关于提供的模型文件RFB512_E_34_4.pth

[achillesli]

你好，我在使用RFB512_E_34_4.pth的时候，
'--version', default=RFB_E_vgg ，'--size', default='512'，cfg = COCO_512，但载入参数的时候有错误，是哪里我忽略了还没选好么？报错如下：

RuntimeError: Error(s) in loading state_dict for RFBNet:
Missing key(s) in state_dict: "Norm.branch3.3.conv.weight", "Norm.branch3.3.bn.bias", "Norm.branch3.3.bn.running_mean", "Norm.branch3.3.bn.running_var", "Norm.branch3.3.bn.weight", "extras.3.conv.weight", "extras.3.bn.bias", "extras.3.bn.running_mean", "extras.3.bn.running_var", "extras.3.bn.weight", "extras.4.conv.weight", "extras.4.bn.bias", "extras.4.bn.running_mean", "extras.4.bn.running_var", "extras.4.bn.weight".
Unexpected key(s) in state_dict: "reduce.conv.weight", "reduce.bn.weight", "reduce.bn.bias", "reduce.bn.running_mean", "reduce.bn.running_var", "up_reduce.conv.weight", "up_reduce.bn.weight", "up_reduce.bn.bias", "up_reduce.bn.running_mean", "up_reduce.bn.running_var", "Norm.branch4.0.conv.weight", "Norm.branch4.0.bn.weight", "Norm.branch4.0.bn.bias", "Norm.branch4.0.bn.running_mean", "Norm.branch4.0.bn.running_var", "Norm.branch4.1.conv.weight", "Norm.branch4.1.bn.weight", "Norm.branch4.1.bn.bias", "Norm.branch4.1.bn.running_mean", "Norm.branch4.1.bn.running_var", "Norm.branch4.2.conv.weight", "Norm.branch4.2.bn.weight", "Norm.branch4.2.bn.bias", "Norm.branch4.2.bn.running_mean", "Norm.branch4.2.bn.running_var", "Norm.branch5.0.conv.weight", "Norm.branch5.0.bn.weight", "Norm.branch5.0.bn.bias", "Norm.branch5.0.bn.running_mean", "Norm.branch5.0.bn.running_var", "Norm.branch5.1.conv.weight", "Norm.branch5.1.bn.weight", "Norm.branch5.1.bn.bias", "Norm.branch5.1.bn.running_mean", "Norm.branch5.1.bn.running_var", "Norm.branch5.2.conv.weight", "Norm.branch5.2.bn.weight", "Norm.branch5.2.bn.bias", "Norm.branch5.2.bn.running_mean", "Norm.branch5.2.bn.running_var", "Norm.branch5.3.conv.weight", "Norm.branch5.3.bn.weight", "Norm.branch5.3.bn.bias", "Norm.branch5.3.bn.running_mean", "Norm.branch5.3.bn.running_var", "Norm.branch6.0.conv.weight", "Norm.branch6.0.bn.weight", "Norm.branch6.0.bn.bias", "Norm.branch6.0.bn.running_mean", "Norm.branch6.0.bn.running_var", "Norm.branch6.1.conv.weight", "Norm.branch6.1.bn.weight", "Norm.branch6.1.bn.bias", "Norm.branch6.1.bn.running_mean", "Norm.branch6.1.bn.running_var", "Norm.branch6.2.conv.weight", "Norm.branch6.2.bn.weight", "Norm.branch6.2.bn.bias", "Norm.branch6.2.bn.running_mean", "Norm.branch6.2.bn.running_var", "Norm.branch6.3.conv.weight", "Norm.branch6.3.bn.weight", "Norm.branch6.3.bn.bias", "Norm.branch6.3.bn.running_mean", "Norm.branch6.3.bn.running_var", "extras.0.branch3.0.conv.weight", "extras.0.branch3.0.bn.weight", "extras.0.branch3.0.bn.bias", "extras.0.branch3.0.bn.running_mean", "extras.0.branch3.0.bn.running_var", "extras.0.branch3.1.conv.weight", "extras.0.branch3.1.bn.weight", "extras.0.branch3.1.bn.bias", "extras.0.branch3.1.bn.running_mean", "extras.0.branch3.1.bn.running_var", "extras.0.branch3.2.conv.weight", "extras.0.branch3.2.bn.weight", "extras.0.branch3.2.bn.bias", "extras.0.branch3.2.bn.running_mean", "extras.0.branch3.2.bn.running_var", "extras.0.branch3.3.conv.weight", "extras.0.branch3.3.bn.weight", "extras.0.branch3.3.bn.bias", "extras.0.branch3.3.bn.running_mean", "extras.0.branch3.3.bn.running_var", "extras.1.branch3.0.conv.weight", "extras.1.branch3.0.bn.weight", "extras.1.branch3.0.bn.bias", "extras.1.branch3.0.bn.running_mean", "extras.1.branch3.0.bn.running_var", "extras.1.branch3.1.conv.weight", "extras.1.branch3.1.bn.weight", "extras.1.branch3.1.bn.bias", "extras.1.branch3.1.bn.running_mean", "extras.1.branch3.1.bn.running_var", "extras.1.branch3.2.conv.weight", "extras.1.branch3.2.bn.weight", "extras.1.branch3.2.bn.bias", "extras.1.branch3.2.bn.running_mean", "extras.1.branch3.2.bn.running_var", "extras.1.branch3.3.conv.weight", "extras.1.branch3.3.bn.weight", "extras.1.branch3.3.bn.bias", "extras.1.branch3.3.bn.running_mean", "extras.1.branch3.3.bn.running_var", "extras.2.branch3.0.conv.weight", "extras.2.branch3.0.bn.weight", "extras.2.branch3.0.bn.bias", "extras.2.branch3.0.bn.running_mean", "extras.2.branch3.0.bn.running_var", "extras.2.branch3.1.conv.weight", "extras.2.branch3.1.