| CPC G06F 9/3001 (2013.01) [G06F 9/345 (2013.01); G06N 3/02 (2013.01); G06N 3/063 (2013.01); G06N 3/048 (2023.01)] | 20 Claims |
|
1. A method for generating output feature map data during operation of neural network processing by a processing unit, the processing unit comprising a plurality of computation resources, the method comprising:
obtaining first, real, data to be processed;
determining, based on a number of computation resources of a set of the plurality of computation resources available for use during a processing cycle of the processing unit, to process at least a portion of the first data using a first subset of the set and to load second, artificial, data into a second subset of the set, disjoint from the first subset of the set, the second data comprising at least one artificial data element;
obtaining the second data;
in the processing cycle:
loading at least the portion of the first data into the first subset of the set, causing the first subset of the set to generate a computational result; and
loading the second data into the second subset of the set, the second subset of the set being an artificially activated subset of the set;
inhibiting the second data from affecting the output feature map data; and
generating the output feature map data based at least in part on the computational result.
|