*   Add `potential_scale_reduction` to
        `tf.contrib.bayesflow.mcmc_diagnostics`.
    *   Add `BatchNormalization`, `Kumaraswamy` bijectors.
    *   Deprecate `tf.contrib.learn`. Please check contrib/learn/README.md for
        instructions on how to convert existing code.
    *   `tf.contrib.data`
    *   Remove deprecated `tf.contrib.data.Dataset`, `tf.contrib.data.Iterator`,

causes, been parted from one another, or mixed together; or otherwise to
find out what sorts of Rays are in the Light by which that Colour is
made, and in what Proportion; and then by the last Problem to learn the
Colour which ought to arise by mixing those Rays (or their Colours) in
that proportion. I speak here of Colours so far as they arise from
Light. For they appear sometimes by other Causes, as when by the power