RuntimeError when training resnext50_32x4d

[NightQing]

Hi,
I have trained a resnet50 model successfully, but when I train resnext50_32x4d, there is an error:
models_lpf/resnet.py", line 147, in forward
out += identity
RuntimeError: The size of tensor a (20) must match the size of tensor b (80) at non-singleton dimension 3.

In addition, in models_lpf/resnet.py, "groups=4, width_per_group=32" in resnext32x4d is differenet from "groups=32, width_per_group=4" in pytorch offical code "torchvision/models/resnet.py"
Do you have any advices?

[honglu-intel]

just run into the exact sample problem, any update?

I tried switching the group and width_pre_group parameters. It messed up the network dimension even further. So that is not the solution.

Also, I noticed that the model size of this implementation is about 4x of the one built by torchvision, so it seems some dim setup is off by 4x.

[honglu-intel]

Just saw this issue is closed. Any updates? Thx much. From: BalalaXiaomoxia <notifications@github.com> Reply-To: adobe/antialiased-cnns <reply@reply.github.com> Date: Friday, May 22, 2020 at 10:39 AM To: adobe/antialiased-cnns <antialiased-cnns@noreply.github.com> Cc: Hong Lu <hong.lu@intel.com>, Comment <comment@noreply.github.com> Subject: Re: [adobe/antialiased-cnns] RuntimeError when training resnext50_32x4d (#33) Closed #33<#33>. — You are receiving this because you commented. Reply to this email directly, view it on GitHub<#33 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACL5KRCMR5XJHZJ5HZ644GDRS22E7ANCNFSM4NC6LUBA>.