IDSGAN RunTime error

1

Hi !
I’m new to pytorch and try to run IDSGAN. (https://github.com/imoken1122/IDSGAN)
I got RunTime error(below) during running IDS-WGAN.py

Traceback (most recent call last):
File “C:/논문/IDSGAN-master/IDSGAN-master/IDS_WGAN.py”, line 47, in
ids_model.load_state_dict(param)
File “C:\anaconda3\envs\tutorial\lib\site-packages\torch\nn\modules\module.py”, line 830, in load_state_dict
self.class.name, “\n\t”.join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for Blackbox_IDS:
size mismatch for layer.6.weight: copying a param with shape torch.Size([33, 66]) from checkpoint, the shape in current model is torch.Size([16, 66]).
size mismatch for layer.6.bias: copying a param with shape torch.Size([33]) from checkpoint, the shape in current model is torch.Size([16]).
size mismatch for layer.9.weight: copying a param with shape torch.Size([16, 33]) from checkpoint, the shape in current model is torch.Size([16, 16]).

Waiting for your help. Thanks.

2

The model from which you are loading the checkpoint from and the model that you are currently using differ in certain layers as mentioned in the traceback. I just had a look at the model. Did you tweak the input to the model instantiation? I have pasted some code from the tutorial below to clarify the question.

D_G_INPUT_DIM = len(train_data.columns) → Are you using different train data?
ids_model = Blackbox_IDS(D_G_INPUT_DIM,2) → What is your D_G_INPUT_DIM?
param = th.load(‘save_model/IDS.pth’)
ids_model.load_state_dict(param)