Samplers / pipelines for imbalanced datasets

Imbalanced datasets, where the classes have very different occurrence rates, can show up in large data sets.

There are many strategies for dealing with imbalanced data. http://contrib.scikit-learn.org/imbalanced-learn/stable/api.html implements a set, some of which could be scaled to large datasets with dask.

Start by reviewing the imbalanced-learn API linked in the issue and the existing Dask ML project structure. The issue does not name files, tests, a specific sampler or pipeline, or a concrete completion criterion, so the scope and definition of done would need to be established first.

Written by the indexing model from the issue text.