diff --git a/README.md b/README.md index 37f5422dd9..0195c7058c 100644 --- a/README.md +++ b/README.md @@ -25,7 +25,7 @@ The tool manages automated machine learning (AutoML) experiments, **dispatches a ## **What's NEW!**   -* **New release**: [v2.5 is available](https://github.com/microsoft/nni/releases/tag/v2.5) - _released on Nov-04-2021_ +* **New release**: [v2.6 is available](https://github.com/microsoft/nni/releases/tag/v2.6) - _released on Jan-19-2022_ * **New demo available**: [Youtube entry](https://www.youtube.com/channel/UCKcafm6861B2mnYhPbZHavw) | [Bilibili 入口](https://space.bilibili.com/1649051673) - _last updated on May-26-2021_ * **New webinar**: [Introducing Retiarii: A deep learning exploratory-training framework on NNI](https://note.microsoft.com/MSR-Webinar-Retiarii-Registration-Live.html) - _scheduled on June-24-2021_ * **New community channel**: [Discussions](https://github.com/microsoft/nni/discussions) @@ -253,7 +253,7 @@ Note: * Download the examples via clone the source code. ```bash - git clone -b v2.5 https://github.com/Microsoft/nni.git + git clone -b v2.6 https://github.com/Microsoft/nni.git ``` * Run the MNIST example. diff --git a/README_zh_CN.md b/README_zh_CN.md index a6fbf441b0..9e551c8ef5 100644 --- a/README_zh_CN.md +++ b/README_zh_CN.md @@ -21,7 +21,7 @@ NNI 管理自动机器学习 (AutoML) 的 Experiment，**调度运行**由调优 ## **最新消息！**  [](#nni-released-reminder) -* **最新版本**：[v2.5 已发布](https://github.com/microsoft/nni/releases/tag/v2.5) - *2021年11月04日* +* **最新版本**：[v2.6 已发布](https://github.com/microsoft/nni/releases/tag/v2.6) - *2022年1月19日* * **最新视频 demo**：[Youtube 入口](https://www.youtube.com/channel/UCKcafm6861B2mnYhPbZHavw) | [Bilibili 入口](https://space.bilibili.com/1649051673) - *上次更新：2021年5月26日* * **最新网络研讨会**: [介绍Retiarii：NNI 上的深度学习探索性训练框架](https://note.microsoft.com/MSR-Webinar-Retiarii-Registration-Live.html) - *2021年6月24日* * **最新互动渠道**: [Discussions](https://github.com/microsoft/nni/discussions) @@ -249,7 +249,7 @@ Linux 和 macOS 下 NNI 系统需求[参考这里](https://nni.readthedocs.io/zh * 通过克隆源代码下载示例。 ```bash - git clone -b v2.5 https://github.com/Microsoft/nni.git + git clone -b v2.6 https://github.com/Microsoft/nni.git ``` * 运行 MNIST 示例。 diff --git a/docs/en_US/Release.rst b/docs/en_US/Release.rst index 44e692efb2..1bb467f745 100644 --- a/docs/en_US/Release.rst +++ b/docs/en_US/Release.rst @@ -5,6 +5,89 @@ Change Log ========== +Release 2.6 - 1/19/2022 +----------------------- + +**NOTE**: NNI v2.6 is the last version that supports Python 3.6. From next release NNI will require Python 3.7+. + +Hyper-Parameter Optimization +^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Experiment +"""""""""" + +* The legacy experiment config format is now deprecated. `(doc of new config) `__ + + * If you are still using legacy format, nnictl will show equivalent new config on start. Please save it to replace the old one. + +* nnictl now uses ``nni.experiment.Experiment`` `APIs `__ as backend. The output message of create, resume, and view commands have changed. +* Added Kubeflow and Frameworkcontroller support to hybrid mode. `(doc) `__ +* The hidden tuner manifest file has been updated. This should be transparent to users, but if you encounter issues like failed to find tuner, please try to remove ``~/.config/nni``. + +Algorithms +"""""""""" + +* Random tuner now supports classArgs ``seed``. `(doc) `__ +* TPE tuner is refactored: `(doc) `__ + + * Support classArgs ``seed``. + * Support classArgs ``tpe_args`` for expert users to customize algorithm behavior. + * Parallel