in conv2d_mtl.py, the mtl_weight has 4-dimensional and the weight has 3-dimensional, new_mtl_weight = self.mtl_weight.expand(self.weight.shape) has error,is there some problem?

Thanks for your interest in our work!

I think you are referring to the following line:

meta-transfer-learning/pytorch/models/conv2d_mtl.py

Line 94 in 5c12cc4

new_mtl_weight = self.mtl_weight.expand(self.weight.shape)

It works well when I am running this project. So I am not sure why there is an error when running this line.
Could you please provide more information? Maybe I can try to find a solution based on your case.
The related information includes your python and PyTorch version, the error log, and so on.

I am looking forward to your reply.

Have a nice day.

Best,
Yaoyao