The size of 'out' and 'rfp_feat' is not match

[sunpeng96]

Thanks for your error report and we appreciate it a lot.

in resnet.py 305 out = out + rfp_feat

The size of 'out' and 'rfp_feat' is not match

Through debugging, I found that the problem lies in resnet.py lines of 607 RFP_ forward(). x = layer.rfp_ forward(x, rfp_ In feat), layer does not operate on img, resulting in x is tensor(img) in this statement. Finally, in the RFP_ In forward (), lead to The size of 'out' and 'rfp_feat' is not match.

In fact, I didn't make any changes to the program. I just called your resnet.py in my program.

I don't know if it's a version problem? We look forward to your reply.

[joe-siyuan-qiao]

Hi @sunpeng96

If the size doesn't match, it's likely that your input size does not satisfy the requirement of the backbone. You can start debugging from this.