From 2e4743ca548a55f1688c8e548dc95c85d4c201b1 Mon Sep 17 00:00:00 2001 From: David Chudzicki Date: Wed, 10 May 2023 21:11:45 -0400 Subject: [PATCH] Fix docstring for batch_size in client.map --- distributed/client.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/distributed/client.py b/distributed/client.py index a0fb7b7885..6219a55f93 100644 --- a/distributed/client.py +++ b/distributed/client.py @@ -2016,12 +2016,12 @@ def map( Whether or not the function is pure. Set ``pure=False`` for impure functions like ``np.random.random``. See :ref:`pure functions` for more details. - batch_size : int, optional + batch_size : int, optional (default: just one batch whose size is the entire iterable) Submit tasks to the scheduler in batches of (at most) ``batch_size``. - Larger batch sizes can be useful for very large ``iterables``, - as the cluster can start processing tasks while later ones are - submitted asynchronously. + The tradeoff in batch size is that large batches avoid more per-batch overhead, + but batches that are too big can take a long time to submit and unreasonably delay + the cluster from starting its processing. **kwargs : dict Extra keyword arguments to send to the function. Large values will be included explicitly in the task graph.