def __init__(self, in_channels, out_channels, *, stride=1, norm="BN"):
"""
Args:
in_channels (int): Number of input channels.
out_channels (int): Number of output channels.
stride (int): Stride for the first conv.
norm (str or callable): normalization for all conv layers.
See :func:`layers.get_norm` for supported format.
"""
super().__init__(in_channels, out_channels, stride)
if in_channels != out_channels:
self.shortcut = Conv2d(
in_channels,
out_channels,
kernel_size=1,
stride=stride,
bias=False,
norm=get_norm(norm, out_channels),
)
else:
self.shortcut = None
self.conv1 = Conv2d(
in_channels,
out_channels,
kernel_size=3,
stride=stride,
padding=1,
bias=False,
norm=get_norm(norm, out_channels),
)
self.conv2 = Conv2d(
out_channels,
out_channels,
kernel_size=3,
stride=1,
padding=1,
bias=False,
norm=get_norm(norm, out_channels),
)
for layer in [self.conv1, self.conv2, self.shortcut]:
if layer is not None: # shortcut can be None
weight_init.c2_msra_fill(layer)
def __init__(self, in_channels=3, out_channels=64, norm="BN"):
"""
Args:
norm (str or callable): norm after the first conv layer.
See :func:`layers.get_norm` for supported format.
"""
super().__init__(in_channels, out_channels, 4)
self.in_channels = in_channels
self.conv1 = Conv2d(
in_channels,
out_channels,
kernel_size=7,
stride=2,
padding=3,
bias=False,
norm=get_norm(norm, out_channels),
)
weight_init.c2_msra_fill(self.conv1)
def __init__(
self,
in_channels,
out_channels,
*,
bottleneck_channels,
stride=1,
num_groups=1,
norm="BN",
stride_in_1x1=False,
dilation=1,
):
"""
Args:
bottleneck_channels (int): number of output channels for the 3x3
"bottleneck" conv layers.
num_groups (int): number of groups for the 3x3 conv layer.
norm (str or callable): normalization for all conv layers.
See :func:`layers.get_norm` for supported format.
stride_in_1x1 (bool): when stride>1, whether to put stride in the
first 1x1 convolution or the bottleneck 3x3 convolution.
dilation (int): the dilation rate of the 3x3 conv layer.
"""
super().__init__(in_channels, out_channels, stride)
if in_channels != out_channels:
self.shortcut = Conv2d(
in_channels,
out_channels,
kernel_size=1,
stride=stride,
bias=False,
norm=get_norm(norm, out_channels),
)
else:
self.shortcut = None
# The original MSRA ResNet models have stride in the first 1x1 conv
# The subsequent fb.torch.resnet and Caffe2 ResNe[X]t implementations have
# stride in the 3x3 conv
stride_1x1, stride_3x3 = (stride, 1) if stride_in_1x1 else (1, stride)
self.conv1 = Conv2d(
in_channels,
bottleneck_channels,
kernel_size=1,
stride=stride_1x1,
bias=False,
norm=get_norm(norm, bottleneck_channels),
)
self.conv2 = Conv2d(
bottleneck_channels,
bottleneck_channels,
kernel_size=3,
stride=stride_3x3,
padding=1 * dilation,
bias=False,
groups=num_groups,
dilation=dilation,
norm=get_norm(norm, bottleneck_channels),
)
self.conv3 = Conv2d(
bottleneck_channels,
out_channels,
kernel_size=1,
bias=False,
norm=get_norm(norm, out_channels),
)
for layer in [self.conv1, self.conv2, self.conv3, self.shortcut]:
if layer is not None: # shortcut can be None
weight_init.c2_msra_fill(layer)
def __init__(
self,
in_channels,
out_channels,
*,
bottleneck_channels,
stride=1,
num_groups=1,
norm="BN",
stride_in_1x1=False,
dilation=1,
deform_modulated=False,
deform_num_groups=1,
):
super().__init__(in_channels, out_channels, stride)
self.deform_modulated = deform_modulated
