Compressing device and method using parameters of quadtree method
AssigneeSamsung Electronics Co., Ltd.
InventorsSungho Jun, Junseok Park, Sangmin Suh, Heonsoo Lee, Hyukjae Jang, Kyungah Jeong
Life of this patent
Estimating the term…
Abstract
A device configured to compress a tensor including a plurality of cells includes: a quadtree generator configured to generate a quadtree searching for a non-zero cell included in the tensor and extract at least one parameter value from the quadtree; a mode selector configured to determine a compression mode based on the at least one parameter; and a bitstream generator configured to generate a bitstream by compressing the tensor based on the compression mode.
Claims20
A compression method of a neural network processor configured to perform computations on a feature map and a weight by using a neural network, the method comprising:
- receiving a tensor comprising a plurality of cells as a result of the computations;
- setting a quantization range of the tensor based on a largest value of a cell among the plurality of cells;
- selectively quantizing the tensor based on the quantization range;
- and
- compressing the quantized tensor.
A device configured to compress a tensor comprising a plurality of cells, the device comprising:
- a quantizer configured to quantize the tensor;
- a quadtree generator configured to generate a quadtree searching for a non-zero cell comprised in the quantized tensor and extract at least one parameter from the quadtree;
- a mode selector configured to determine a compression mode based on the at least one parameter;
- and
- a bitstream generator configured to generate a bitstream by compressing the quantized tensor based on the compression mode.
A neural network processor comprising:
- an arithmetic circuit configured to generate a tensor comprising a plurality of cells by performing a computation on input data by using a neural network;
- and
- a neural tensor compressor configured to output a bitstream by compressing the tensor,
- wherein the neural tensor compressor is further configured to quantize the tensor, to generate a quadtree corresponding to a repetitive spatial division method to search for a non-zero cell comprised in the quantized tensor, extract at least one parameter from the quadtree, and determine a compression mode of the bitstream based on the at least one parameter.
Dependencies are inferred from the claim text.