def build_metrics(cfg):
if isinstance(cfg, (list, tuple)):
metrics = [build_metrics(c) for c in cfg]
else:
metrics = [build_from_cfg(cfg, registry, 'metrics')]
return metrics
def __call__(self, model):
if hasattr(model, 'module'):
model = model.module
optimizer_cfg = self.optimizer_cfg.copy()
# if no paramwise option is specified, just use the global setting
if not self.paramwise_cfg:
optimizer_cfg['params'] = model.parameters()
return build_from_cfg(optimizer_cfg, registry, 'optimizer')
# set param-wise lr and weight decay recursively
params = []
self.add_params(params, model)
optimizer_cfg['params'] = params
return build_from_cfg(optimizer_cfg, registry, 'optimizer')
def build_dataset(cfg, default_args=None):
if isinstance(cfg, (list, tuple)):
dataset = ConcatDataset([build_dataset(c, default_args) for c in cfg])
else:
dataset = build_from_cfg(cfg, registry, 'dataset', default_args)
return dataset
def __init__(self, transforms):
assert isinstance(transforms, collections.abc.Sequence)
self.transforms = []
for transform in transforms:
if isinstance(transform, dict):
transform = build_from_cfg(transform, registry, 'pipeline')
self.transforms.append(transform)
elif callable(transform):
self.transforms.append(transform)
else:
raise TypeError('transform must be callable or a dict')
def build_activation_layer(cfg):
"""Build activation layer.
Args:
cfg (dict): The activation layer config, which should contain:
- type (str): Layer type.
- layer args: Args needed to instantiate an activation layer.
Returns:
nn.Module: Created activation layer.
"""
return build_from_cfg(cfg, registry, 'activation')
def build(cfg, module_name, default_args=None):
"""Build a module.
Args:
cfg (dict, list[dict]): The config of modules, is is either a dict
or a list of configs.
registry (:obj:`Registry`): A registry the module belongs to.
default_args (dict, optional): Default arguments to build the module.
Defaults to None.
Returns:
nn.Module: A built nn module.
"""
if isinstance(cfg, list):
modules = [
build_from_cfg(cfg_, registry, module_name, default_args)
for cfg_ in cfg
]
return nn.Sequential(*modules)
else:
return build_from_cfg(cfg, registry, module_name, default_args)
def build_dataset(cfg, default_args=None):
from .dataset_wrappers import (ClassBalancedDataset, ConcatDataset,
RepeatDataset)
if isinstance(cfg, (list, tuple)):
dataset = ConcatDataset([build_dataset(c, default_args) for c in cfg])
elif cfg['typename'] == 'RepeatDataset':
dataset = RepeatDataset(
build_dataset(cfg['dataset'], default_args), cfg['times'])
elif cfg['typename'] == 'ClassBalancedDataset':
dataset = ClassBalancedDataset(
build_dataset(cfg['dataset'], default_args), cfg['oversample_thr'])
elif isinstance(cfg.get('ann_file'), (list, tuple)):
dataset = _concat_dataset(cfg, default_args)
else:
dataset = build_from_cfg(cfg, registry, 'dataset', default_args)
return dataset
def build_engine(cfg):
return build_from_cfg(cfg, registry, 'engine')
def build_loss(cfg):
return build_from_cfg(cfg, registry, 'loss')
def build_converter(cfg):
return build_from_cfg(cfg, registry, 'converter')
def build_loss(cfg):
loss = build_from_cfg(cfg, registry, 'loss')
return loss
def build_model(cfg):
return build_from_cfg(cfg, registry, 'sal_detector')
def build_meshgrid(cfg):
return build_from_cfg(cfg, registry, 'meshgrid')
def build_backbone(cfg):
"""Build backbone."""
return build_from_cfg(cfg, registry, 'backbone')
def build_assigner(cfg, **default_args):
"""Builder of box assigner."""
return build_from_cfg(cfg, registry, 'bbox_assigner', default_args)
def build_bbox_coder(cfg, **default_args):
bbox_coder = build_from_cfg(cfg, registry, 'bbox_coder', default_args)
return bbox_coder
def build_optimizer_constructor(cfg):
return build_from_cfg(cfg, registry, 'optimizer_builder')
def build_iou_calculator(cfg, default_args=None):
"""Builder of IoU calculator."""
return build_from_cfg(cfg, registry, 'iou_calculator', default_args)
def build_head(cfg):
"""Build head."""
return build_from_cfg(cfg, registry, 'head')
def build_base_anchor(cfg):
return build_from_cfg(cfg, registry, 'base_anchor')
def build_detector(cfg):
return build_from_cfg(cfg, registry, 'detector')
def build_hook(cfg):
return build_from_cfg(cfg, registry, 'hook')
def build_neck(cfg):
"""Build neck."""
return build_from_cfg(cfg, registry, 'neck')
def build_criterion(cfg):
return build_from_cfg(cfg, registry, 'criterion')
def build_sampler(cfg, **default_args):
"""Builder of box sampler."""
return build_from_cfg(cfg, registry, 'bbox_sampler', default_args)
def build_sequencer(cfg):
"""Build sequencer."""
return build_from_cfg(cfg, registry, 'sequencer')
