train problrm with pre_model

[successhaha]

When I train CUB with pre_model that you offer.
Runtime Error:Error(s) in loading state_dict for WideResNet:
Missing key(s) in state_dict: "conv1.weight", "block1.layer.0.bn1.weight", "block1.layer.0.bn1.bias", "block1.layer.0.bn1.running_mean", "block1.layer.0.bn1.running_var", "block1.layer.0.conv1.weight", "block1.layer.0.bn2.weight", "block1.layer.0.bn2.bias", "block1.layer.0.bn2.running_mean",

[ThatNight]

python train.py --dataset miniImagenet --method S2M2_R --model WideResNet28_10 --train_aug
When I use this command to train, the program reports an error: the resume_file in torch.load(resume_file) is empty.
Why do I need to load the checkpoint file when I train? How can I train it?Looking forward to your reply!

[monsters-s]

python train.py --dataset miniImagenet --method S2M2_R --model WideResNet28_10 --train_aug
When I use this command to train, the program reports an error: the resume_file in torch.load(resume_file) is empty.
Why do I need to load the checkpoint file when I train? How can I train it?Looking forward to your reply!

@ThatNight
I think the code is similar to S2M2_fewshot and 'S2M2_R' need train on the 'rotation' method.So you need first train on method 'rotation' then train on method 'S2M2_R'.
I don't know whether I'm right, but I follow it and it can run correctly.