You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
PyramidNet's residual block(d) is not the same as pre-resnet's(a). It deletes the first ReLU and adds a new BN at the end.
I noticed that in the pytorch version pyramidnet.py, the model is built just on pre_conv1x1_block.
Waitting for your verification.
The text was updated successfully, but these errors were encountered:
PyramidNet's residual block(d) is not the same as pre-resnet's(a). It deletes the first ReLU and adds a new BN at the end.
I noticed that in the pytorch version pyramidnet.py, the model is built just on pre_conv1x1_block.
Waitting for your verification.
The text was updated successfully, but these errors were encountered: