Why you choose the Convolution 1d layer with input feature channel(16,1024) and output(1,1024)?
What is your relevant formula?.
I look forward to hearing from you soon. Thank you so much
The corresponding equations are (7) and (8) in https://arxiv.org/pdf/2212.02967. Note that the features are computed for every RIS element. Therefore, a conv layer with kernel size of 1 can be applied.
Why you choose the Convolution 1d layer with input feature channel(16,1024) and output(1,1024)? What is your relevant formula?. I look forward to hearing from you soon. Thank you so much