The class implements a layer that adds up its inputs element by element.
This layer has no settings.
This layer has no trainable parameters.
This layer has 2 inputs.
The first input accepts a blob containing BatchLength * BatchWidth * ListSize
objects of size Height * Width * Depth * Channels
.
The second input accepts a blob of the dimensions:
Height
,Width
,Depth
andChannels
are equal to these dimensions of the first input- all other dimensions are equal to
1
The single output returns a blob of size equal to the first input. This blob contains the sums of the corresponding objects of the first input with the contents of the second input.