公开(公告)号：US12124939B1

公开(公告)日：2024-10-22

申请号：US17199221

申请日：2021-03-11

申请人： Perceive Corporation

发明人： Justin Tantiongloc ,  Brian Thomas ,  Steven L. Teig

IPC分类号： G06N3/048 ,  G06F9/48 ,  G06F9/50 ,  G06F18/21 ,  G06N3/063

CPC分类号： G06N3/048 ,  G06F9/4881 ,  G06F9/5016 ,  G06F18/2163 ,  G06N3/063 ,  G06F2209/5017 ,  G06F2209/506

摘要： Some embodiments provide a method for generating neural network program instructions for a neural network inference circuit to execute a neural network. The neural network inference circuit includes a particular amount of available memory. The method receives a specification of the neural network including multiple layers. The method determines (i) a required amount of weight memory for the neural network and (ii) required amounts of activation memory for each of a set of layers of the neural network. When the required amount of weight memory and the required amount of activation memory for at least one layer is greater than the particular amount of available memory, the method generates the program instructions for the neural network inference circuit to execute a first set of the layers of the neural network multiple times for different blocks of input data and execute a second set of the layers in a single pass.