Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Release timelines:

StepDateComments
Code Freeze08/02Track ongoing PRs
RC008/10Depending on stability


Proposed content:

Project

Lead Contributor

Project Docs

Link to design discussion on dev@

Notes

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

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 ProposalDiscussionexperimental feature
Clojure packageCarin MeierMXNet ClojureDiscussionexperimental clojure language binding
TensorRT Runtime Integration

Marek Kolodziej,

Clement Fuji Tsang,

Kellen Sunderland

Runtime Integration with TensorRT

DiscussionThis feature introduces runtime integration of TensorRT into MxNet, in order to accelerate inference.
Proper support for Types

Sebastian (Wolfram)

(sebastianbod@gmail.com) or taliesinb@gmail.com



Backend for general purpose numerical packages (support for multipe types, proper testing for all supported types). Derive test priorities from Mathematica internal suite or integrate Mathematica into CI?

Feedback: Users are observing instability, incomplete behavior vs. quality delivered by numpy / Pytorch.

Haibin: many tests are not well written, most test only cover fp32, don't have good guidelines for community, operator tutorial lacks guidance on test requirements, suggest to develop guidelines (results are deterministic, cover all supported types), need to set quality standards as community grows

Sebastian: presentation how it is done in Mathematica based on experience and learnings

Fused RNN Operators for CPUPatric ZhaoTao LvFused CPU RNN DesignDiscussionvRNN/LSTM/GRU PR has been merged


Open PRs to track:


PR

Title

Contributor

Status

#11325

TensorRT runtime integration

MarekOpen
#11482

make gluon rnn layers hybrid blocks

Sheng ZhaOpen
#11753

RNN Example for Scala

Qing LanOpen
#11621

NeuralStyle Example for Scala

Qing LanOpen
#11866

Add download util

Qing Lan

Merged


...