centernet-pytorch icon indicating copy to clipboard operation
centernet-pytorch copied to clipboard

KeyError: 'backbone.0.weight'

Open JiangZQYW opened this issue 3 years ago • 3 comments

使用的是pytorch版本的centernet代码,主干特征提取网络是resnet50,主干网没动。 在代码: self.backbone = resnet50(pretrain=pretrain) self.decoder = resnet50_Decoder(2048) self.head = resnet50_Head(channel=64, num_classes=num_classes) 中,self.backbone = resnet50(pretrain=pretrain)没动,只更改了self.decoder ,想要迁移使用model_data/centernet_resnet50_voc.pth中的resnet50权重,下面的权重自己训练。 更改了pretrain = True。

却出现错误如下: pretrained_dict = {k: v for k, v in pretrained_dict.items() if np.shape(model_dict[k]) == np.shape(v)} KeyError: 'backbone.0.weight' 我的邮箱:[email protected] 谢谢!

JiangZQYW avatar May 03 '21 11:05 JiangZQYW

已解决,谢谢

JiangZQYW avatar May 04 '21 14:05 JiangZQYW

好的加油

bubbliiiing avatar May 07 '21 06:05 bubbliiiing

@JiangZQYW 具体怎么解决,能否参考一下你的代码 我的邮箱:[email protected] 谢谢!

chenpeng68 avatar Jul 07 '21 08:07 chenpeng68