model = torch.jit.load('fod.pth')
torch.save(model.state_dict(), 'weights.pt')
u2net = U2NETP()
u2net.eval()
u2net.load_state_dict(torch.load('/content/weights.pt'), strict=False)
I am using this code to load my model weights to the U2NETP() network. But it keeps on giving me Incompatible keys. I tried visualizing the model using netron and the architecture is similar to that of the one used in U2NETP().
Please help me what should I do ?
P.S. Here are the keys that seem to be missing in the fod.pth file
_IncompatibleKeys(missing_keys=['stage1.rebnconvin.bn_s1.running_mean', 'stage1.rebnconvin.bn_s1.running_var', 'stage1.rebnconv1.bn_s1.running_mean', 'stage1.rebnconv1.bn_s1.running_var', 'stage1.rebnconv2.bn_s1.running_mean', 'stage1.rebnconv2.bn_s1.running_var', 'stage1.rebnconv3.bn_s1.running_mean', 'stage1.rebnconv3.bn_s1.running_var', 'stage1.rebnconv4.bn_s1.running_mean', 'stage1.rebnconv4.bn_s1.running_var', 'stage1.rebnconv5.bn_s1.running_mean', 'stage1.rebnconv5.bn_s1.running_var', 'stage1.rebnconv6.bn_s1.running_mean', 'stage1.rebnconv6.bn_s1.running_var', 'stage1.rebnconv7.bn_s1.running_mean', 'stage1.rebnconv7.bn_s1.running_var', 'stage1.rebnconv6d.bn_s1.running_mean', 'stage1.rebnconv6d.bn_s1.running_var', 'stage1.rebnconv5d.bn_s1.running_mean', 'stage1.rebnconv5d.bn_s1.running_var', 'stage1.rebnconv4d.bn_s1.running_mean', 'stage1.rebnconv4d.bn_s1.running_var', 'stage1.rebnconv3d.bn_s1.running_mean', 'stage1.rebnconv3d.bn_s1.running_var', 'stage1.rebnconv2d.bn_s1.running_mean', 'stage1.rebnconv2d.bn_s1.running_var', 'stage1.rebnconv1d.bn_s1.running_mean', 'stage1.rebnconv1d.bn_s1.running_var', 'stage2.rebnconvin.bn_s1.running_mean', 'stage2.rebnconvin.bn_s1.running_var', 'stage2.rebnconv1.bn_s1.running_mean', 'stage2.rebnconv1.bn_s1.running_var', 'stage2.rebnconv2.bn_s1.running_mean', 'stage2.rebnconv2.bn_s1.running_var', 'stage2.rebnconv3.bn_s1.running_mean', 'stage2.rebnconv3.bn_s1.running_var', 'stage2.rebnconv4.bn_s1.running_mean', 'stage2.rebnconv4.bn_s1.running_var', 'stage2.rebnconv5.bn_s1.running_mean', 'stage2.rebnconv5.bn_s1.running_var', 'stage2.rebnconv6.bn_s1.running_mean', 'stage2.rebnconv6.bn_s1.running_var', 'stage2.rebnconv5d.bn_s1.running_mean', 'stage2.rebnconv5d.bn_s1.running_var', 'stage2.rebnconv4d.bn_s1.running_mean', 'stage2.rebnconv4d.bn_s1.running_var', 'stage2.rebnconv3d.bn_s1.running_mean', 'stage2.rebnconv3d.bn_s1.running_var', 'stage2.rebnconv2d.bn_s1.running_mean', 'stage2.rebnconv2d.bn_s1.running_var', 'stage2.rebnconv1d.bn_s1.running_mean', 'stage2.rebnconv1d.bn_s1.running_var', 'stage3.rebnconvin.bn_s1.running_mean', 'stage3.rebnconvin.bn_s1.running_var', 'stage3.rebnconv1.bn_s1.running_mean', 'stage3.rebnconv1.bn_s1.running_var', 'stage3.rebnconv2.bn_s1.running_mean', 'stage3.rebnconv2.bn_s1.running_var', 'stage3.rebnconv3.bn_s1.running_mean', 'stage3.rebnconv3.bn_s1.running_var', 'stage3.rebnconv4.bn_s1.running_mean', 'stage3.rebnconv4.bn_s1.running_var', 'stage3.rebnconv5.bn_s1.running_mean', 'stage3.rebnconv5.bn_s1.running_var', 'stage3.rebnconv4d.bn_s1.running_mean', 