def build_fcos_resnet_fpn_backbone(cfg, input_shape: ShapeSpec):
"""
Args:
cfg: a detectron2 CfgNode
Returns:
backbone (Backbone): backbone module, must be a subclass of :class:`Backbone`.
"""
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.FPN.IN_FEATURES
out_channels = cfg.MODEL.FPN.OUT_CHANNELS
top_levels = cfg.MODEL.FCOS.TOP_LEVELS
in_channels_top = out_channels
if top_levels == 2:
top_block = LastLevelP6P7(in_channels_top, out_channels, "p5")
if top_levels == 1:
top_block = LastLevelP6(in_channels_top, out_channels, "p5")
elif top_levels == 0:
top_block = None
backbone = FPN(
bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
norm=cfg.MODEL.FPN.NORM,
top_block=top_block,
fuse_type=cfg.MODEL.FPN.FUSE_TYPE,
)
return backbone
def build_resnest_fpn_backbone(cfg, input_shape: ShapeSpec):
if cfg.MODEL.RESNEST:
bottom_up = build_resnest_backbone(cfg, input_shape)
else:
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.FPN.IN_FEATURES
out_channels = cfg.MODEL.FPN.OUT_CHANNELS
top_levels = cfg.MODEL.FCOS.TOP_LEVELS
in_channels_top = out_channels
if top_levels == 2:
top_block = LastLevelP6P7(in_channels_top, out_channels, "p5")
if top_levels == 1:
top_block = LastLevelP6(in_channels_top, out_channels, "p5")
elif top_levels == 0:
top_block = None
backbone = FPN(
bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
norm=cfg.MODEL.FPN.NORM,
top_block=top_block,
fuse_type=cfg.MODEL.FPN.FUSE_TYPE,
)
return backbone
#BiFPN
#https://github.com/sxhxliang/detectron2_backbone/blob/master/detectron2_backbone/backbone/bifpn.py
def test_resnet_scriptability(self):
cfg = get_cfg()
resnet = build_resnet_backbone(cfg, ShapeSpec(channels=3))
scripted_resnet = torch.jit.script(resnet)
inp = torch.rand(2, 3, 100, 100)
out1 = resnet(inp)["res4"]
out2 = scripted_resnet(inp)["res4"]
self.assertTrue(torch.allclose(out1, out2))
def __init__(self, cfg, input_shape):
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.FPN.IN_FEATURES
out_channels = cfg.MODEL.FPN.OUT_CHANNELS
super().__init__(
bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
norm=cfg.MODEL.FPN.NORM,
top_block=LastLevelMaxPool(),
fuse_type=cfg.MODEL.FPN.FUSE_TYPE,
)
input_shapes = bottom_up.output_shape()
self.quantizer = build_quantizer(cfg, input_shapes)
def build_resnet_fpn_p5_backbone(cfg, input_shape: ShapeSpec):
"""
Build ResNet-FPN backbone with P6 and P7 from P5 feature.
Returns:
backbone (Backbone): backbone module, must be a subclass of :class:`Backbone`.
"""
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.FPN.IN_FEATURES
out_channels = cfg.MODEL.FPN.OUT_CHANNELS
in_channels_p6p7 = out_channels
backbone = FPN(
bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
norm=cfg.MODEL.FPN.NORM,
top_block=LastLevelP6P7fromP5(in_channels_p6p7, out_channels),
fuse_type=cfg.MODEL.FPN.FUSE_TYPE,
)
return backbone
def build_resnet_pafpn_backbone(cfg, input_shape: ShapeSpec):
"""
Args:
cfg: a detectron2 CfgNode
Returns:
backbone (Backbone): backbone module, must be a subclass of :class:`Backbone`.
"""
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.FPN.IN_FEATURES
out_channels = cfg.MODEL.FPN.OUT_CHANNELS
backbone = PAFPN(
bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
norm=cfg.MODEL.FPN.NORM,
top_block=LastLevelMaxPool(),
fuse_type=cfg.MODEL.FPN.FUSE_TYPE,
)
return backbone
def build_p35_fcos_resnet_bifpn_backbone(cfg, input_shape: ShapeSpec):
"""
Args:
cfg: a detectron2 CfgNode
Returns:
backbone (Backbone): backbone module, must be a subclass of :class:`Backbone`.
"""
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.FPN.IN_FEATURES
out_channels = cfg.MODEL.BIFPN.OUT_CHANNELS
num_repeats = cfg.MODEL.BIFPN.NUM_BIFPN
top_levels = 0
backbone = BiFPN(bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
num_top_levels=top_levels,
num_repeats=num_repeats,
norm=cfg.MODEL.BIFPN.NORM)
return backbone
def build_retinanet_resnet_vt_fpn_backbone_use_p5(cfg, input_shape: ShapeSpec):
"""
Args:
cfg: a detectron2 CfgNode
Returns:
backbone (Backbone): backbone module, must be a subclass of :class:`Backbone`.