bn.weight", "extras.2.branch3.1.bn.bias", "extras.2.branch3.1.bn.running_mean", "extras.2.branch3.1.bn.running_var", "extras.2.branch3.2.conv.weight", "extras.2.branch3.2.bn.weight", "extras.2.branch3.2.bn.bias", "extras.2.branch3.2.bn.running_mean", "extras.2.branch3.2.bn.running_var", "extras.2.branch3.3.conv.weight", "extras.2.branch3.3.bn.weight", "extras.2.branch3.3.bn.bias", "extras.2.branch3.3.bn.running_mean", "extras.2.branch3.3.bn.running_var", "extras.3.branch0.0.conv.weight", "extras.3.branch0.0.bn.weight", "extras.3.branch0.0.bn.bias", "extras.3.branch0.0.bn.running_mean", "extras.3.branch0.0.bn.running_var", "extras.3.branch0.1.conv.weight", "extras.3.branch0.1.bn.weight", "extras.3.branch0.1.bn.bias", "extras.3.branch0.1.bn.running_mean", "extras.3.branch0.1.bn.running_var", "extras.3.branch1.0.conv.weight", "extras.3.branch1.0.bn.weight", "extras.3.branch1.0.bn.bias", "extras.3.branch1.0.bn.running_mean", "extras.3.branch1.0.bn.running_var", "extras.3.branch1.1.conv.weight", "extras.3.branch1.1.bn.weight", "extras.3.branch1.1.bn.bias", "extras.3.branch1.1.bn.running_mean", "extras.3.branch1.1.bn.running_var", "extras.3.branch1.2.conv.weight", "extras.3.branch1.2.bn.weight", "extras.3.branch1.2.bn.bias", "extras.3.branch1.2.bn.running_mean", "extras.3.branch1.2.bn.running_var", "extras.3.branch2.0.conv.weight", "extras.3.branch2.0.bn.weight", "extras.3.branch2.0.bn.bias", "extras.3.branch2.0.bn.running_mean", "extras.3.branch2.0.bn.running_var", "extras.3.branch2.1.conv.weight", "extras.3.branch2.1.bn.weight", "extras.3.branch2.1.bn.bias", "extras.3.branch2.1.bn.running_mean", "extras.3.branch2.1.bn.running_var", "extras.3.branch2.2.conv.weight", "extras.3.branch2.2.bn.weight", "extras.3.branch2.2.bn.bias", "extras.3.branch2.2.bn.running_mean", "extras.3.branch2.2.bn.running_var", "extras.3.branch2.3.conv.weight", "extras.3.branch2.3.bn.weight", "extras.3.branch2.3.bn.bias", "extras.3.branch2.3.bn.running_mean", "extras.3.branch2.3.bn.running_var", "extras.3.ConvLinear.conv.weight", "extras.3.ConvLinear.bn.weight", "extras.3.ConvLinear.bn.bias", "extras.3.ConvLinear.bn.running_mean", "extras.3.ConvLinear.bn.running_var", "extras.3.shortcut.conv.weight", "extras.3.shortcut.bn.weight", "extras.3.shortcut.bn.bias", "extras.3.shortcut.bn.running_mean", "extras.3.shortcut.bn.running_var", "extras.4.branch0.0.conv.weight", "extras.4.branch0.0.bn.weight", "extras.4.branch0.0.bn.bias", "extras.4.branch0.0.bn.running_mean", "extras.4.branch0.0.bn.running_var", "extras.4.branch0.1.conv.weight", "extras.4.branch0.1.bn.weight", "extras.4.branch0.1.bn.bias", "extras.4.branch0.1.bn.running_mean", "extras.4.branch0.1.bn.running_var", "extras.4.branch1.0.conv.weight", "extras.4.branch1.0.bn.weight", "extras.4.branch1.0.bn.bias", "extras.4.branch1.0.bn.running_mean", "extras.4.branch1.0.bn.running_var", "extras.4.branch1.1.conv.weight", "extras.4.branch1.1.bn.weight", "extras.4.branch1.1.bn.bias", "extras.4.branch1.1.bn.running_mean", "extras.4.branch1.1.bn.running_var", "extras.4.branch1.2.conv.weight", "extras.4.branch1.2.bn.weight", "extras.4.branch1.2.bn.bias", "extras.4.branch1.2.bn.running_mean", "extras.4.branch1.2.bn.running_var", "extras.4.branch2.0.conv.weight", "extras.4.branch2.0.bn.weight", "extras.4.branch2.0.bn.bias", "extras.4.branch2.0.bn.running_mean", "extras.4.branch2.0.bn.running_var", "extras.4.branch2.1.conv.weight", "extras.4.branch2.1.bn.weight", "extras.4.branch2.1.bn.bias", "extras.4.branch2.1.bn.running_mean", "extras.4.branch2.1.bn.running_var", "extras.4.branch2.2.conv.weight", "extras.4.branch2.2.bn.weight", "extras.4.branch2.2.bn.bias", "extras.4.branch2.2.bn.running_mean", "extras.4.branch2.2.bn.running_var", "extras.4.branch2.3.conv.weight", "extras.4.branch2.3.bn.weight", "extras.4.branch2.3.bn.bias", "extras.4.branch2.3.bn.running_mean", "extras.4.branch2.3.bn.running_var", "extras.4.ConvLinear.conv.weight", "extras.4.ConvLinear.bn.weight", "extras.4.ConvLinear.bn.bias", "extras.4.ConvLinear.bn.running_mean", "extras.4.ConvLinear.bn.running_var", "extras.4.shortcut.conv.weight", "extras.4.shortcut.bn.weight", "extras.4.shortcut.bn.bias", "extras.4.shortcut.bn.running_mean", "extras.4.shortcut.bn.running_var", "loc.6.weight", "loc.6.bias", "conf.6.weight", "conf.6.bias".
