Industrial Internet Application Development

Continuous training of a model

In the IIoT, to manage a fleet of equipment, it is often required to control the processes using machine learning techniques. The machine learning models help to predict and control the outcome. In such scenarios, model prediction performance often needs to be kept high. In a number of special scenarios, due to bad datasets or some unexpected events or changes in control settings, the performance of the model may drift. In such a setup, it is always advantageous to have the capability of continuous training of machine learning models and automation of deployment with little or no human intervention. Continuous training of a model provides a way of listening to the latest data and adjusting the model coefficients ...