I was studying convolutional neural networks, and creating a module. While using nn.conv2d, I can't figure out exactly how to specify my out_channels.
I understand that it is the number of filters that are going to be used. For a black and white picture, (meaning in_channels to be 1), if I put 8 output_channels, does it mean that I will get 8 different feature maps?
Also if I increase my input_channel to 3, let's say RGB, does this mean that I will get total of 24 (3*8 = 24) feature maps, 8 red, 8 green, 8 black?