def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str,
run_script=run_script,
start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
args = default_argument_parser().parse_args(args=[])
args = config2args(myargs.config.args, args)
args.opts += ['OUTPUT_DIR', args1.outdir + '/detectron2']
print("Command Line Args:", args)
myargs = D2Utils.unset_myargs_for_multiple_processing(
myargs, num_gpus=args.num_gpus)
launch(
main,
args.num_gpus,
num_machines=args.num_machines,
machine_rank=args.machine_rank,
dist_url=args.dist_url,
args=(args, myargs),
)
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str,
run_script=run_script,
start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
if hasattr(myargs.config, 'datasets'):
prepare_dataset(myargs.config.datasets, cfg=myargs.config)
parser = build_parser()
args = parser.parse_args([])
args = config2args(myargs.config.args, args)
outdir = os.path.join(myargs.args.outdir, 'NAO')
os.makedirs(outdir, exist_ok=True)
args.output_dir = outdir
log_format = '%(asctime)s %(message)s'
logging.basicConfig(stream=sys.stdout,
level=logging.INFO,
format=log_format,
datefmt='%m/%d %I:%M:%S %p')
main(args, myargs)
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str,
run_script=run_script,
start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
if hasattr(myargs.config, 'datasets'):
prepare_dataset(myargs.config.datasets, cfg=myargs.config)
helper = Helper(save_dir=os.path.join(myargs.args.outdir, 'robust_darts'),
args_list=[])
args = helper.config
args = config2args(myargs.config.args, args)
if args.drop_path_prob == 0:
args.cutout = False
log_format = '%(asctime)s %(message)s'
# logging.basicConfig()
logging.basicConfig(level=logging.INFO,
format=log_format,
datefmt='%m/%d %I:%M:%S %p')
logging.getLogger().handlers = myargs.logger.handlers
# for handler in myargs.logger.handlers:
# logging.getLogger().addHandler(handler)
space = spaces_dict[args.space]
main(space, args=args, helper=helper, myargs=myargs)
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str,
run_script=run_script,
start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
args = config2args(myargs.config, args1)
build_start(cfg=args, myargs=myargs)
pass
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str, run_script=run_script, start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
if hasattr(myargs.config, 'datasets'):
prepare_dataset(myargs.config.datasets, cfg=myargs.config)
args = parser.parse_args([])
args = config2args(myargs.config.args, args)
args.path = os.path.join(myargs.args.outdir, args.path)
main(args, myargs)
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str, run_script=run_script, start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
if hasattr(myargs.config, 'datasets'):
prepare_dataset(myargs.config.datasets, cfg=myargs.config)
opt = TestOptions().parse(print_options=False)
opt = config2args(myargs.config.args, opt)
main(opt, myargs)
pass
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
args1, myargs, _ = parse_args_and_setup_myargs(argv_str, start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
args = parse_args()
args = config2args(myargs.config, args)
args.datadir = os.path.expanduser(args.datadir)
args.outdir = os.path.expanduser(args.outdir)
if args.dataset == "cityscapes_instance_only":
convert_cityscapes_instance_only(args.datadir, args.outdir)
elif args.dataset == "cocostuff":
convert_coco_stuff_mat(args.datadir, args.outdir)
else:
print("Dataset not supported: %s" % args.dataset)
def run1(argv_str=None): from template_lib.utils.config import parse_args_and_setup_myargs, config2args from template_lib.utils.modelarts_utils import prepare_dataset run_script = os.path.relpath(__file__, os.getcwd()) args1, myargs, _ = parse_args_and_setup_myargs(argv_str, run_script=run_script, start_tb=False) myargs.args = args1 myargs.config = getattr(myargs.config, args1.command) # prepare_dataset(myargs.config.dataset) parser = prepare_parser() args = parser.parse_args([]) args = config2args(myargs.config, args) args.data_root = os.path.expanduser(args.data_root) print(args) config = vars(args) run(config, stdout=myargs.stdout)
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str,
run_script=run_script,
start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
if hasattr(myargs.config, 'datasets'):
prepare_dataset(myargs.config.datasets, cfg=myargs.config)
parser = build_parser()
args = parser.parse_args()
if args.rand_seed is None or args.rand_seed < 0:
args.rand_seed = random.randint(1, 100000)
args = config2args(config=myargs.config.args, args=args)
args.save_dir = os.path.join(args1.outdir, '201')
main(args, myargs)
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str,
run_script=run_script,
start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
if hasattr(myargs.config, 'datasets'):
prepare_dataset(myargs.config.datasets, cfg=myargs.config)
parser = argparse.ArgumentParser()
parser.add_argument('--out', type=str)
parser.add_argument('--n_worker', type=int, default=8)
parser.add_argument('--path', type=str)
args = parser.parse_args([])
args = config2args(myargs.config.args, args)
main(args, myargs)
def main(myargs):
parser = argparse.ArgumentParser(description="PyTorch Object Detection Training")
parser.add_argument(
"--config-file",
default="",
metavar="FILE",
help="path to config file",
type=str,
)
parser.add_argument("--local_rank", type=int, default=0)
parser.add_argument(
"--skip-test",
dest="skip_test",
