delete old joblib notes

doc/impl-tips.rst

@@ -63,20 +63,3 @@ that use randomness at predict or recommendation time, not just training time, s
 value ``'user'`` for the ``rng`` parameter, and if it is passed, derive a new seed for each user
 using :py:func:`seedbank.derive_seed` to allow reproducibility in the face of parallelism for common
 experimental designs.  :py:func:`lenskit.util.derivable_rng` automates this logic.
-
-Memory Map Friendliness
------------------------
-
-LensKit uses :py:class:`joblib.Parallel` to parallelize internal operations (when it isn't using Numba).
-Joblib is pretty good about using shared memory to minimize memory overhead in parallel computations,
-and LensKit has some tricks to maximize this use. However, it does require a bit of attention in
-your algorithm implementation.
-
-The easiest way to make this fail is to use many small NumPy or Pandas data structures.  If you have
-a dictionary of :py:class:`np.ndarray` objects, for instance, it will cause a problem.  This is because
-each array will be memory-mapped, and each map will *reopen* the file.  Having too many active
-open files will cause your process to run out of file descriptors on many systems.  Keep your
-object count to a small, ideally fixed number; in :py:class:`lenskit.algorithms.basic.UnratedItemSelector`,
-we do this by storing user and item indexes along with a :py:class:`matrix.CSR` containing the items
-rated by each user.  The old implementation had a dictionary mapping user IDs to ``ndarray``s with
-each user's rated items.  This is a change from :math:`|U|+1` arrays to 5 arrays.

min-constraints.txt

@@ -4,5 +4,4 @@ scipy==1.2.1
 numba==0.51.0
 pyarrow==0.15.1
 cffi==1.12.2
-joblib==0.13.0
 binpickle==0.3.2