About the BN!!!
zyxu1996 opened this issue · comments
I add other block to replace EMAU, but get some warning. I guess it's bn_lib you used not suitable for my block.
`2020-07-30 20:05:26,727 - INFO - step: 1 loss: 2.429 lr: 0.009
WARNING batched routines are designed for small sizes. It might be better to use the
Native/Hybrid classical routines if you want good performance.
=========================================================================================
WARNING batched routines are designed for small sizes. It might be better to use the
Native/Hybrid classical routines if you want good performance.
=========================================================================================
WARNING batched routines are designed for small sizes. It might be better to use the
Native/Hybrid classical routines if you want good performance.
=========================================================================================
WARNING batched routines are designed for small sizes. It might be better to use the
Native/Hybrid classical routines if you want good performance.
2020-07-30 20:05:29,586 - INFO - step: 2 loss: 2.398 lr: 0.009`
Yeap, the BN used in my repo is not the best choice.
As we may have no plan to update it, I suggest you replace it with Pytorch's official version of SyncBN