help="Do not test the final model",
action="store_true",
)
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_args([])
args = config2args(myargs.config, args)
cfg.OUTPUT_DIR = os.path.join(myargs.args.outdir, 'maskrcnn_benchmark')
num_gpus = int(os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1
args.distributed = num_gpus > 1
if args.distributed:
myargs = setup_myargs_for_multiple_processing(myargs)
torch.cuda.set_device(args.local_rank)
torch.distributed.init_process_group(
backend="nccl", init_method="env://"
)
synchronize()
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
if 'opts_private' in args:
cfg.merge_from_list(args.opts_private)
cfg.freeze()
output_dir = cfg.OUTPUT_DIR
if output_dir:
mkdir(output_dir)
logger = setup_logger("maskrcnn_benchmark", output_dir, get_rank())
logger.info("Using {} GPUs".format(num_gpus))
logger.info(args)
if comm.is_main_process():
# Note: some of our scripts may expect the existence of
# config.yaml in output directory
path = os.path.join(output_dir, "config.yaml")
with open(path, "w") as f:
f.write(cfg.dump())
logger.info("Full config saved to {}".format(os.path.abspath(path)))
logger.info("Collecting env info (might take some time)")
logger.info("\n" + collect_env_info())
logger.info("Loaded configuration file {}".format(args.config_file))
with open(args.config_file, "r") as cf:
config_str = "\n" + cf.read()
logger.info(config_str)
logger.info("Running with config:\n{}".format(cfg))
model = train(cfg, args.local_rank, args.distributed, myargs=myargs)
if not args.skip_test:
test(cfg, model, args.distributed)
modelarts_sync_results(args=myargs.args, myargs=myargs, join=True, end=True)
def run(argv_str=None):
from template_lib.utils.config import parse_args_and_setup_myargs, config2args
from template_lib.utils.modelarts_utils import prepare_dataset
run_script = os.path.relpath(__file__, os.getcwd())
args1, myargs, _ = parse_args_and_setup_myargs(argv_str,
run_script=run_script,
start_tb=False)
myargs.args = args1
myargs.config = getattr(myargs.config, args1.command)
if hasattr(myargs.config, 'datasets'):
prepare_dataset(myargs.config.datasets, cfg=myargs.config)
parser = argparse.ArgumentParser(description='Progressive Growing of GANs')
parser.add_argument('--path', type=str, help='path of specified dataset')
parser.add_argument(
'--phase',
type=int,
default=600_000,
help='number of samples used for each training phases',
)
parser.add_argument('--lr',
default=0.001,
type=float,
help='learning rate')
parser.add_argument('--sched',
action='store_true',
help='use lr scheduling')
parser.add_argument('--init_size',
default=8,
type=int,
help='initial image size')
parser.add_argument('--max_size',
default=1024,
type=int,
help='max image size')
parser.add_argument('--ckpt',
default=None,
type=str,
help='load from previous checkpoints')
parser.add_argument(
'--no_from_rgb_activate',
action='store_true',
help='use activate in from_rgb (original implementation)',
)
parser.add_argument('--mixing',
action='store_true',
help='use mixing regularization')
parser.add_argument(
'--loss',
type=str,
default='wgan-gp',
choices=['wgan-gp', 'r1'],
help='class of gan loss',
)
args = parser.parse_args([])
args = config2args(myargs.config.args, args)
main(args, myargs)
def main(myargs):
parser = argparse.ArgumentParser(
description="PyTorch Object Detection Inference")
parser.add_argument(
"--config-file",
default=
"/private/home/fmassa/github/detectron.pytorch_v2/configs/e2e_faster_rcnn_R_50_C4_1x_caffe2.yaml",
metavar="FILE",
help="path to config file",
)
parser.add_argument("--local_rank", type=int, default=0)
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_args([])
args = config2args(myargs.config, args)
num_gpus = int(
os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1
distributed = num_gpus > 1
if distributed:
torch.cuda.set_device(args.local_rank)
torch.distributed.init_process_group(backend="nccl",
init_method="env://")
synchronize()
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.OUTPUT_DIR = myargs.args.outdir
cfg.freeze()
output_dir = cfg.OUTPUT_DIR
if output_dir:
mkdir(output_dir)
save_dir = output_dir
logger = setup_logger("maskrcnn_benchmark", save_dir, get_rank())
logger.info("Using {} GPUs".format(num_gpus))
logger.info(cfg)
logger.info("Collecting env info (might take some time)")
logger.info("\n" + collect_env_info())
model = build_detection_model(cfg)
model.to(cfg.MODEL.DEVICE)
checkpointer = DetectronCheckpointer(cfg, model, save_dir=output_dir)
_ = checkpointer.load(cfg.MODEL.WEIGHT)
iou_types = ("bbox", )
if cfg.MODEL.MASK_ON:
iou_types = iou_types + ("segm", )
if cfg.MODEL.KEYPOINT_ON:
iou_types = iou_types + ("keypoints", )
output_folders = [None] * len(cfg.DATASETS.TEST)
dataset_names = cfg.DATASETS.TEST
if cfg.OUTPUT_DIR:
for idx, dataset_name in enumerate(dataset_names):
output_folder = os.path.join(cfg.OUTPUT_DIR, "inference",
dataset_name)
mkdir(output_folder)
output_folders[idx] = output_folder
data_loaders_val = make_data_loader(cfg,
is_train=False,
is_distributed=distributed)
for output_folder, dataset_name, data_loader_val in zip(
output_folders, dataset_names, data_loaders_val):
inference(
model,
data_loader_val,
dataset_name=dataset_name,
iou_types=iou_types,
box_only=False if cfg.MODEL.RETINANET_ON else cfg.MODEL.RPN_ONLY,
device=cfg.MODEL.DEVICE,
expected_results=cfg.TEST.EXPECTED_RESULTS,
expected_results_sigma_tol=cfg.TEST.EXPECTED_RESULTS_SIGMA_TOL,
output_folder=output_folder,
)
synchronize()