incompatibility in code and article

[maralzar]

In your article after concatenating Q and output of the transfer attention layer, we see two convolution layers as illustrated in Fig 2
,however, In code you applied one Conv why do not they match?
T=self.conv1[i](T)

[chunmeifeng]

Hi Maral, Thanks for your email. We are sorry that the paper is inconsistent with the code. However, there's no big difference between one convolution layer and two layers. Maral ***@***.***> 于2022年8月22日周一 23:32写道：

…

In your article after concatenating Q and output of the transfer attention layer, we see two convolution layers as illustrated in Fig 2 ,however, In code you applied one Conv why do not match? T=self.conv1[i](T) — Reply to this email directly, view it on GitHub <#12>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AR75XN6RMROCRATB4A3M75LV2PWXRANCNFSM57JFIUOQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[maralzar]

Thank you for the quick response and also amazing research. So the results of the articles are based on two convs ?
??

[chunmeifeng]

All the results are obtained from our released code. Maral ***@***.***> 于2022年8月23日周二 00:36写道：

…

Thank you for the quick response and also amazing research. So the results of the articles are based on two convs ? ?? — Reply to this email directly, view it on GitHub <#12 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AR75XN3IJKL45K7IJX34GE3V2P6FRANCNFSM57JFIUOQ> . You are receiving this because you commented.Message ID: ***@***.***>

[maralzar]

Thank you.