Calculate the coefficient of variation without calculating the mean twice?

[Boscop]

With this crate, is it possible to calculate the coefficient of variation / relative std dev (data.std_dev()/data.mean()) without calculating the mean twice?

[YeungOnion]

If the distribution is specified by a family of distributions, e.g. Normal, Exponential, then it certainly won't need to be computed twice.

If however, you need to determine this from data, perhaps from an iterable, then the Empirical distribution's implementation add to add a datum will stream the variance and mean.

average may also be useful for you.