'stage3.rebnconv4d.bn_s1.running_var', 'stage3.rebnconv3d.bn_s1.running_mean', 'stage3.rebnconv3d.bn_s1.running_var', 'stage3.rebnconv2d.bn_s1.running_mean', 'stage3.rebnconv2d.bn_s1.running_var', 'stage3.rebnconv1d.bn_s1.running_mean', 'stage3.rebnconv1d.bn_s1.running_var', 'stage4.rebnconvin.bn_s1.running_mean', 'stage4.rebnconvin.bn_s1.running_var', 'stage4.rebnconv1.bn_s1.running_mean', 'stage4.rebnconv1.bn_s1.running_var', 'stage4.rebnconv2.bn_s1.running_mean', 'stage4.rebnconv2.bn_s1.running_var', 'stage4.rebnconv3.bn_s1.running_mean', 'stage4.rebnconv3.bn_s1.running_var', 'stage4.rebnconv4.bn_s1.running_mean', 'stage4.rebnconv4.bn_s1.running_var', 'stage4.rebnconv3d.bn_s1.running_mean', 'stage4.rebnconv3d.bn_s1.running_var', 'stage4.rebnconv2d.bn_s1.running_mean', 'stage4.rebnconv2d.bn_s1.running_var', 'stage4.rebnconv1d.bn_s1.running_mean', 'stage4.rebnconv1d.bn_s1.running_var', 'stage5.rebnconvin.bn_s1.running_mean', 'stage5.rebnconvin.bn_s1.running_var', 'stage5.rebnconv1.bn_s1.running_mean', 'stage5.rebnconv1.bn_s1.running_var', 'stage5.rebnconv2.bn_s1.running_mean', 'stage5.rebnconv2.bn_s1.running_var', 'stage5.rebnconv3.bn_s1.running_mean', 'stage5.rebnconv3.bn_s1.running_var', 'stage5.rebnconv4.bn_s1.running_mean', 'stage5.rebnconv4.bn_s1.running_var', 'stage5.rebnconv3d.bn_s1.running_mean', 'stage5.rebnconv3d.bn_s1.running_var', 'stage5.rebnconv2d.bn_s1.running_mean', 'stage5.rebnconv2d.bn_s1.running_var', 'stage5.rebnconv1d.bn_s1.running_mean', 'stage5.rebnconv1d.bn_s1.running_var', 'stage6.rebnconvin.bn_s1.running_mean', 'stage6.rebnconvin.bn_s1.running_var', 'stage6.rebnconv1.bn_s1.running_mean', 'stage6.rebnconv1.bn_s1.running_var', 'stage6.rebnconv2.bn_s1.running_mean', 'stage6.rebnconv2.bn_s1.running_var', 'stage6.rebnconv3.bn_s1.running_mean', 'stage6.rebnconv3.bn_s1.running_var', 'stage6.rebnconv4.bn_s1.running_mean', 'stage6.rebnconv4.bn_s1.running_var', 'stage6.rebnconv3d.bn_s1.running_mean', 'stage6.rebnconv3d.bn_s1.running_var', 'stage6.rebnconv2d.bn_s1.running_mean', 'stage6.rebnconv2d.bn_s1.running_var', 'stage6.rebnconv1d.bn_s1.running_mean', 'stage6.rebnconv1d.bn_s1.running_var', 'stage5d.rebnconvin.bn_s1.running_mean', 'stage5d.rebnconvin.bn_s1.running_var', 'stage5d.rebnconv1.bn_s1.running_mean', 'stage5d.rebnconv1.bn_s1.running_var', 'stage5d.rebnconv2.bn_s1.running_mean', 'stage5d.rebnconv2.bn_s1.running_var', 'stage5d.rebnconv3.bn_s1.running_mean', 'stage5d.rebnconv3.bn_s1.running_var', 'stage5d.rebnconv4.bn_s1.running_mean', 'stage5d.rebnconv4.bn_s1.running_var', 'stage5d.rebnconv3d.bn_s1.running_mean', 'stage5d.rebnconv3d.bn_s1.running_var', 'stage5d.rebnconv2d.bn_s1.running_mean', 'stage5d.rebnconv2d.bn_s1.running_var', 'stage5d.rebnconv1d.bn_s1.running_mean', 'stage5d.rebnconv1d.bn_s1.running_var', 'stage4d.rebnconvin.bn_s1.running_mean', 'stage4d.rebnconvin.bn_s1.running_var', 'stage4d.rebnconv1.bn_s1.running_mean', 'stage4d.rebnconv1.bn_s1.running_var', 'stage4d.rebnconv2.bn_s1.running_mean', 'stage4d.rebnconv2.bn_s1.running_var', 'stage4d.rebnconv3.bn_s1.running_mean', 'stage4d.rebnconv3.bn_s1.running_var', 