model_builders

A model builder is a method use to build a (typically data-driven) model in a workflow like greedy.

As an example, we can consider a simple training algorithm where we iteratively run FOM samples, use them to construct a ROM, evaulate the ROM, and then continue to add samples until the ROM is of a desired accuracy. In this context, the model_builder interface will provide context on how to construct the ROM given the FOM samples.

Contents:

• Basic model builder tutorial