def run_test(cfg, model, distributed):
if distributed:
model = model.module
torch.cuda.empty_cache() # TODO check if it helps
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,
bbox_aug=cfg.TEST.BBOX_AUG.ENABLED,
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()
def cache_url(url, model_dir=None, progress=True):
r"""Loads the Torch serialized object at the given URL.
If the object is already present in `model_dir`, it's deserialized and
returned. The filename part of the URL should follow the naming convention
``filename-<sha256>.ext`` where ``<sha256>`` is the first eight or more
digits of the SHA256 hash of the contents of the file. The hash is used to
ensure unique names and to verify the contents of the file.
The default value of `model_dir` is ``$TORCH_HOME/models`` where
``$TORCH_HOME`` defaults to ``~/.torch``. The default directory can be
overridden with the ``$TORCH_MODEL_ZOO`` environment variable.
Args:
url (string): URL of the object to download
model_dir (string, optional): directory in which to save the object
progress (bool, optional): whether or not to display a progress bar to stderr
Example:
>>> cached_file = asynet_mask_rcnn.utils.model_zoo.cache_url('https://s3.amazonaws.com/pytorch/models/resnet18-5c106cde.pth')
"""
if model_dir is None:
torch_home = os.path.expanduser(os.getenv("TORCH_HOME", "~/.torch"))
model_dir = os.getenv("TORCH_MODEL_ZOO",
os.path.join(torch_home, "models"))
if not os.path.exists(model_dir):
os.makedirs(model_dir)
parts = urlparse(url)
filename = os.path.basename(parts.path)
if filename == "model_final.pkl":
# workaround as pre-trained Caffe2 models from Detectron have all the same filename
# so make the full path the filename by replacing / with _
filename = parts.path.replace("/", "_")
cached_file = os.path.join(model_dir, filename)
if not os.path.exists(cached_file) and is_main_process():
sys.stderr.write('Downloading: "{}" to {}\n'.format(url, cached_file))
hash_prefix = HASH_REGEX.search(filename)
if hash_prefix is not None:
hash_prefix = hash_prefix.group(1)
# workaround: Caffe2 models don't have a hash, but follow the R-50 convention,
# which matches the hash PyTorch uses. So we skip the hash matching
# if the hash_prefix is less than 6 characters
if len(hash_prefix) < 6:
hash_prefix = None
_download_url_to_file(url, cached_file, hash_prefix, progress=progress)
synchronize()
return cached_file
def do_train(
cfg,
model,
data_loader,
data_loader_val,
optimizer,
scheduler,
checkpointer,
device,
checkpoint_period,
test_period,
arguments,
):
logger = logging.getLogger("asynet_mask_rcnn.trainer")
logger.info("Start training")
meters = MetricLogger(delimiter=" ")
max_iter = len(data_loader)
start_iter = arguments["iteration"]
model.train()
start_training_time = time.time()
end = time.time()
iou_types = ("bbox", )
if cfg.MODEL.MASK_ON:
iou_types = iou_types + ("segm", )
if cfg.MODEL.KEYPOINT_ON:
iou_types = iou_types + ("keypoints", )
dataset_names = cfg.DATASETS.TEST
for iteration, (images, targets, _) in enumerate(data_loader, start_iter):
if any(len(target) < 1 for target in targets):
logger.error(
f"Iteration={iteration + 1} || Image Ids used for training {_} || targets Length={[len(target) for target in targets]}"
)
continue
data_time = time.time() - end
iteration = iteration + 1
arguments["iteration"] = iteration
images = images.to(device)
targets = [target.to(device) for target in targets]
loss_dict = model(images, targets)
losses = sum(loss for loss in loss_dict.values())
# reduce losses over all GPUs for logging purposes
loss_dict_reduced = reduce_loss_dict(loss_dict)
losses_reduced = sum(loss for loss in loss_dict_reduced.values())
meters.update(loss=losses_reduced, **loss_dict_reduced)
optimizer.zero_grad()
# Note: If mixed precision is not used, this ends up doing nothing
# Otherwise apply loss scaling for mixed-precision recipe
with amp.scale_loss(losses, optimizer) as scaled_losses:
scaled_losses.backward()
optimizer.step()
scheduler.step()
batch_time = time.time() - end
end = time.time()
