Changelog update (#891)

* [skip ci] adding comments to changelog

* adding date to changelog

* [skip ci] minor edit

CHANGELOG.md

@@ -4,12 +4,22 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [0.4.4] - TBD
+## [0.4.5] - TBD
+
+### Added
+
+### Changed
+
+## [0.4.4] - 2021-12-22
 
 ### Fixed
-- inf/nan check on all gradient tensors in ShardedGradScaler [#890]
+- Inf/nan check on all gradient tensors in ShardedGradScaler [#890]
+- Allow user to supress warning on trainable_parameter in sharded_ddp.py [#886]
+- KeyError in certain FSDP wrapping scenarios [#881]
 
 ### Added
+- Support eval mode in MEVO [#884]
+- A new benchmark for the MOE model [#866]
 
 ### Changed
 - Fixed a corner case of FSDP init order and losing one of the flags [#880]

README.md

@@ -22,6 +22,7 @@ FairScale was designed with the following values in mind:
 
 ## What's New:
 
+* December 2021 [fairscale 0.4.4 was released](https://github.com/facebookresearch/fairscale/releases/tag/v0.4.4).
 * FairScale is tested with the following PyTorch versions (with CUDA 11.2): 1.8.1, 1.10.0 and 1.11.0.dev20211101+cu111.
 * November 2021 [fairscale 0.4.3 was released](https://github.com/facebookresearch/fairscale/releases/tag/v0.4.3).
 * We have experimental support for offloading params to disk when using the FSDP API for evaluation workloads.