While copying the parameter named "Norm.branch0.0.conv.weight", whose dimensions in the model are torch.Size([128, 512, 1, 1]) and whose dimensions in the checkpoint are torch.Size([64, 512, 1, 1]).
While copying the parameter named "Norm.branch0.0.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch0.0.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch0.0.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch0.0.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch0.1.conv.weight", whose dimensions in the model are torch.Size([128, 128, 3, 3]) and whose dimensions in the checkpoint are torch.Size([64, 64, 3, 3]).
While copying the parameter named "Norm.branch0.1.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch0.1.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch0.1.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch0.1.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.0.conv.weight", whose dimensions in the model are torch.Size([128, 512, 1, 1]) and whose dimensions in the checkpoint are torch.Size([64, 512, 1, 1]).
While copying the parameter named "Norm.branch1.0.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.0.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.0.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.0.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.1.conv.weight", whose dimensions in the model are torch.Size([128, 128, 3, 1]) and whose dimensions in the checkpoint are torch.Size([64, 64, 3, 1]).
While copying the parameter named "Norm.branch1.1.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.1.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.1.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.1.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.2.conv.weight", whose dimensions in the model are torch.Size([128, 128, 3, 3]) and whose dimensions in the checkpoint are torch.Size([64, 64, 3, 3]).
While copying the parameter named "Norm.branch1.2.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.2.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.2.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch1.2.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.0.conv.weight", whose dimensions in the model are torch.Size([128, 512, 1, 1]) and whose dimensions in the checkpoint are torch.Size([64, 512, 1, 1]).
While copying the parameter named "Norm.branch2.0.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.0.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.0.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.0.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.1.conv.weight", whose dimensions in the model are torch.Size([128, 128, 1, 3]) and whose dimensions in the checkpoint are torch.Size([64, 64, 1, 3]).
While copying the parameter named "Norm.branch2.1.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.1.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.1.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.1.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.2.conv.weight", whose dimensions in the model are torch.Size([128, 128, 3, 3]) and whose dimensions in the checkpoint are torch.Size([64, 64, 3, 3]).
While copying the parameter named "Norm.branch2.2.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.2.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.2.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch2.2.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.1.conv.weight", whose dimensions in the model are torch.Size([96, 64, 1, 3]) and whose dimensions in the checkpoint are torch.Size([64, 64, 3, 1]).
While copying the parameter named "Norm.branch3.1.bn.bias", whose dimensions in the model are torch.Size([96]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.1.bn.running_mean", whose dimensions in the model are torch.Size([96]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.1.bn.running_var", whose dimensions in the model are torch.Size([96]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.1.bn.weight", whose dimensions in the model are torch.Size([96]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.2.conv.weight", whose dimensions in the model are torch.Size([128, 96, 3, 1]) and whose dimensions in the checkpoint are torch.Size([64, 64, 3, 3]).
