printing samplesets fails when one of the datafields is more than one dimension and has float biases

[arcondello]

        rec = np.recarray(5, dtype=[('sample', '<f8', 2),
                                    ('energy', '<f8'),
                                    ('num_occurrences', 'int8'),
                                    ('violations', '<i1', 10),
                                    ])

        rec[:] = 1

        ss = dimod.SampleSet(rec, range(2), {}, 'SPIN')

        print(ss)

works fine, but

        rec = np.recarray(5, dtype=[('sample', '<f8', 2),
                                    ('energy', '<f8'),
                                    ('num_occurrences', 'int8'),
                                    ('violations', '<f8', 10),
                                    ])

        rec[:] = 1

        ss = dimod.SampleSet(rec, range(2), {}, 'SPIN')

        print(ss)

raises

TypeError: only size-1 arrays can be converted to Python scalars