The class implements a layer that divides its first input by the second element by element.
This layer has no settings.
This layer has no trainable parameters.
The layer should two inputs. The blobs of those inputs should have the same dimensions.
The layer has one output of the same size as each of the inputs. The output blob contains the element-wise division of the input blobs.