While copying the parameter named "Norm.branch3.2.bn.bias", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.2.bn.running_mean", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.2.bn.running_var", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.branch3.2.bn.weight", whose dimensions in the model are torch.Size([128]) and whose dimensions in the checkpoint are torch.Size([64]).
While copying the parameter named "Norm.ConvLinear.conv.weight", whose dimensions in the model are torch.Size([512, 512, 1, 1]) and whose dimensions in the checkpoint are torch.Size([512, 448, 1, 1]).
While copying the parameter named "extras.0.ConvLinear.conv.weight", whose dimensions in the model are torch.Size([1024, 768, 1, 1]) and whose dimensions in the checkpoint are torch.Size([1024, 1024, 1, 1]).
While copying the parameter named "extras.1.ConvLinear.conv.weight", whose dimensions in the model are torch.Size([512, 768, 1, 1]) and whose dimensions in the checkpoint are torch.Size([512, 1024, 1, 1]).
While copying the parameter named "extras.2.ConvLinear.conv.weight", whose dimensions in the model are torch.Size([256, 384, 1, 1]) and whose dimensions in the checkpoint are torch.Size([256, 512, 1, 1]).
While copying the parameter named "extras.6.conv.weight", whose dimensions in the model are torch.Size([256, 128, 3, 3]) and whose dimensions in the checkpoint are torch.Size([256, 128, 4, 4]).
While copying the parameter named "loc.4.bias", whose dimensions in the model are torch.Size([16]) and whose dimensions in the checkpoint are torch.Size([24]).
While copying the parameter named "loc.4.weight", whose dimensions in the model are torch.Size([16, 256, 3, 3]) and whose dimensions in the checkpoint are torch.Size([24, 256, 3, 3]).
While copying the parameter named "conf.0.bias", whose dimensions in the model are torch.Size([126]) and whose dimensions in the checkpoint are torch.Size([486]).
While copying the parameter named "conf.0.weight", whose dimensions in the model are torch.Size([126, 512, 3, 3]) and whose dimensions in the checkpoint are torch.Size([486, 512, 3, 3]).
While copying the parameter named "conf.1.bias", whose dimensions in the model are torch.Size([126]) and whose dimensions in the checkpoint are torch.Size([486]).
While copying the parameter named "conf.1.weight", whose dimensions in the model are torch.Size([126, 1024, 3, 3]) and whose dimensions in the checkpoint are torch.Size([486, 1024, 3, 3]).
While copying the parameter named "conf.2.bias", whose dimensions in the model are torch.Size([126]) and whose dimensions in the checkpoint are torch.Size([486]).
While copying the parameter named "conf.2.weight", whose dimensions in the model are torch.Size([126, 512, 3, 3]) and whose dimensions in the checkpoint are torch.Size([486, 512, 3, 3]).
While copying the parameter named "conf.3.bias", whose dimensions in the model are torch.Size([126]) and whose dimensions in the checkpoint are torch.Size([486]).
While copying the parameter named "conf.3.weight", whose dimensions in the model are torch.Size([126, 256, 3, 3]) and whose dimensions in the checkpoint are torch.Size([486, 256, 3, 3]).
While copying the parameter named "conf.4.bias", whose dimensions in the model are torch.Size([84]) and whose dimensions in the checkpoint are torch.Size([486]).
While copying the parameter named "conf.4.weight", whose dimensions in the model are torch.Size([84, 256, 3, 3]) and whose dimensions in the checkpoint are torch.Size([486, 256, 3, 3]).
While copying the parameter named "conf.5.bias", whose dimensions in the model are torch.Size([84]) and whose dimensions in the checkpoint are torch.Size([324]).
While copying the parameter named "conf.5.weight", whose dimensions in the model are torch.Size([84, 256, 3, 3]) and whose dimensions in the checkpoint are torch.Size([324, 256, 3, 3]).

[dazhangzhang]

@achillesli hello,so, this problem has be solved?

[DonghoonPark12]

@achillesli Did you solve the problem??