'stage4d.rebnconv4.bn_s1.running_mean', 'stage4d.rebnconv4.bn_s1.running_var', 'stage4d.rebnconv3d.bn_s1.running_mean', 'stage4d.rebnconv3d.bn_s1.running_var', 'stage4d.rebnconv2d.bn_s1.running_mean', 'stage4d.rebnconv2d.bn_s1.running_var', 'stage4d.rebnconv1d.bn_s1.running_mean', 'stage4d.rebnconv1d.bn_s1.running_var', 'stage3d.rebnconvin.bn_s1.running_mean', 'stage3d.rebnconvin.bn_s1.running_var', 'stage3d.rebnconv1.bn_s1.running_mean', 'stage3d.rebnconv1.bn_s1.running_var', 'stage3d.rebnconv2.bn_s1.running_mean', 'stage3d.rebnconv2.bn_s1.running_var', 'stage3d.rebnconv3.bn_s1.running_mean', 'stage3d.rebnconv3.bn_s1.running_var', 'stage3d.rebnconv4.bn_s1.running_mean', 'stage3d.rebnconv4.bn_s1.running_var', 'stage3d.rebnconv5.bn_s1.running_mean', 'stage3d.rebnconv5.bn_s1.running_var', 'stage3d.rebnconv4d.bn_s1.running_mean', 'stage3d.rebnconv4d.bn_s1.running_var', 'stage3d.rebnconv3d.bn_s1.running_mean', 'stage3d.rebnconv3d.bn_s1.running_var', 'stage3d.rebnconv2d.bn_s1.running_mean', 'stage3d.rebnconv2d.bn_s1.running_var', 'stage3d.rebnconv1d.bn_s1.running_mean', 'stage3d.rebnconv1d.bn_s1.running_var', 'stage2d.rebnconvin.bn_s1.running_mean', 'stage2d.rebnconvin.bn_s1.running_var', 'stage2d.rebnconv1.bn_s1.running_mean', 'stage2d.rebnconv1.bn_s1.running_var', 'stage2d.rebnconv2.bn_s1.running_mean', 'stage2d.rebnconv2.bn_s1.running_var', 'stage2d.rebnconv3.bn_s1.running_mean', 'stage2d.rebnconv3.bn_s1.running_var', 'stage2d.rebnconv4.bn_s1.running_mean', 'stage2d.rebnconv4.bn_s1.running_var', 'stage2d.rebnconv5.bn_s1.running_mean', 'stage2d.rebnconv5.bn_s1.running_var', 'stage2d.rebnconv6.bn_s1.running_mean', 'stage2d.rebnconv6.bn_s1.running_var', 'stage2d.rebnconv5d.bn_s1.running_mean', 'stage2d.rebnconv5d.bn_s1.running_var', 'stage2d.rebnconv4d.bn_s1.running_mean', 'stage2d.rebnconv4d.bn_s1.running_var', 'stage2d.rebnconv3d.bn_s1.running_mean', 'stage2d.rebnconv3d.bn_s1.running_var', 'stage2d.rebnconv2d.bn_s1.running_mean', 'stage2d.rebnconv2d.bn_s1.running_var', 'stage2d.rebnconv1d.bn_s1.running_mean', 'stage2d.rebnconv1d.bn_s1.running_var', 'stage1d.rebnconvin.bn_s1.running_mean', 'stage1d.rebnconvin.bn_s1.running_var', 'stage1d.rebnconv1.bn_s1.running_mean', 'stage1d.rebnconv1.bn_s1.running_var', 'stage1d.rebnconv2.bn_s1.running_mean', 'stage1d.rebnconv2.bn_s1.running_var', 'stage1d.rebnconv3.bn_s1.running_mean', 'stage1d.rebnconv3.bn_s1.running_var', 'stage1d.rebnconv4.bn_s1.running_mean', 'stage1d.rebnconv4.bn_s1.running_var', 'stage1d.rebnconv5.bn_s1.running_mean', 'stage1d.rebnconv5.bn_s1.running_var', 'stage1d.rebnconv6.bn_s1.running_mean', 'stage1d.rebnconv6.bn_s1.running_var', 'stage1d.rebnconv7.bn_s1.running_mean', 'stage1d.rebnconv7.bn_s1.running_var', 'stage1d.rebnconv6d.bn_s1.running_mean', 'stage1d.rebnconv6d.bn_s1.running_var', 'stage1d.rebnconv5d.bn_s1.running_mean', 'stage1d.rebnconv5d.bn_s1.running_var', 'stage1d.rebnconv4d.bn_s1.running_mean', 'stage1d.rebnconv4d.bn_s1.running_var', 'stage1d.rebnconv3d.bn_s1.running_mean', 'stage1d.rebnconv3d.bn_s1.running_var', 'stage1d.rebnconv2d.bn_s1.running_mean', 'stage1d.rebnconv2d.bn_s1.running_var', 'stage1d.rebnconv1d.bn_s1.running_mean', 'stage1d.rebnconv1d.bn_s1.running_var'], unexpected_keys=[])
