公开(公告)号：US20210124588A1

公开(公告)日：2021-04-29

申请号：US16838971

申请日：2020-04-02

Applicant: Samsung Electronics Co., Ltd.

Inventor： Titash Rakshit ,  Malik Aqeel Anwar ,  Ryan Hatcher

IPC: G06F9/38 ,  G06N3/02 ,  G06F17/16 ,  G06T1/20

Abstract: A method of pipelining inference of a neural network, which includes an i-th layer (i being an integer greater than zero), an (i+1)-th layer, and an (i+2)-th layer, includes processing a first set of i-th values of the i-th layer to generate (i+1)-th values for the (i+1)-th layer, determining a quantity of the (i+1)-th values as being sufficient for processing, and in response to the determining, processing the (i+1)-th values to generate an output value for the (i+2)-th layer while concurrently processing a second set of i-th values of the i-th layer.

公开(公告)号：US11769043B2

公开(公告)日：2023-09-26

申请号：US16839043

申请日：2020-04-02

Applicant: Samsung Electronics Co., Ltd.

Inventor： Titash Rakshit ,  Malik Aqeel Anwar ,  Ryan Hatcher

IPC: G06T1/20 ,  G06N3/063 ,  G06F17/16 ,  G06N3/04 ,  G06N3/08 ,  G06F18/21

CPC classification number: G06N3/063 ,  G06F17/16 ,  G06F18/21 ,  G06N3/04 ,  G06N3/08

Abstract: A method of pipelining inference of a neural network, which includes an i-th layer (i being an integer greater than zero) and an (i+1)-th layer, includes processing, for a first input image, first i-th values of the i-th layer to generate first (i+1)-th values for the (i+1)-th layer, processing, for the first input image, the first (i+1)-th values of the (i+1)-th layer to generate output values, and concurrently with processing, for the first image, the (i+1)-th values, processing, for a second input image, second i-th values of the i-th layer to generate second (i+1)-th values.

公开(公告)号：US20210124984A1

公开(公告)日：2021-04-29

申请号：US16839043

申请日：2020-04-02

Applicant: Samsung Electronics Co., Ltd.

Inventor： Titash Rakshit ,  Malik Aqeel Anwar ,  Ryan Hatcher

IPC: G06K9/62 ,  G06N3/04 ,  G06N3/08 ,  G06F17/16 ,  G06K9/46

Abstract: A method of pipelining inference of a neural network, which includes an i-th layer (i being an integer greater than zero) and an (i+1)-th layer, includes processing, for a first input image, first i-th values of the i-th layer to generate first (i+1)-th values for the (i+1)-th layer, processing, for the first input image, the first (i+1)-th values of the (i+1)-th layer to generate output values, and concurrently with processing, for the first image, the (i+1)-th values, processing, for a second input image, second i-th values of the i-th layer to generate second (i+1)-th values.