SYSTEMS AND METHODS FOR DISTRIBUTED DATA ANALYTICS
The invention provides systems and method for generating device-specific artificial neural network (ANN) models for distribution across user devices. Sample datasets are collected from devices in a particular environment or use case and include predictions by device-specific ANN models executing the user devices. The received datasets are used with existing datasets and stored ANN models to generate updated device-specific ANN models from each of the stored instances of the device ANN models based on the training data.
1 .- 28 . (canceled)
29 . A method for optimizing the execution of device-specific trained artificial neural network (ANN) models on devices, the method comprising:
receiving, by a processor, a first trained ANN model and a second ANN model, wherein the first ANN model and the second ANN model each execute different inferences on input data;
merging the first ANN model, the second ANN model, and control flow execution instructions into a combined software package; and
deploying the combined software package to an edge device for execution thereon according to the control flow instructions.
30 . The method of claim 29 , wherein output of the first ANN model serves as input to the second ANN model.
31 . The method of claim 29 , wherein the first trained ANN model and second trained ANN model each comprise respective analytics criteria and use case data.
32 . The method of claim 29 , wherein the processor selects the first and second ANN models based, at least in part, on the analytics criteria therein.
33 . The method of claim 29 , further comprising generating a parent ANN as a meta-architecture based on the first ANN model architecture and the second ANN model architecture, and the meta-architecture is delivered to the edge device such that it executes as a single ANN model.
34 . The method of claim 29 , wherein the edge device comprises a camera.
35 . The method of claim 34 , wherein execution of the first ANN model and second ANN model on the camera identifies an object of interest in an image file captured on the camera.
36 . A system for optimizing the execution of device-specific trained artificial neural network (ANN) models on edge devices, the system comprising:
one or more processors; and
a memory coupled with the one or more processors wherein the one or more processors executes computer-executable instructions stored in the memory, that
when executed:
identify a first trained ANN model and a second ANN model, wherein the first ANN model and the second ANN model each execute different inferences on input data;
merge the first ANN model, the second ANN model, and control flow execution instructions into a combined software package; and
deploy, by a distribution module, the combined software package to an edge device for execution thereon according to the control flow instructions.
37 . The system of claim 36 , wherein the output of the first ANN model serves as input to the second ANN model.
38 . The system of claim 36 , wherein the first trained ANN model and second trained ANN model each comprise respective analytics criteria and use case data.
39 . The system of claim 36 , wherein the processor selects the first and second ANN models based, at least in part, on the analytics criteria therein.
40 . The system of claim 36 , wherein execution of the instructions further generates a parent ANN as a meta-architecture based on the first ANN model architecture and the second ANN model architecture, and the meta-architecture is delivered to the edge device such that it executes as a single ANN model.
41 . The system of claim 36 , wherein the edge device comprises a camera.
42 . The system of claim 41 , wherein execution of the first ANN model and second ANN model on the camera identifies an object of interest in an image file captured on the camera.
43 . A non-transitory computer-readable storage medium storing computer-readable instructions that, when executed by a computer, cause the computer to perform a method, the method comprising:
receiving, by a processor, a first trained ANN model and a second ANN model, wherein the first ANN model and the second ANN model each execute different inferences on input data;
merging the first ANN model, the second ANN model, and control flow execution instructions into a combined software package; and
deploying the combined software package to an edge device for execution thereon according to the control flow instructions.