My model architecture is like this
class REBNCONV(nn.Module):
def __init__(self,in_ch=3,out_ch=3,dirate=1):
super(REBNCONV,self).__init__()
self.conv_s1 = nn.Conv2d(in_ch,out_ch,3,padding=1*dirate,dilation=1*dirate)
self.bn_s1 = nn.BatchNorm2d(out_ch)
self.relu_s1 = nn.ReLU(inplace=True)
def forward(self,x):
hx = x
xout = self.relu_s1(self.bn_s1(self.conv_s1(hx)))
return xout
## upsample tensor 'src' to have the same spatial size with tensor 'tar'
def _upsample_like(src,tar):
src = F.upsample(src,size=tar.shape[2:],mode='bilinear', align_corners=True)
return src
### RSU-7 ###
class RSU7(nn.Module):#UNet07DRES(nn.Module):
def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
super(RSU7,self).__init__()
self.rebnconvin = REBNCONV(in_ch,out_ch,dirate=1)
self.rebnconv1 = REBNCONV(out_ch,mid_ch,dirate=1)
self.pool1 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv2 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool2 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv3 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool3 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv4 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool4 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv5 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool5 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv6 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.rebnconv7 = REBNCONV(mid_ch,mid_ch,dirate=2)
self.rebnconv6d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv5d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv4d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv3d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv2d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv1d = REBNCONV(mid_ch*2,out_ch,dirate=1)
def forward(self,x):
hx = x
hxin = self.rebnconvin(hx)
hx1 = self.rebnconv1(hxin)
hx = self.pool1(hx1)
hx2 = self.rebnconv2(hx)
hx = self.pool2(hx2)
hx3 = self.rebnconv3(hx)
hx = self.pool3(hx3)
hx4 = self.rebnconv4(hx)
hx = self.pool4(hx4)
hx5 = self.rebnconv5(hx)
hx = self.pool5(hx5)
hx6 = self.rebnconv6(hx)
hx7 = self.rebnconv7(hx6)
hx6d = self.rebnconv6d(torch.cat((hx7,hx6),1))
hx6dup = _upsample_like(hx6d,hx5)
hx5d = self.rebnconv5d(torch.cat((hx6dup,hx5),1))
hx5dup = _upsample_like(hx5d,hx4)
hx4d = self.rebnconv4d(torch.cat((hx5dup,hx4),1))
hx4dup = _upsample_like(hx4d,hx3)
hx3d = self.rebnconv3d(torch.cat((hx4dup,hx3),1))
hx3dup = _upsample_like(hx3d,hx2)
hx2d = self.rebnconv2d(torch.cat((hx3dup,hx2),1))
hx2dup = _upsample_like(hx2d,hx1)
hx1d = self.rebnconv1d(torch.cat((hx2dup,hx1),1))
return hx1d + hxin
### RSU-6 ###
class RSU6(nn.Module):#UNet06DRES(nn.Module):
def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
super(RSU6,self).__init__()
self.rebnconvin = REBNCONV(in_ch,out_ch,dirate=1)
self.rebnconv1 = REBNCONV(out_ch,mid_ch,dirate=1)
