From 08b53632a288cf48a8d8553fc10570a0fbd524b0 Mon Sep 17 00:00:00 2001 From: Paul Andrey <paul.andrey@inria.fr> Date: Fri, 1 Mar 2024 15:19:15 +0100 Subject: [PATCH] Update release notes for '2.4.0'. --- docs/release-notes/v2.4.0.md | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/docs/release-notes/v2.4.0.md b/docs/release-notes/v2.4.0.md index af8b7633..43ce1197 100644 --- a/docs/release-notes/v2.4.0.md +++ b/docs/release-notes/v2.4.0.md @@ -249,3 +249,23 @@ can be specified in various ways. Previously, they could only be passed as `(func, args)` tuples. Now, they can either be passed as `(func, args)`, `(func, kwargs)` or `(func, args, kwargs)`, where `args` is still a tuple of positional arguments, and `kwargs` is a dict of keyword ones. + +## Other changes + +### Fix redundant sharing of model weights with clients + +`FederatedServer` now keeps track of clients having received the latest global +model weights, and avoids sending them redundantly with future training (or +evaluation) requests. To achieve this, `TrainRequest` and `EvaluationRequest` +now support setting their `weights` field to `None`. + +### Update TensorFlow supported versions + +Older TensorFlow versions (v2.5 to 2.10 included) were improperly marked as +supported in spite of `TensorflowOptiModule` requiring at least version 2.11 +to work (due to changes of the Keras Optimizer API). This has been corrected. + +Upcoming TensorFlow versions (v2.16 and onwards) introduce backward-breaking +changes, which are probably due to the backend swap from Keras 2 to Keras 3. +To keep safe, these versions are currently marked as unsupported, awaiting +further investigation once version 2.16 is finalized. -- GitLab