System, method, and computer program for a machine learning model-agnostic cloud-agnostic no-code onboarding platform

1. A non-transitory computer readable medium storing computer code executable by a processor to perform a method comprising:

providing a universal unified interface to a Model-Agnostic Onboarding Workflow (MAOW) platform, wherein the MAOW platform is configured as a no-code, model-agnostic, cloud-agnostic platform, and wherein the universal unified interface includes a dictionary of machine learning model input structured query language (SQL) statements and a dictionary of machine learning model output destinations;

onboarding a chain of machine learning models using the MAOW platform, based on the universal unified interface, wherein the MAOW platform:

creates an onboarding configuration file that indicates the dictionary of machine learning model input SQL statements for use in providing the chain of machine learning models with input datasets and that indicates the dictionary of machine learning model output destinations for use in storing output datasets of the chain from machine learning models,

creates a machine learning model pipeline for deployment of the chain of machine learning models as multiple machine learning models running one after another,

uses the dictionary of machine learning model input SQL statements to provide the chain of machine learning models with the input datasets,

caches input and output data of each machine learning model in the chain of machine learning models such that output data of one machine learning model in the chain of machine learning models is provided from the cache as input to a next machine learning model in the chain of machine learning models in accordance with the machine learning model pipeline, and

uses the dictionary of machine learning model output destinations to store the output datasets from the chain of machine learning models; and

deploying the chain of machine learning models from the MAOW platform to at least one target environment.

2. The non-transitory computer readable medium of claim 1 , wherein MAOW platform interfaces a development environment in which the chain of machine learning models is created.

3. The non-transitory computer readable medium of claim 1 , wherein the universal unified interface is a Unified Universal Interface Contract (UUIC) that serves as an interface between the chain of machine learning models and the MAOW platform.

4. The non-transitory computer readable medium of claim 3 , wherein the UUIC serves both input and output between main methods of the chain of machine learning models and the MAOW platform.

5. The non-transitory computer readable medium of claim 4 , wherein the UUIC is defined as a dictionary map of dictionary maps.

6. The non-transitory computer readable medium of claim 1 , wherein the MAOW platform enables automatic no-code configuration and customization of the chain of machine learning models.

7. The non-transitory computer readable medium of claim 6 , wherein the no-code configuration defines machine learning model-agnostic pre-processing and post-processing.

8. The non-transitory computer readable medium of claim 7 , wherein the pre-processing includes reading input for the chain of machine learning models from one or more defined input sources.

9. The non-transitory computer readable medium of claim 7 , wherein the post-processing includes writing output of the chain of machine learning models to one or more defined output destinations.

10. The non-transitory computer readable medium of claim 6 , wherein the no-code configuration further defines data pre- and post-transformations.

11. The non-transitory computer readable medium of claim 1 , wherein the chain of machine learning models is deployed and run as the machine learning pipeline on the target environment.

12. The non-transitory computer readable medium of claim 1 , wherein the target environment is one of a plurality of supported environments, the plurality of supported environments including a cloud environment, a hybrid cloud environment, an on-premises environment, a data center environment, and a local computer of a development and test environment in which the at least one machine learning model is created and tested.

13. The non-transitory computer readable medium of claim 1 , further comprising:

running the onboarded chain of machine learning models locally as a prototype to validate correct onboarding of components and their configuration and interaction, wherein a main method of the onboarded chain of machine learning models is replaced with an empty body leaving only an input and output parameters signature.

14. The non-transitory computer readable medium of claim 1 , further comprising:

using a MAOW Model Wrapper application programming interface (APIA to produce a run-time configuration MAOW Model Wrapper API main onboarding program for a particular input/output for the chain of machine learning models and testing the onboarded chain of machine learning models locally using a preliminary defined generic test case.

15. The non-transitory computer readable medium of claim 1 , wherein the MAOW platform includes:

a workflow that utilizes a directed acyclic graph (DAG) for running multiple models as a set of tasks with no directed cycles.

16. The non-transitory computer readable medium of claim 1 , wherein the MAOW platform provides output database tables metadata allowing machine learning model output dataset validation to accelerate a machine learning model productizing phase.

17. A method, comprising:

providing a universal unified interface to a Model-Agnostic Onboarding Workflow (MAOW) platform, wherein the MAOW platform is configured as a no-code, model-agnostic, cloud-agnostic platform, and wherein the universal unified interface includes a dictionary of machine learning model input structured query language (SQL) statements and a dictionary of machine learning model output destinations;

onboarding a chain of machine learning models using the MAOW platform, based on the universal unified interface, wherein the MAOW platform:

creates an onboarding configuration file that indicates the dictionary of machine learning model input SQL statements for use in providing the chain of machine learning models with input datasets and that indicates the dictionary of machine learning model output destinations for use in storing output datasets of the chain from machine learning models,

creates a machine learning model pipeline for deployment of the chain of machine learning models as multiple machine learning models running one after another,

uses the dictionary of machine learning model input SQL statements to provide the chain of machine learning models with the input datasets,

caches input and output data of each machine learning model in the chain of machine learning models such that output data of one machine learning model in the chain of machine learning models is provided from the cache as input to a next machine learning model in the chain of machine learning models in accordance with the machine learning model pipeline, and

uses the dictionary of machine learning model output destinations to store the output datasets from the chain of machine learning models; and

deploying the chain of machine learning models from the MAOW platform to at least one target environment.

18. A system, comprising:

a non-transitory memory storing instructions; and

one or more processors in communication with the non-transitory memory that execute the instructions to perform a method comprising:

providing a universal unified interface to a Model-Agnostic Onboarding Workflow (MAOW) platform, wherein the MAOW platform is configured as a no-code, model-agnostic, cloud-agnostic platform, and wherein the universal unified interface includes a dictionary of machine learning model input structured query language (SQL) statements and a dictionary of machine learning model output destinations;

onboarding a chain of machine learning models using the MAOW platform, based on the universal unified interface, wherein the MAOW platform:

creates an onboarding configuration file that indicates the dictionary of machine learning model input SQL statements for use in providing the chain of machine learning models with input datasets and that indicates the dictionary of machine learning model output destinations for use in storing output datasets of the chain from machine learning models,

creates a machine learning model pipeline for deployment of the chain of machine learning models as multiple machine learning models running one after another,

uses the dictionary of machine learning model input SQL statements to provide the chain of machine learning models with the input datasets,

caches input and output data of each machine learning model in the chain of machine learning models such that output data of one machine learning model in the chain of machine learning models is provided from the cache as input to a next machine learning model in the chain of machine learning models in accordance with the machine learning model pipeline, and

uses the dictionary of machine learning model output destinations to store the output datasets from the chain of machine learning models; and

deploying the chain of machine learning models from the MAOW platform to at least one target environment.