self.pool1 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv2 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool2 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv3 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool3 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv4 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool4 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv5 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.rebnconv6 = REBNCONV(mid_ch,mid_ch,dirate=2)
self.rebnconv5d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv4d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv3d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv2d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv1d = REBNCONV(mid_ch*2,out_ch,dirate=1)
def forward(self,x):
hx = x
hxin = self.rebnconvin(hx)
hx1 = self.rebnconv1(hxin)
hx = self.pool1(hx1)
hx2 = self.rebnconv2(hx)
hx = self.pool2(hx2)
hx3 = self.rebnconv3(hx)
hx = self.pool3(hx3)
hx4 = self.rebnconv4(hx)
hx = self.pool4(hx4)
hx5 = self.rebnconv5(hx)
hx6 = self.rebnconv6(hx5)
hx5d = self.rebnconv5d(torch.cat((hx6,hx5),1))
hx5dup = _upsample_like(hx5d,hx4)
hx4d = self.rebnconv4d(torch.cat((hx5dup,hx4),1))
hx4dup = _upsample_like(hx4d,hx3)
hx3d = self.rebnconv3d(torch.cat((hx4dup,hx3),1))
hx3dup = _upsample_like(hx3d,hx2)
hx2d = self.rebnconv2d(torch.cat((hx3dup,hx2),1))
hx2dup = _upsample_like(hx2d,hx1)
hx1d = self.rebnconv1d(torch.cat((hx2dup,hx1),1))
return hx1d + hxin
### RSU-5 ###
class RSU5(nn.Module):#UNet05DRES(nn.Module):
def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
super(RSU5,self).__init__()
self.rebnconvin = REBNCONV(in_ch,out_ch,dirate=1)
self.rebnconv1 = REBNCONV(out_ch,mid_ch,dirate=1)
self.pool1 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv2 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool2 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv3 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool3 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv4 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.rebnconv5 = REBNCONV(mid_ch,mid_ch,dirate=2)
self.rebnconv4d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv3d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv2d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv1d = REBNCONV(mid_ch*2,out_ch,dirate=1)
def forward(self,x):
hx = x
hxin = self.rebnconvin(hx)
hx1 = self.rebnconv1(hxin)
hx = self.pool1(hx1)
hx2 = self.rebnconv2(hx)
hx = self.pool2(hx2)
hx3 = self.rebnconv3(hx)
hx = self.pool3(hx3)
hx4 = self.rebnconv4(hx)
hx5 = self.rebnconv5(hx4)
hx4d = self.rebnconv4d(torch.cat((hx5,hx4),1))
hx4dup = _upsample_like(hx4d,hx3)
hx3d = self.rebnconv3d(torch.cat((hx4dup,hx3),1))
hx3dup = _upsample_like(hx3d,hx2)
hx2d = self.rebnconv2d(torch.cat((hx3dup,hx2),1))
hx2dup = _upsample_like(hx2d,hx1)
hx1d = self.rebnconv1d(torch.cat((hx2dup,hx1),1))
return hx1d + hxin
### RSU-4 ###
class RSU4(nn.Module):#UNet04DRES(nn.Module):
def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
super(RSU4,self).__init__()
self.rebnconvin = REBNCONV(in_ch,out_ch,dirate=1)
self.rebnconv1 = REBNCONV(out_ch,mid_ch,dirate=1)
