Need somebody with expertise in interpreting model results

[palatiner]

Depending on the model output we need a way to assess accuracy, usefulness, etc.

[GuidosLabs]

With data split into train and test, we should be able to train each model and easily evaluate it on the test data. That's the standard way. There are several standard scoring methods to evaluate how well the model does on the test data. We can discuss to select the best.