US 11,681,511 B2
Systems and methods for building and deploying machine learning applications
Jitender Arora, Westford, MA (US); Amir H. Atai, New Boston, NH (US); Michael Piecuch, Cary, NC (US); and Hassan Ahmed, Concord, MA (US)
Assigned to Sway AI Inc., Chelmsford, MA (US)
Filed by Sway AI Inc., Chelmsford, MA (US)
Filed on Dec. 23, 2021, as Appl. No. 17/560,576.
Claims priority of provisional application 63/130,472, filed on Dec. 24, 2020.
Prior Publication US 2022/0206774 A1, Jun. 30, 2022
Int. Cl. G06K 9/62 (2022.01); G06F 9/445 (2018.01); G06F 16/22 (2019.01); G06F 9/455 (2018.01); G06F 8/60 (2018.01); G06F 9/48 (2006.01); G06N 20/00 (2019.01)
CPC G06F 8/60 (2013.01) [G06F 9/4881 (2013.01); G06N 20/00 (2019.01)] 22 Claims
OG exemplary drawing
 
1. A method of operating a data processing system configured to deploy one or more machine learning applications, each machine learning application configured to process input data using a respective machine learning model to generate output data, the method comprising, by one or more processors:
receiving user input defining at least a portion of a first machine learning application, the portion of the first machine learning application comprising a first component, the first component defining a first data type as input to the first component and identifying input data having the first data type;
automatically selecting a machine learning model for the first machine learning application by:
identifying one or more machine learning models each associated with the first data type;
for each of the one or more machine learning models associated with the first data type, generating a respective metric value indicating a performance of the machine learning model over at least a first portion of the input data having the first data type; and
selecting a machine learning model from among the one or more machine learning models by comparing metric values generated for each of the one or more machine learning models; and
configuring the first component of the first machine learning application so that the first component is configured to utilize the selected machine learning model.