search

hughperkins / DeepCL

OpenCL library to train deep convolutional neural networks
Mozilla Public License 2.0
866 stars 200 forks source link

How to set stride of conv layer? #153

Open glgoo opened 4 years ago

glgoo commented 4 years ago

I must use conv kernel 3x3 with stride 2, pooling kernel 2x2 with stride 2. But, xxxxLayerMaker without stride option. So, how to setup? :)

hughperkins commented 4 years ago

I believe that stride 1 is the default?

On Wed, Jul 22, 2020, 05:32 Willem Chen notifications@github.com wrote:

I must use conv kernel 3x3 with stride 1 So how to setup?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/hughperkins/DeepCL/issues/153, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAA6FKEPIW5OEEHXNBZTEETR42WZPANCNFSM4PEQOY2A .

glgoo commented 4 years ago

I believe that stride 1 is the default? … On Wed, Jul 22, 2020, 05:32 Willem Chen @.***> wrote: I must use conv kernel 3x3 with stride 1 So how to setup? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#153>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAA6FKEPIW5OEEHXNBZTEETR42WZPANCNFSM4PEQOY2A .

yeah, but I want to set stride 2. I use net->print(), then it show conv layer skip=0. I cannot find the variable. :)