`average` track will actually run two other tracks...

# datascience

[orangy]

average track will actually run two other tracks sum and count (those can be folded with other tracks, and also run in parallel/cluster), get their results and combine them into average