"""
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.FPN.IN_FEATURES
out_channels = cfg.MODEL.FPN.OUT_CHANNELS
in_channels_p6p7 = out_channels
backbone = VT_FPN(
bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
norm=cfg.MODEL.FPN.NORM,
top_block=LastLevelP6P7(in_channels_p6p7,
out_channels,
in_feature="p5"),
fuse_type=cfg.MODEL.FPN.FUSE_TYPE,
)
return backbone
def __init__(self,
cfg=None,
model_path=None,
depth=101,
vec_dim=128,
max_pool=False):
super(ResNetbasedNet, self).__init__()
if cfg is not None:
model = build_resnet_backbone(
cfg, ShapeSpec(channels=len(cfg.MODEL.PIXEL_MEAN)))
self.backbone = nn.Sequential(*list(model.children()))
else:
model = torch.hub.load('pytorch/vision:v0.6.0',
'resnet{}'.format(depth),
pretrained=True)
self.backbone = nn.Sequential(*list(model.children())[:-2])
self.max_pool = nn.AdaptiveMaxPool2d(
(1, 1)) if max_pool else nn.AdaptiveAvgPool2d((1, 1))
self.fc = nn.Linear(2048, vec_dim)
weight_init.c2_xavier_fill(self.fc)
if model_path is not None:
model.load_state_dict(torch.load(model_path))
def build_fcos_resnet_bifpn_backbone(cfg, input_shape: ShapeSpec):
"""
Args:
cfg: a detectron2 CfgNode
Returns:
backbone (Backbone): backbone module, must be a subclass of :class:`Backbone`.
"""
if cfg.MODEL.MOBILENET:
bottom_up = build_mnv2_backbone(cfg, input_shape)
else:
bottom_up = build_resnet_backbone(cfg, input_shape)
in_features = cfg.MODEL.BiFPN.IN_FEATURES
out_channels = cfg.MODEL.BiFPN.OUT_CHANNELS
num_repeats = cfg.MODEL.BiFPN.NUM_REPEATS
top_levels = cfg.MODEL.FCOS.TOP_LEVELS
backbone = BiFPN(bottom_up=bottom_up,
in_features=in_features,
out_channels=out_channels,
num_top_levels=top_levels,
num_repeats=num_repeats,
norm=cfg.MODEL.BiFPN.NORM)
return backbone
import torch
from detectron2.config import get_cfg
from detectron2 import model_zoo
from detectron2.layers import ShapeSpec
from detectron2.modeling.backbone import build_resnet_backbone
cfg = get_cfg()
cfg.merge_from_file(
model_zoo.get_config_file("COCO-Detection/faster_rcnn_R_50_FPN_3x.yaml"))
cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 512
cfg.MODEL.ROI_HEADS.NUM_CLASSES = 2
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.85
net = build_resnet_backbone(cfg, ShapeSpec(channels=3))
temp = torch.load("weight.pt")
net.load_state_dict({k: temp[k] for k in net.state_dict()})
net.eval()
with torch.no_grad():
torch.save(net(torch.load("data.pt")), "res1.pt")
def __init__(self,
cfg=None,
load_path=None,
depth=101,
vec_dim=128,
max_pool=False,
clf1_num=None,
clf2_num=None,
adv_eta=None):
super(ResNetbasedNet, self).__init__()
self.load = True if load_path is not None else False
self.clf1 = True if clf1_num is not None else False
self.clf2 = True if clf2_num is not None else False
self.adv_eta = Variable(
torch.tensor(adv_eta).type(torch.float),
requires_grad=False) if adv_eta is not None else None
if cfg is not None:
model = build_resnet_backbone(
cfg, ShapeSpec(channels=len(cfg.MODEL.PIXEL_MEAN)))
pretrained_model = torch.load(cfg.MODEL.WEIGHTS)
cur_state = model.state_dict()
mapped_dict = {}
for name, param in pretrained_model.items():
if name == 'model':
for p in param:
if p.replace('backbone.bottom_up.', '') in cur_state:
mapped_dict[p.replace('backbone.bottom_up.',
'')] = param[p]
model.load_state_dict(mapped_dict)
self.backbone = nn.Sequential(*list(model.children()))
else:
model = torch.hub.load('pytorch/vision:v0.6.0',
'resnet{}'.format(depth),
pretrained=not self.load)
self.backbone = nn.Sequential(*list(model.children())[:-2])
self.max_pool = nn.AdaptiveMaxPool2d(
(1, 1)) if max_pool else nn.AdaptiveAvgPool2d((1, 1))
self.fc = nn.Linear(2048, vec_dim)
if self.clf1:
self.clf1_layer = nn.Sequential(nn.Linear(vec_dim, vec_dim),
nn.BatchNorm1d(vec_dim), nn.ReLU(),
nn.Linear(vec_dim, clf1_num))
if self.clf2:
self.clf2_layer = nn.Sequential(nn.Linear(vec_dim, vec_dim),
nn.BatchNorm1d(vec_dim), nn.ReLU(),
nn.Linear(vec_dim, clf2_num))
if self.load:
load_model = torch.load(load_path)
mapped_dict = {
'backbone': (self.backbone, {}),
'fc': (self.fc, {})
}
if self.clf1:
mapped_dict['clf1_layer'] = (self.clf1_layer, {})
if self.clf2:
# print(self.clf2_layer.state_dict())
mapped_dict['clf2_layer'] = (self.clf2_layer, {})
for name, param in load_model.items():
if name.split('.')[0] in mapped_dict.keys():
mapped_dict[name.split('.')[0]][1]['.'.join(
name.split('.')[1:])] = param
for layers in mapped_dict.keys():
mapped_dict[layers][0].load_state_dict(mapped_dict[layers][1])
