def main():
args = parse_args()
cfg = mmcv.Config.fromfile(args.config)
# set cudnn_benchmark
if cfg.get('cudnn_benchmark', False):
torch.backends.cudnn.benchmark = True
cfg.model.pretrained = None
# init distributed env first, since logger depends on the dist info.
if args.launcher == 'none':
distributed = False
else:
distributed = True
init_dist(args.launcher, **cfg.dist_params)
rank, _ = get_dist_info()
# set random seeds
if args.seed is not None:
if rank == 0:
print('set random seed to', args.seed)
set_random_seed(args.seed, deterministic=args.deterministic)
# build the dataloader
# TODO: support multiple images per gpu (only minor changes are needed)
dataset = build_dataset(cfg.data.test)
loader_cfg = {
**dict((k, cfg.data[k]) for k in ['workers_per_gpu'] if k in cfg.data),
**dict(samples_per_gpu=1,
drop_last=False,
shuffle=False,
dist=distributed),
**cfg.data.get('test_dataloader', {})
}
data_loader = build_dataloader(dataset, **loader_cfg)
# build the model and load checkpoint
model = build_model(cfg.model, train_cfg=None, test_cfg=cfg.test_cfg)
args.save_image = args.save_path is not None
empty_cache = cfg.get('empty_cache', False)
if not distributed:
_ = load_checkpoint(model, args.checkpoint, map_location='cpu')
model = MMDataParallel(model, device_ids=[0])
outputs = single_gpu_test(model,
data_loader,
save_path=args.save_path,
save_image=args.save_image)
else:
find_unused_parameters = cfg.get('find_unused_parameters', False)
model = DistributedDataParallelWrapper(
model,
device_ids=[torch.cuda.current_device()],
broadcast_buffers=False,
find_unused_parameters=find_unused_parameters)
device_id = torch.cuda.current_device()
_ = load_checkpoint(
model,
args.checkpoint,
map_location=lambda storage, loc: storage.cuda(device_id))
outputs = multi_gpu_test(model,
data_loader,
args.tmpdir,
args.gpu_collect,
save_path=args.save_path,
save_image=args.save_image,
empty_cache=empty_cache)
if rank == 0:
print('')
# print metrics
stats = dataset.evaluate(outputs)
for stat in stats:
print('Eval-{}: {}'.format(stat, stats[stat]))
# save result pickle
if args.out:
print('writing results to {}'.format(args.out))
mmcv.dump(outputs, args.out)
def _dist_train(model,
dataset,
cfg,
validate=False,
logger=None,
timestamp=None,
meta=None):
"""Distributed training function.
Args:
model (nn.Module): The model to be trained.
dataset (:obj:`Dataset`): Train dataset.
cfg (dict): The config dict for training.
validate (bool): Whether to do evaluation. Default: False.
logger (logging.Logger | None): Logger for training. Default: None.
timestamp (str | None): Local time for runner. Default: None.
meta (dict | None): Meta dict to record some important information.
Default: None.
"""
dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
# step 1: give default values and override (if exist) from cfg.data
loader_cfg = dict(
seed=cfg.get('seed'),
drop_last=False,
dist=True,
**({} if torch.__version__ != 'parrots' else dict(
prefetch_num=2,
pin_memory=False,
)),
**dict((k, cfg.data[k]) for k in [
'samples_per_gpu',
'workers_per_gpu',
'shuffle',
'seed',
'drop_last',
'prefetch_num',
'pin_memory',
] if k in cfg.data))
# step 2: cfg.data.train_dataloader has highest priority
train_loader_cfg = dict(loader_cfg, **cfg.data.get('train_dataloader', {}))
data_loaders = [build_dataloader(ds, **train_loader_cfg) for ds in dataset]
# put model on gpus
find_unused_parameters = cfg.get('find_unused_parameters', False)
model = DistributedDataParallelWrapper(
model,
device_ids=[torch.cuda.current_device()],
broadcast_buffers=False,
find_unused_parameters=find_unused_parameters)
# build runner
optimizer = build_optimizers(model, cfg.optimizers)
runner = IterBasedRunner(
model,
optimizer=optimizer,
work_dir=cfg.work_dir,
logger=logger,
meta=meta)
# an ugly walkaround to make the .log and .log.json filenames the same
runner.timestamp = timestamp
# register hooks
runner.register_training_hooks(
cfg.lr_config,
checkpoint_config=cfg.checkpoint_config,
log_config=cfg.log_config)
# visual hook
if cfg.get('visual_config', None) is not None:
cfg.visual_config['output_dir'] = os.path.join(
cfg.work_dir, cfg.visual_config['output_dir'])
runner.register_hook(mmcv.build_from_cfg(cfg.visual_config, HOOKS))
# evaluation hook
if validate and cfg.get('evaluation', None) is not None:
dataset = build_dataset(cfg.data.val)
if ('val_samples_per_gpu' in cfg.data
or 'val_workers_per_gpu' in cfg.data):
warnings.warn('"val_samples_per_gpu/val_workers_per_gpu" have '
'been deprecated. Please use '
'"val_dataloader=dict(samples_per_gpu=1)" instead. '
'Details see '
'https://github.com/open-mmlab/mmediting/pull/201')
val_loader_cfg = dict(
loader_cfg,
shuffle=False,
drop_last=False,
**dict((newk, cfg.data[oldk]) for oldk, newk in [
('val_samples_per_gpu', 'samples_per_gpu'),
('val_workers_per_gpu', 'workers_per_gpu'),
] if oldk in cfg.data),
**cfg.data.get('val_dataloader', {}))
data_loader = build_dataloader(dataset, **val_loader_cfg)
save_path = osp.join(cfg.work_dir, 'val_visuals')
runner.register_hook(
DistEvalIterHook(
data_loader, save_path=save_path, **cfg.evaluation))
if cfg.resume_from:
runner.resume(cfg.resume_from)
elif cfg.load_from:
runner.load_checkpoint(cfg.load_from)
runner.run(data_loaders, cfg.workflow, cfg.total_iters)
def _dist_train(model,
dataset,
cfg,
validate=False,
logger=None,
timestamp=None,
meta=None):
"""Distributed training function.
