cfg_kernelsize

[tehreemnaqvi]

Hi,

I am referring to your code. My question is how you are taking cfg_kernel_size?

I didn't get the forward function, how you kept cfg_kernel_size? Below is the snippet of your forward function.

Can you please tell me how did you do that?

torch.zeros(batch_size, cfg_cnn[0][1], cfg_kernel[0], cfg_kernel[0], device=device))

Thanks

[nguyenhongson1902]

@tehreemnaqvi, I have the same problems when reading the code

[QianpengLi577]

@tehreemnaqvi
Hi, input image size is [batch_size, 28, 28]

if conv1 is 3*3 kernel ,4 channel(for example) and padding ,then the c1_mem size is [batch_size, 4, 28, 28]

if conv1 is no padding ,then the c1_mem size is [batch_size, 4 ,26, 26]

thus cfg_cnn[ i ][ 1 ] means the channels of i^{th} conv layer

cfg_kernel[0] means the picture size of the first layer output

cfg_kernel[1] means the picture size of the second layer output