Radeon ProRender


Parameters for RML_OP_SPACE_TO_DEPTH operation.

RML_OP_SPACE_TO_DEPTH operation rearranges blocks of spatial tensor data into channels. More specifically, this operation outputs a copy of the input tensor where values from the height and width dimensions are moved to the channel dimension. E.g., rearranges tensor of shape (N, C, H * bs, W * bs) to a tensor of shape (N, C * bs^2, H, W).


Graph Manipulation


Name Type Description
input rml_op Operation that outputs tensor containing input data.
block_size uint32_t Size of data blocks for rearrangement.