Evaluation Library for the Kepler Model Server

What would you like to be added?

Incorporate a standardized way of evaluating specifically Regression Sklearn (if not other models) Models and calculating essential error metrics like Mean Absolute Percentage Error and R2.

Why is this needed?

We need a better way of discussing how good our models are during tech conferences. This library will make it so new models implemented do not need to worry about including Evaluation metrics.

WIP: https://github.com/sthaha/kepler-model-validation/tree/main