Using "padding = same " in a dau_conv2d_tf layer ??

[cloudscapes]

Hi,
Can I use "padding = same" in a dau_conv2d_tf layer the way one can usually use it in a regular Keras/tensorflow layer(Conv2D(filters,(num_row, num_col), strides=strides, padding="same")) ?

cheers,
H

[skokec]

Hi,

For use in Keras you need to use the class version DAUConv2dTF(filters, dau_units, max_kernel_size), which only works as padding = same and cannot be changed. There is actually no padding parameter at all.

Best,
Domen

[cloudscapes]

Thank you very much for your quick response Domen :)
So if I understand you correctly, I don't even need to specify to the DAUCOnv2dTF layer the padding = 'same' argument? Cause it already does that implicitly(default behavior)?
cheers,
H

[skokec]

That is correct.

Best,
Domen