if in_channels != out_channels:
self.shortcut = Conv2d(
in_channels,
out_channels,
kernel_size=1,
stride=stride,
bias=False,
norm=get_norm(norm, out_channels),
)
else:
self.shortcut = None
stride_1x1, stride_3x3 = (stride, 1) if stride_in_1x1 else (1, stride)
self.conv1 = Conv2d(
in_channels,
bottleneck_channels,
kernel_size=1,
stride=stride_1x1,
bias=False,
norm=get_norm(norm, bottleneck_channels),
)
if deform_modulated:
deform_conv_op = ModulatedDeformConv
# offset channels are 2 or 3 (if with modulated) * kernel_size * kernel_size
offset_channels = 27
else:
deform_conv_op = DeformConv
offset_channels = 18
self.conv2_offset = Conv2d(
bottleneck_channels,
offset_channels * deform_num_groups,
kernel_size=3,
stride=stride_3x3,
padding=1 * dilation,
dilation=dilation,
)
self.conv2 = deform_conv_op(
bottleneck_channels,
bottleneck_channels,
kernel_size=3,
stride=stride_3x3,
padding=1 * dilation,
bias=False,
groups=num_groups,
dilation=dilation,
deformable_groups=deform_num_groups,
norm=get_norm(norm, bottleneck_channels),
)
self.conv3 = Conv2d(
bottleneck_channels,
out_channels,
kernel_size=1,
bias=False,
norm=get_norm(norm, out_channels),
)
for layer in [self.conv1, self.conv2, self.conv3, self.shortcut]:
if layer is not None: # shortcut can be None
weight_init.c2_msra_fill(layer)
nn.init.constant_(self.conv2_offset.weight, 0)
nn.init.constant_(self.conv2_offset.bias, 0)
def __init__(
self,
*,
input_shape: Dict[str,ShapeSpec],
num_classes: Dict[str, int],
norm="",
prior_prob=0.01,
final_channel=512,
hierarchy_loss=False,
hierarchy_prior=None,
):
"""
NOTE: this interface is experimental.
Args:
input_shape (List[ShapeSpec]): input shape
num_classes (int): number of classes. Used to label background proposals.
num_anchors (int): number of generated anchors
conv_dims (List[int]): dimensions for each convolution layer
norm (str or callable):
Normalization for conv layers except for the two output layers.
See :func:`herbarium.layers.get_norm` for supported types.
prior_prob (float): Prior weight for computing bias
"""
super().__init__()
if norm == "BN" or norm == "SyncBN":
logger = logging.getLogger(__name__)
logger.warn("Shared norm does not work well for BN, SyncBN, expect poor results")
cls_subnet = defaultdict(list)
for input_key in list(input_shape.keys()):
in_channels = input_shape[input_key].channels
out_channels = final_channel
cls_subnet[input_key] = [AttentionConv(in_channels, out_channels, kernel_size=3, stride=1, padding=1)]
if norm:
cls_subnet[input_key].append(get_norm(norm, out_channels))
cls_subnet[input_key].append(nn.ReLU())
cls_subnet[input_key].append(nn.AdaptiveAvgPool2d((1,1)))
cls_subnet[input_key] = nn.Sequential(*cls_subnet[input_key])
self.cls_subnet = nn.ModuleDict(cls_subnet)
cls_score = defaultdict(list)
for cls_level in list(num_classes.keys()):
cls_score[cls_level] = nn.Conv2d(
final_channel, num_classes[cls_level], kernel_size=1, stride=1
)
self.cls_score = nn.ModuleDict(cls_score)
self.hierarchy_prior = nn.ParameterDict({
k: nn.Parameter(v, requires_grad=False) for k, v in hierarchy_prior.items()
})
# Initialization
for modules in [self.cls_subnet, self.cls_score]:
for layer in modules.modules():
if isinstance(layer, nn.Conv2d):
if layer.weight is not None:
torch.nn.init.normal_(layer.weight, mean=0, std=0.01)
if layer.bias is not None:
torch.nn.init.constant_(layer.bias, 0)