zhang xu

Please show me the screenshot of your tensorboard. [email protected]

(1) Yeah, when the backbone network is resnet, OctaveConv only needs OctaveUpsampleLayer. OctaveConv does not change the way the convolution is done. Convolution will only differ depending on the backbone...

@A-suozhang thanks

1.this project no relationship with the paper—“MOON”. 2.Why use the loss function —“EuclideanLoss”？It is better use “Sigmoid Cross-Entropy” or “HingeLoss”？