This class implements a layer that concatenates the input objects along the Height
, Width
, Depth
, and Channels
dimensions.
This layer has no settings.
This layer has no trainable parameters.
The layer accepts an arbitrary number of inputs, each containing a blob with data:
BatchLength
,BatchWidth
, andListSize
dimensions must be the same for all inputs.- other dimensions may vary.
The layer has one output which contains a blob with the result of concatenation. The dimensions of the blob are:
BatchLength
,BatchWidth
, andListSize
are equal to the corresponding inputs' dimensions.Height
,Width
, andDepth
are equal to1
.Channels
is equal to the sum ofHeight * Width * Depth * Channels
over all the inputs.