yzspku/CQTNet

How to solve the Augment.position problem

Opened this issue · 2 comments

in CQTNet, there is no Augment.position layer, but when i load the state of the pretrained model, it will occur wrong msg,

yzspku commented

in CQTNet, there is no Augment.position layer, but when i load the state of the pretrained model, it will occur wrong msg,

This problem has not been encountered before, could you provide more details?

Because I can't reproduce the model you have in the google disk, that model has specAug data augmentation, can you provide me with a relevant training code? my email: 2662345338@qq.com