitor_path, config,
args.seed)
task.run()
def make_agent(name: str, state_dimensions: str, action_space: str, RNN: bool = False, **args):
"""Make an agent of a given name, possibly using extra arguments."""
try:
Agent = agent_registry[name]
Agent = next((agent_type for agent_type in Agent if
agent_type["action_space"] == action_space and
agent_type["state_dimensions"] == state_dimensions and
agent_type["RNN"] == RNN))
Agent = Agent["entry_point"]
if not callable(Agent):
Agent = load(Agent)
except (KeyError, StopIteration):
raise ClassNotRegisteredError(
"The agent {} for state dimensionality {}, action space {} and RNN={} is not registered.".format(
name,
state_dimensions,
action_space,
RNN))
return Agent(**args)
