Dear TSC members,
Please find below the release plan for our upcoming release 0.2.0-alpha. Starting next
release we will present the release planning to TSC at the beginning of the month for
Please also note that this is not the final releasenote, which will contain additional
information on bugfixes and so forth.
Feel free to ask any questions or provide any feedback through the list !
Planned Features and Improvements For 0.2.0-alpha release
Training and Inference Framework
・ New models
* SSD: Single Shot MultiBox Detector.
* MobileNetV2: Inverted Residuals and Linear Bottlenecks.
* ResNet101: Deep Residual Learning for Image Recognition.
・ Frontend and User Interface
* Support for all python comparison operators.
* Support for math operators **,//,%. Support for other python operators like
and/or/not/is/is not/ in/ not in.
* Support for the gradients of function with variable arguments.
* Support for tensor indexing assignment for certain indexing type.
* Support for dynamic learning rate.
* User interfaces change log
* DepthwiseConv2dNative, DepthwiseConv2dNativeBackpropFilter,
・ Executor and Performance Optimization
* Support parallel execution of data prefetching and forward/backward computing.
* Support parallel execution of gradient aggregation and forward/backward computing
in distributed training scenarios.
* Support operator fusion optimization.
* Optimize compilation process and improve the performance.
・ Data processing, augmentation, and save format
* Support multi-process of GeneratorDataset/PyFunc for high performance
* Support variable batchsize
* Support new Dataset operators, such as filter,skip,take,TextLineDataset
Other Hardware Support
* GPU platform
* Support device memory swap in/out during training process.
* Quantization aware training (including training and inference).
* Add GPU kernels for Bert.
* CPU platform
* Support for windows 10 OS.
* Provides a common graph-level option, and multiple requirements can also share this
mechanism in the future.
* Improve graph compilation performance.
* Optimize memory allocation.
* Optimize serveral operators e.g., Slice, StridedSlice, ScatterMax etc.
* Add a white-box attack method:
* Add three neuron coverage metrics: KMNCov, NBCov,
* Add a coverage-guided fuzzing test framework for deep neural
* Update the MNIST Lenet5 example.
* Remove some duplicate code.
* Parameter distribution graph (Histogram). Now you can use
and MindInsight to record and visualize distribution info of tensors. See our
* Lineage support Custom information
* GPU support
* Model and dataset tracking linkage support
黄之鹏 Zhipeng (Howard) Huang
主任工程师 - 智能计算&IT开源生态部
Principle Engineer - Intelligent Computing & IT Open Source Ecosystem Department
Huawei Technologies Co., Ltd.
Tel : +86 755 28780808 / 18576658966
Email : huangzhipeng(a)huawei.com
This e-mail and its attachments contain confidential information from HUAWEI, which is
intended only for the person or entity whose address is listed above. Any use of the
information contained herein in any way (including, but not limited to, total or partial
disclosure, reproduction, or dissemination) by persons other than the intended
recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender
by phone or email immediately and delete it!