...
- Code Freeze: End of June
- Release published: July
Proposed content:
Project | Lead Contributor | Project Docs | Notes |
---|---|---|---|
Scala Inference API | Naveen Swamy | MXNet Scala Inference API | |
ONNX Export from MXNet | Roshani Nagmote | Proposal: ImportExport module | |
Gluon Vision Model Zoo | zhasheng@apache.org | ||
Fix Flaky Tests | zhasheng@apache.org | Github Flaky Tests | |
MKL-DNN | Patric Zhao, Da Zheng | MKL-DNN integration design | experimental in v1.2, upgrade to GA in v1.3 |
Proper support for Types | Sebastian (Wolfram) | 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 | |
Sparse Tensor support for Gluon | Haibin Lin | experimental feature | |
Android support | Anton | experimental feature or can it be GA? | |
Topology-aware AllReduce | Topology-aware AllReduce Proposal | experimental feature | |
control flow operators | Da Zheng, | Optimize dynamic neural networks | This is the first step towards optimizing dynamic neural networks by adding symbolic and imperative control flow operators (foreach, while_loop, maybe ifelse). This step extends Gluon to hybridize models with control flow. |
Clojure package | Carin Meier | MXNet Clojure | experimental clojure language binding |
Fused RNN Operators for CPU | Patric Zhao, Tao Lv | Fused CPU RNN Design | LSTM/GRU PR has been merged; vRNN will PR soon |
Completed in v1.2 | Lead Contributor | Project Docs | Notes |
---|---|---|---|
Distributed Training for FP16 | https://github.com/apache/incubator-mxnet/pull/10183 | v1.2 | |
Model Quantization - Experimental | Experimental quantization | v1.2 | |
Support for TensorBoard | Logging MXNet Data for Visualization in TensorBoard (a separate repo) | v1.2 | |
Friendly exception handling | Improved Exception Handling in MXNet | v1.2 |
...