meters.update(time=batch_time, data=data_time)
eta_seconds = meters.time.global_avg * (max_iter - iteration)
eta_string = str(datetime.timedelta(seconds=int(eta_seconds)))
if iteration % 20 == 0 or iteration == max_iter:
logger.info(
meters.delimiter.join([
"eta: {eta}",
"iter: {iter}",
"{meters}",
"lr: {lr:.6f}",
"max mem: {memory:.0f}",
]).format(
eta=eta_string,
iter=iteration,
meters=str(meters),
lr=optimizer.param_groups[0]["lr"],
memory=torch.cuda.max_memory_allocated() / 1024.0 / 1024.0,
))
if iteration % checkpoint_period == 0:
checkpointer.save("model_{:07d}".format(iteration), **arguments)
if data_loader_val is not None and test_period > 0 and iteration % test_period == 0:
meters_val = MetricLogger(delimiter=" ")
synchronize()
_ = inference( # The result can be used for additional logging, e. g. for TensorBoard
model,
# The method changes the segmentation mask format in a data loader,
# so every time a new data loader is created:
make_data_loader(cfg,
is_train=False,
is_distributed=(get_world_size() > 1),
is_for_period=True),
dataset_name="[Validation]",
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=None,
)
synchronize()
model.train()
with torch.no_grad():
# Should be one image for each GPU:
for iteration_val, (images_val, targets_val,
_) in enumerate(tqdm(data_loader_val)):
images_val = images_val.to(device)
targets_val = [target.to(device) for target in targets_val]
loss_dict = model(images_val, targets_val)
losses = sum(loss for loss in loss_dict.values())
loss_dict_reduced = reduce_loss_dict(loss_dict)
losses_reduced = sum(
loss for loss in loss_dict_reduced.values())
meters_val.update(loss=losses_reduced, **loss_dict_reduced)
synchronize()
logger.info(
meters_val.delimiter.join([
"[Validation]: ",
"eta: {eta}",
"iter: {iter}",
"{meters}",
"lr: {lr:.6f}",
"max mem: {memory:.0f}",
]).format(
eta=eta_string,
iter=iteration,
meters=str(meters_val),
lr=optimizer.param_groups[0]["lr"],
memory=torch.cuda.max_memory_allocated() / 1024.0 / 1024.0,
))
if iteration == max_iter:
checkpointer.save("model_final", **arguments)
total_training_time = time.time() - start_training_time
total_time_str = str(datetime.timedelta(seconds=total_training_time))
logger.info("Total training time: {} ({:.4f} s / it)".format(
total_time_str, total_training_time / (max_iter)))
def main():
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(
"--ckpt",
help=
"The path to the checkpoint for test, default is the latest checkpoint.",
default=None,
)
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_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.freeze()
save_dir = ""
logger = setup_logger("asynet_mask_rcnn", 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)
# Initialize mixed-precision if necessary
use_mixed_precision = cfg.DTYPE == 'float16'
amp_handle = amp.init(enabled=use_mixed_precision, verbose=cfg.AMP_VERBOSE)
output_dir = cfg.OUTPUT_DIR
checkpointer = DetectronCheckpointer(cfg, model, save_dir=output_dir)
ckpt = cfg.MODEL.WEIGHT if args.ckpt is None else args.ckpt
_ = checkpointer.load(ckpt, use_latest=args.ckpt is None)
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,
bbox_aug=cfg.TEST.BBOX_AUG.ENABLED,
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()
def main():
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()
num_gpus = int(
os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1
args.distributed = num_gpus > 1
if args.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.freeze()
output_dir = cfg.OUTPUT_DIR
if output_dir:
mkdir(output_dir)
logger = setup_logger("asynet_mask_rcnn", output_dir, get_rank())
logger.info("Using {} GPUs".format(num_gpus))
logger.info(args)
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))
output_config_path = os.path.join(cfg.OUTPUT_DIR, 'config.yml')
logger.info("Saving config into: {}".format(output_config_path))
# save overloaded model config in the output directory
save_config(cfg, output_config_path)
model = train(cfg, args.local_rank, args.distributed)
if not args.skip_test:
run_test(cfg, model, args.distributed)