some errors about loading pretrained model

[qk-huang]

Hi, when I loaded the detectron_100_resnet_most_data model, some problems happened:

While copying the parameter named "module.image_embedding_models_list.0.0.image_attention_model.modal_combine.Fa_image.main.0.weight_g", whose dimensions in the model are torch.Size([]) and whose dimensions in the checkpoint are torch.Size([1]). While copying the parameter named "module.image_embedding_models_list.0.0.image_attention_model.modal_combine.Fa_txt.main.0.weight_g", whose dimensions in the model are torch.Size([]) and whose dimensions in the checkpoint are torch.Size([1]).

I think might weight_norm caused it, but I don't know the reason. Can you give me some advices.

[YuJiang01]

@Ken-Con which version of pytorch are you using?

[qk-huang]

@YuJiang01 pytorch 0.4.0. I have found the problem, because my pytorch creates a different type for weight_norm parameter. Just need to convert single element tensor to scalar type by use a = a[0].