US 11,775,806 B2
Method of compressing neural network model and electronic apparatus for performing the same
Yoo Chan Kim, Seoul (KR); Jong Won Baek, Uiwang-si (KR); and Geun Jae Lee, Seoul (KR)
Assigned to NOTA, INC., Daejeon (KR)
Filed by NOTA, INC., Daejeon (KR)
Filed on Feb. 2, 2023, as Appl. No. 18/163,527.
Claims priority of application No. 10-2022-0017230 (KR), filed on Feb. 10, 2022; application No. 10-2022-0017231 (KR), filed on Feb. 10, 2022; application No. 10-2022-0023385 (KR), filed on Feb. 23, 2022; application No. 10-2022-0048201 (KR), filed on Apr. 19, 2022; application No. 10-2022-0057599 (KR), filed on May 11, 2022; and application No. 10-2022-0104355 (KR), filed on Aug. 19, 2022.
Prior Publication US 2023/0252293 A1, Aug. 10, 2023
Int. Cl. G06N 3/08 (2023.01); G06F 3/04847 (2022.01)
CPC G06N 3/08 (2013.01) [G06F 3/04847 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of compressing a neural network model that is performed by a computing device, comprising:
receiving, at a processor of the computing device, a trained model and compression method instructions for compressing the trained model;
identifying, via the processor, a compressible block and a non-compressible block among a plurality of blocks included in the trained model based on the compression method instructions;
transmitting, via a computer network, a command to a user device that causes the user device to:
display a structure of the trained model representing a connection relationship between the plurality of blocks on a first screen such that the compressible block and the non-compressible block are visually distinguished, and
display, on a second screen, an input field operable to receive a parameter value entered by a user for compression of the compressible block; and
compressing the trained model based on the parameter value entered by the user in the input field.