Evaluations

Model comparison across benchmark and custom dataset runs.