Develop evaluation methods for matching models

We will want to compare, select, and evaluate matching models. This requires generating and storing metrics (see https://github.com/dssg/pgdedupe/issues/20 for some possibilities) and, perhaps comparing Type I and Type II error rates on labeled pairs not used in the training data (see #20).

This will likely entail storing metrics in a metrics table and a notebook/methods/workflow for conducting comparisons and evaluations.

Start by reading issue #20 for the suggested metrics and labeled-pair evaluation possibilities. Then define the metrics table and the notebook or methods workflow needed to compare matching models. Done means metrics can be generated, stored, and used to compare models, including Type I and Type II error rates where applicable.

Written by the indexing model from the issue text.