US 12,254,401 B2
Artificial neural network model selection
Poorna Kale, Folsom, CA (US)
Assigned to Micron Technology, Inc., Boise, ID (US)
Filed by Micron Technology, Inc., Boise, ID (US)
Filed on Dec. 9, 2020, as Appl. No. 17/116,847.
Prior Publication US 2022/0180185 A1, Jun. 9, 2022
Int. Cl. G06N 3/08 (2023.01); G06F 18/21 (2023.01); G06N 3/045 (2023.01); G06T 7/00 (2017.01)
CPC G06N 3/08 (2013.01) [G06F 18/217 (2023.01); G06N 3/045 (2023.01); G06T 7/0002 (2013.01); G06T 2207/20084 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving, at a memory device on which a plurality of artificial neural network (ANN) models are implemented, first data, second data, and third data from an imaging device,
wherein the first data corresponds to a first image captured by the imaging device, the second data corresponds to a second image captured by the imaging device subsequent to the first image, and the third data corresponds to a third image captured by the imaging device subsequent to the second image;
executing only a first ANN model on the first data;
executing only a second ANN model on the second data;
determining, by the memory device, whether an accuracy value of results yielded from the execution of the second ANN model on the second data is less than a threshold accuracy value; and
responsive to determining that the accuracy value is less than the threshold accuracy value, executing only the first ANN model on the third data.