Args:
model (nn.Module): The model to be trained.
dataset (:obj:`Dataset`): Train dataset.
cfg (dict): The config dict for training.
validate (bool): Whether to do evaluation. Default: False.
logger (logging.Logger | None): Logger for training. Default: None.
timestamp (str | None): Local time for runner. Default: None.
meta (dict | None): Meta dict to record some important information.
Default: None.
"""
# prepare data loaders
dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
data_loaders = [
build_dataloader(ds,
cfg.data.samples_per_gpu,
cfg.data.workers_per_gpu,
dist=True,
drop_last=cfg.data.get('drop_last', False),
seed=cfg.seed) for ds in dataset
]
# put model on gpus
find_unused_parameters = cfg.get('find_unused_parameters', False)
model = DistributedDataParallelWrapper(
model,
device_ids=[torch.cuda.current_device()],
broadcast_buffers=False,
find_unused_parameters=find_unused_parameters)
# build runner
optimizer = build_optimizers(model, cfg.optimizers)
runner = IterBasedRunner(model,
optimizer=optimizer,
work_dir=cfg.work_dir,
logger=logger,
meta=meta)
# an ugly walkaround to make the .log and .log.json filenames the same
runner.timestamp = timestamp
# register hooks
runner.register_training_hooks(cfg.lr_config,
checkpoint_config=cfg.checkpoint_config,
log_config=cfg.log_config)
# visual hook
if cfg.get('visual_config', None) is not None:
cfg.visual_config['output_dir'] = os.path.join(
cfg.work_dir, cfg.visual_config['output_dir'])
runner.register_hook(mmcv.build_from_cfg(cfg.visual_config, HOOKS))
# evaluation hook
if validate and cfg.get('evaluation', None) is not None:
dataset = build_dataset(cfg.data.val)
samples_per_gpu = cfg.data.get('val_samples_per_gpu',
cfg.data.samples_per_gpu)
workers_per_gpu = cfg.data.get('val_workers_per_gpu',
cfg.data.workers_per_gpu)
data_loader = build_dataloader(dataset,
samples_per_gpu=samples_per_gpu,
workers_per_gpu=workers_per_gpu,
dist=True,
shuffle=False)
save_path = osp.join(cfg.work_dir, 'val_visuals')
runner.register_hook(
DistEvalIterHook(data_loader,
save_path=save_path,
**cfg.evaluation))
if cfg.resume_from:
runner.resume(cfg.resume_from)
elif cfg.load_from:
runner.load_checkpoint(cfg.load_from)
runner.run(data_loaders, cfg.workflow, cfg.total_iters)
def main():
args = parse_args()
checkpoint_list = os.listdir(args.checkpoint_dir)
print(checkpoint_list)
for checkpoint in checkpoint_list:
if '.pth' in checkpoint:
cfg = mmcv.Config.fromfile(args.config)
# set cudnn_benchmark
if cfg.get('cudnn_benchmark', False):
torch.backends.cudnn.benchmark = True
cfg.model.pretrained = None
# init distributed env first, since logger depends on the dist info.
if args.launcher == 'none':
distributed = False
else:
distributed = True
init_dist(args.launcher, **cfg.dist_params)
rank, _ = get_dist_info()
# set random seeds
if args.seed is not None:
if rank == 0:
print('set random seed to', args.seed)
set_random_seed(args.seed, deterministic=args.deterministic)
# build the dataloader
# TODO: support multiple images per gpu (only minor changes are needed)
dataset = build_dataset(cfg.data.test)
data_loader = build_dataloader(dataset,
samples_per_gpu=1,
workers_per_gpu=cfg.data.get(
'val_workers_per_gpu',
cfg.data.workers_per_gpu),
dist=distributed,
shuffle=False)
# build the model and load checkpoint
model = build_model(cfg.model,
train_cfg=None,
test_cfg=cfg.test_cfg)
args.save_image = args.save_path is not None
# distributed test
find_unused_parameters = cfg.get('find_unused_parameters', False)
model = DistributedDataParallelWrapper(
model,
device_ids=[torch.cuda.current_device()],
broadcast_buffers=False,
find_unused_parameters=find_unused_parameters)
device_id = torch.cuda.current_device()
_ = load_checkpoint(
model,
os.path.join(args.checkpoint_dir, checkpoint),
map_location=lambda storage, loc: storage.cuda(device_id))
outputs = multi_gpu_test(model,
data_loader,
args.tmpdir,
args.gpu_collect,
save_path=args.save_path,
save_image=args.save_image)
if rank == 0:
# print metrics
stats = dataset.evaluate(outputs)
write_file = open(
os.path.join(args.checkpoint_dir, 'eval_result_new.txt'),
'a')
for stat in stats:
print('{}: Eval-{}: {}'.format(checkpoint, stat,
stats[stat]))
write_file.write('{}: Eval-{}: {} '.format(
checkpoint, stat, stats[stat]))
write_file.write('\n')
write_file.close()
# save result pickle
if args.out:
print('writing results to {}'.format(args.out))
mmcv.dump(outputs, args.out)