CPC G01R 31/31704 (2013.01) [G01R 31/318357 (2013.01); G06N 20/00 (2019.01)] | 29 Claims |
18. A method to support device under test (DUT) validation reuse across a plurality of platforms, comprising:
generating an inference profile used for an inference operation of a machine learning (ML) application based on one or more of a set of profile configurations, a set of test parameters, and a set of randomized constraints for the ML application;
accepting and verifying a plurality of math functions statically and/or dynamically to identify one or more DUT issues, wherein the plurality of math functions are specified for implementing the ML application;
building an inference model based on the inference profile and the plurality of verified math functions, wherein the inference model is reused for DUT validation across the plurality of platforms.
|