self.pool1 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv2 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.pool2 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.rebnconv3 = REBNCONV(mid_ch,mid_ch,dirate=1)
self.rebnconv4 = REBNCONV(mid_ch,mid_ch,dirate=2)
self.rebnconv3d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv2d = REBNCONV(mid_ch*2,mid_ch,dirate=1)
self.rebnconv1d = REBNCONV(mid_ch*2,out_ch,dirate=1)
def forward(self,x):
hx = x
hxin = self.rebnconvin(hx)
hx1 = self.rebnconv1(hxin)
hx = self.pool1(hx1)
hx2 = self.rebnconv2(hx)
hx = self.pool2(hx2)
hx3 = self.rebnconv3(hx)
hx4 = self.rebnconv4(hx3)
hx3d = self.rebnconv3d(torch.cat((hx4,hx3),1))
hx3dup = _upsample_like(hx3d,hx2)
hx2d = self.rebnconv2d(torch.cat((hx3dup,hx2),1))
hx2dup = _upsample_like(hx2d,hx1)
hx1d = self.rebnconv1d(torch.cat((hx2dup,hx1),1))
return hx1d + hxin
### RSU-4F ###
class RSU4F(nn.Module):#UNet04FRES(nn.Module):
def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
super(RSU4F,self).__init__()
self.rebnconvin = REBNCONV(in_ch,out_ch,dirate=1)
self.rebnconv1 = REBNCONV(out_ch,mid_ch,dirate=1)
self.rebnconv2 = REBNCONV(mid_ch,mid_ch,dirate=2)
self.rebnconv3 = REBNCONV(mid_ch,mid_ch,dirate=4)
self.rebnconv4 = REBNCONV(mid_ch,mid_ch,dirate=8)
self.rebnconv3d = REBNCONV(mid_ch*2,mid_ch,dirate=4)
self.rebnconv2d = REBNCONV(mid_ch*2,mid_ch,dirate=2)
self.rebnconv1d = REBNCONV(mid_ch*2,out_ch,dirate=1)
def forward(self,x):
hx = x
hxin = self.rebnconvin(hx)
hx1 = self.rebnconv1(hxin)
hx2 = self.rebnconv2(hx1)
hx3 = self.rebnconv3(hx2)
hx4 = self.rebnconv4(hx3)
hx3d = self.rebnconv3d(torch.cat((hx4,hx3),1))
hx2d = self.rebnconv2d(torch.cat((hx3d,hx2),1))
hx1d = self.rebnconv1d(torch.cat((hx2d,hx1),1))
return hx1d + hxin
##### U^2-Net ####
class U2NET(nn.Module):
def __init__(self,in_ch=3,out_ch=1):
super(U2NET,self).__init__()
self.stage1 = RSU7(in_ch,32,64)
self.pool12 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage2 = RSU6(64,32,128)
self.pool23 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage3 = RSU5(128,64,256)
self.pool34 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage4 = RSU4(256,128,512)
self.pool45 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage5 = RSU4F(512,256,512)
self.pool56 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage6 = RSU4F(512,256,512)
# decoder
self.stage5d = RSU4F(1024,256,512)
self.stage4d = RSU4(1024,128,256)
self.stage3d = RSU5(512,64,128)
self.stage2d = RSU6(256,32,64)
self.stage1d = RSU7(128,16,64)
self.side1 = nn.Conv2d(64,out_ch,3,padding=1)
self.side2 = nn.Conv2d(64,out_ch,3,padding=1)
self.side3 = nn.Conv2d(128,out_ch,3,padding=1)
self.side4 = nn.Conv2d(256,out_ch,3,padding=1)
self.side5 = nn.Conv2d(512,out_ch,3,padding=1)
self.side6 = nn.Conv2d(512,out_ch,3,padding=1)
self.outconv = nn.Conv2d(6*out_ch,out_ch,1)
def forward(self,x):
hx = x
#stage 1
hx1 = self.stage1(hx)
hx = self.pool12(hx1)
#stage 2
hx2 = self.stage2(hx)
hx = self.pool23(hx2)
#stage 3
hx3 = self.stage3(hx)
hx = self.pool34(hx3)
#stage 4
hx4 = self.stage4(hx)
hx = self.pool45(hx4)
#stage 5
hx5 = self.stage5(hx)
hx = self.pool56(hx5)
