US 12,248,867 B2
Data processing device for neural network, method of operating the same, and program
Shunsuke Okumura, Tokyo (JP); and Koichi Nose, Tokyo (JP)
Assigned to RENESAS ELECTRONICS CORPORATION, Tokyo (JP)
Filed by RENESAS ELECTRONICS CORPORATION, Tokyo (JP)
Filed on Nov. 5, 2020, as Appl. No. 17/090,534.
Claims priority of application No. 2019-200609 (JP), filed on Nov. 5, 2019.
Prior Publication US 2021/0132866 A1, May 6, 2021
Int. Cl. G06F 7/50 (2006.01); G06F 3/06 (2006.01); G06F 18/21 (2023.01); G06N 3/04 (2023.01); G06N 3/063 (2023.01); G06V 10/94 (2022.01)
CPC G06N 3/063 (2013.01) [G06F 3/0604 (2013.01); G06F 3/0659 (2013.01); G06F 3/0673 (2013.01); G06F 7/50 (2013.01); G06F 18/2193 (2023.01); G06N 3/04 (2013.01); G06V 10/95 (2022.01); G06V 10/955 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A data processing device comprising:
a processor; and
a first memory storing instructions that, when executed by the processor, cause the processor to:
determine whether or not each of binarized input data is a predetermined value;
store, in a second memory, a plurality of coefficients and coefficient address information including information related to coefficient addresses where the plurality of coefficients are stored;
read the coefficient address from the second memory based on a determination result of the binarized input data being the predetermined value or not and read the coefficient from the second memory based on the coefficient address; and
execute an arithmetic operation related to the acquired coefficient.