Radeon ProRender


Parameters for the RML_OP_LOCAL_RESPONSE_NORM operation.


Graph Manipulation


Name Type Description
input rml_op Operation that outputs tensor containing input tensor data.
size uint32_t The number of channels to sum over (for cross channel LRN) or the side length of the square region to sum over (for within channel LRN).
alpha float Scaling parameter.
beta float Exponent parameter.
bias float Bias parameter.
cross_channel rml_bool Whether to sum over adjacent channels (for cross channel LRN) or nearby spatial locations (for within channel LRN).