#stage 6
hx6 = self.stage6(hx)
hx6up = _upsample_like(hx6,hx5)
#-------------------- decoder --------------------
hx5d = self.stage5d(torch.cat((hx6up,hx5),1))
hx5dup = _upsample_like(hx5d,hx4)
hx4d = self.stage4d(torch.cat((hx5dup,hx4),1))
hx4dup = _upsample_like(hx4d,hx3)
hx3d = self.stage3d(torch.cat((hx4dup,hx3),1))
hx3dup = _upsample_like(hx3d,hx2)
hx2d = self.stage2d(torch.cat((hx3dup,hx2),1))
hx2dup = _upsample_like(hx2d,hx1)
hx1d = self.stage1d(torch.cat((hx2dup,hx1),1))
#side output
d1 = self.side1(hx1d)
d2 = self.side2(hx2d)
d2 = _upsample_like(d2,d1)
d3 = self.side3(hx3d)
d3 = _upsample_like(d3,d1)
d4 = self.side4(hx4d)
d4 = _upsample_like(d4,d1)
d5 = self.side5(hx5d)
d5 = _upsample_like(d5,d1)
d6 = self.side6(hx6)
d6 = _upsample_like(d6,d1)
d0 = self.outconv(torch.cat((d1,d2,d3,d4,d5,d6),1))
return F.sigmoid(d0), F.sigmoid(d1), F.sigmoid(d2), F.sigmoid(d3), F.sigmoid(d4), F.sigmoid(d5), F.sigmoid(d6)
### U^2-Net small ###
class U2NETP(nn.Module):
def __init__(self,in_ch=3,out_ch=1):
super(U2NETP,self).__init__()
self.stage1 = RSU7(in_ch,16,64)
self.pool12 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage2 = RSU6(64,16,64)
self.pool23 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage3 = RSU5(64,16,64)
self.pool34 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage4 = RSU4(64,16,64)
self.pool45 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage5 = RSU4F(64,16,64)
self.pool56 = nn.MaxPool2d(2,stride=2,ceil_mode=True)
self.stage6 = RSU4F(64,16,64)
# decoder
self.stage5d = RSU4F(128,16,64)
self.stage4d = RSU4(128,16,64)
self.stage3d = RSU5(128,16,64)
self.stage2d = RSU6(128,16,64)
self.stage1d = RSU7(128,16,64)
self.side1 = nn.Conv2d(64,out_ch,3,padding=1)
self.side2 = nn.Conv2d(64,out_ch,3,padding=1)
self.side3 = nn.Conv2d(64,out_ch,3,padding=1)
self.side4 = nn.Conv2d(64,out_ch,3,padding=1)
self.side5 = nn.Conv2d(64,out_ch,3,padding=1)
self.side6 = nn.Conv2d(64,out_ch,3,padding=1)
self.outconv = nn.Conv2d(6*out_ch,out_ch,1)
def forward(self,x):
hx = x
#stage 1
hx1 = self.stage1(hx)
hx = self.pool12(hx1)
#stage 2
hx2 = self.stage2(hx)
hx = self.pool23(hx2)
#stage 3
hx3 = self.stage3(hx)
hx = self.pool34(hx3)
#stage 4
hx4 = self.stage4(hx)
hx = self.pool45(hx4)
#stage 5
hx5 = self.stage5(hx)
hx = self.pool56(hx5)
#stage 6
hx6 = self.stage6(hx)
hx6up = _upsample_like(hx6,hx5)
#decoder
hx5d = self.stage5d(torch.cat((hx6up,hx5),1))
hx5dup = _upsample_like(hx5d,hx4)
hx4d = self.stage4d(torch.cat((hx5dup,hx4),1))
hx4dup = _upsample_like(hx4d,hx3)
hx3d = self.stage3d(torch.cat((hx4dup,hx3),1))
hx3dup = _upsample_like(hx3d,hx2)
hx2d = self.stage2d(torch.cat((hx3dup,hx2),1))
hx2dup = _upsample_like(hx2d,hx1)
hx1d = self.stage1d(torch.cat((hx2dup,hx1),1))
#side output
d1 = self.side1(hx1d)
d2 = self.side2(hx2d)
d2 = _upsample_like(d2,d1)
d3 = self.side3(hx3d)
d3 = _upsample_like(d3,d1)
d4 = self.side4(hx4d)
d4 = _upsample_like(d4,d1)
d5 = self.side5(hx5d)
d5 = _upsample_like(d5,d1)
d6 = self.side6(hx6)
d6 = _upsample_like(d6,d1)
d0 = self.outconv(torch.cat((d1,d2,d3,d4,d5,d6),1))
return F.sigmoid(d0)