As noted in PR #33 the default partitioning results with very lopsided partitions in our standard tests, which in turn results in a disturbingly high error. At some point we should change the default partitioning strategy and rebaseline the tests accordingly.
As noted in PR #33 the default partitioning results with very lopsided partitions in our standard tests, which in turn results in a disturbingly high error. At some point we should change the default partitioning strategy and rebaseline the tests accordingly.