Automated processing of multiple prediction generation including model tuning

A non-transitory computer-readable storage medium comprising stored instructions executable by a processor system, the instructions when executed causing the processor system to:

• access a set of machine-learning models trained using a dataset, wherein the dataset comprises at least a set of keys and key-values for the set of keys, and wherein each machine-learning model is trained using a respective subset of the dataset corresponding to a respective set of key-values for the set of keys;
• receive, via an interface, a current dataset that is an updated dataset;
• determine whether drift has occurred with respect to the dataset used to train the set of machine-learning models, comprising determining whether a difference between the current dataset and the dataset exceeds a drift threshold;
• responsive to a determination the drift has occurred and the difference exceeds the drift threshold, identify one or more machine-learning models in the set of machine-learning models that were trained using a drifted portion of the dataset;
• tune parameters of the one or more machine-learning models in the set of machine-learning models using the current dataset to generate another set of machine-learning models;
• for a selected machine-learning model that is not determined to have drift, obtain an accuracy of the machine-learning model;
• responsive to a determination that the accuracy is below an accuracy threshold, iteratively update parameters of the selected machine-learning model by using a set of optimizers that are threads, compute nodes, or processor cores until the accuracy of the machine-learning model exceeds the accuracy threshold,
• wherein the set of optimizers retain a history of the iterative updates and stops when a maximum iterations are reached, stops when there is no improvement in the accuracy, or the accuracy decreases;
• and
• expose the another set of machine-learning models via an interface of a computing service.

A computer system, comprising:

• a processor system;
• and
• a non-transitory computer-readable storage medium comprising stored instructions executable by a processor system, the instructions when executed causing the processor system to:
• access a set of machine-learning models trained using a dataset, wherein the dataset comprises at least a set of keys and key-values for the set of keys, and wherein each machine-learning model is trained using a respective subset of the dataset corresponding to a respective set of key-values for the set of keys;
• receive, via an interface, a current dataset that is an updated dataset;
• determine whether drift has occurred with respect to the dataset used to train the set of machine-learning models, comprising determining whether a difference between the current dataset and the dataset exceeds a drift threshold;
• responsive to a determination the drift has occurred and the difference exceeds the drift threshold, identify one or more machine-learning models in the set of machined-learning models that were trained using a drifted portion of the dataset;
• tune parameters of the one or more machine-learning models in the set of machine-learning models using the current dataset to generate another set of machine-learning models;
• for a selected machine-learning model that is not determined to have drift, obtain an accuracy of the machine-learning model;
• responsive to a determination the accuracy is below an accuracy threshold, iteratively update parameters of the selected machine-learning model by using a set of optimizers that are threads, compute nodes, or processor cores until the accuracy of the machine-learning model exceeds the accuracy threshold,
• wherein the set of optimizers retain a history of the iterative updates and stops when a maximum iterations are reached, stops when there is no improvement in the accuracy, or the accuracy decreases;
• and
• expose the another set of machine-learning models via an interface of a computing service.