Release timelines:

Following timeline is based on everything goes well.(Added some buffer time)

StepTasksCompleted byComments
Code Freeze and release startTrack ongoing PRs08/02 08/07 08/10Delayed due to unmerged PRs(#11482, #11325)

Cut the release branchCheck license headers08/17Depending on stability

Make code changes with necessary version updates

Cut the release branch

Test the release and tag the releaseNightly test, Jenkins CI08/22

RAT check

Tag RC0

Package artifacts and validateCreate release artifacts08/28

Validate release package

Test release package

Scala release process

Begin apache votingStart vote on dev@08/28-09/05

Start vote on general@

Finalizing and posting the releaseCreate the final release tag on github09/12

Rename, resign and upload the src tar to final dir

Update the website using tag

Release the official pip package

Release the official docker images

After 24 hrs, validate the packages are uploaded

Draft the offical announce email and review

Send out the email on announce@

Update the apache blog

update the aws blog

send internal announcement

Update the version on master

New Features and Improvements:


Lead Contributor

Project Docs

Link to design discussion on dev@


Scala API ImprovementsQing LanNaveen Swamy 
MXNet Scala Usability ImprovementsDiscussion #10660, #10787, #10991 Improvements to MXNet Scala API usability: type safety, API documentation.
Scala Example CIQing LanNaveen SwamyAndrew Ayres

#11753, #11621 Refurnished Scala Examples with improved API, documentation and CI test coverage
ONNX Export from MXNetRoshani NagmoteProposal: ImportExport moduleDiscussion #11213 MXNet to ONNX exporter APIs to export MXNet trained models to ONNX protobuf so that those models can be imported in other frameworks for inference.
Topology-aware AllReduce

Carl Yang

Topology-aware AllReduce ProposalDiscussion#11591 experimental feature
MKL-DNNPatric ZhaoDa ZhengMKL-DNN integration designDiscussion#experimental in v1.2, upgrade to GA in v1.3
Clojure packageCarin MeierMXNet ClojureDiscussion

#11205 experimental clojure language binding

TensorRT Runtime Integration

Marek Kolodziej,

Clement Fuji Tsang,

Kellen Sunderland

Runtime Integration with TensorRT

Discussion#11325 This feature introduces runtime integration of TensorRT into MxNet, in order to accelerate inference.
Sparse Tensor support for GluonHaibin LinGluon Sparse SupportDiscussion#11429, #11231, #11197, #11001, #10924 experimental feature
Fused RNN Operators for CPUPatric ZhaoTao LvFused CPU RNN DesignDiscussion#10104 #11399 #10311 vRNN/LSTM/GRU PR has been merged
Fix Flaky Testszhasheng@apache.orgGithub Flaky TestsDiscussion
Control flow operatorsDa Zheng,Optimize dynamic neural networksDiscussionThis is the first step towards optimizing dynamic neural networks by adding symbolic and imperative control flow operators (foreach, while_loop, maybe ifelse). foreach and while_loop have been merged. ifelse has been submitted for review, but is unlikely merged in this release.

Open PRs to track:






TensorRT runtime integration


make gluon rnn layers hybrid blocks

Sheng ZhaOpen

RNN Example for Scala

Qing LanNot going in

NeuralStyle Example for Scala

Qing LanMerged

Add download util

Qing Lan


  • No labels