optimization has been turned on by default. To turn it off set ``tpe_args.constant_liar_type`` to ``null`` (or ``None`` in Python). + * ``parallel_optimize`` and ``constant_liar_type`` has been removed. If you are using them please update your config to use ``tpe_args.constant_liar_type`` instead. + +* Grid search tuner now supports all search space types, including uniform, normal, and nested choice. `(doc) `__ + +Neural Architecture Search +^^^^^^^^^^^^^^^^^^^^^^^^^^ + +* Enhancement to serialization utilities `(doc) `__ and changes to recommended practice of customizing evaluators. `(doc) `__ +* Support latency constraint on edge device for ProxylessNAS based on nn-Meter. `(doc) `__ +* Trial parameters are showed more friendly in Retiarii experiments. +* Refactor NAS examples of ProxylessNAS and SPOS. + +Model Compression +^^^^^^^^^^^^^^^^^ + +* New Pruner Supported in Pruning V2 + + * Auto-Compress Pruner `(doc) `__ + * AMC Pruner `(doc) `__ + * Movement Pruning Pruner `(doc) `__ + +* Support ``nni.trace`` wrapped ``Optimizer`` in Pruning V2. In the case of not affecting the user experience as much as possible, trace the input parameters of the optimizer. `(doc) `__ +* Optimize Taylor Pruner, APoZ Activation Pruner, Mean Activation Pruner in V2 memory usage. +* Add more examples for Pruning V2. +* Add document for pruning config list. `(doc) `__ +* Parameter ``masks_file`` of ``ModelSpeedup`` now accepts `pathlib.Path` object. (Thanks to @dosemeion) `(doc) `__ +* Bug Fix + + * Fix Slim Pruner in V2 not sparsify the BN weight. + * Fix Simulator Annealing Task Generator generates config ignoring 0 sparsity. + +Documentation +^^^^^^^^^^^^^ + +* Supported GitHub feature "Cite this repository". +* Updated index page of readthedocs. +* Updated Chinese documentation. + + * From now on NNI only maintains translation for most import docs and ensures they are up to date. + +* Reorganized HPO tuners' doc. + +Bugfixes +^^^^^^^^ + +* Fixed a bug where numpy array is used as a truth value. (Thanks to @khituras) +* Fixed a bug in updating search space. +* Fixed a bug that HPO search space file does not support scientific notation and tab indent. + + * For now NNI does not support mixing scientific notation and YAML features. We are waiting for PyYAML to update. + +* Fixed a bug that causes DARTS 2nd order to crash. +* Fixed a bug that causes deep copy of mutation primitives (e.g., LayerChoice) to crash. +* Removed blank at bottom in Web UI overview page. + Release 2.5 - 11/2/2021 ----------------------- diff --git a/docs/en_US/Tutorial/InstallationLinux.rst b/docs/en_US/Tutorial/InstallationLinux.rst index 251c44dfdf..b29e219310 100644 --- a/docs/en_US/Tutorial/InstallationLinux.rst +++ b/docs/en_US/Tutorial/InstallationLinux.rst @@ -24,7 +24,7 @@ Install NNI through source code .. code-block:: bash - git clone -b v2.5 https://github.com/Microsoft/nni.git + git clone -b v2.6 https://github.com/Microsoft/nni.git cd nni python3 -m pip install -U -r dependencies/setup.txt python3 -m pip install -r dependencies/develop.txt @@ -38,7 +38,7 @@ If you want to perform a persist install instead, we recommend to build your own .. code-block:: bash - git clone -b v2.5 https://github.com/Microsoft/nni.git + git clone -b v2.6 https://github.com/Microsoft/nni.git cd nni export NNI_RELEASE=2.0 python3 -m pip install -U -r dependencies/setup.txt @@ -61,7 +61,7 @@ Verify installation .. code-block:: bash - git clone -b v2.5 https://github.com/Microsoft/nni.git + git clone -b v2.6 https://github.com/Microsoft/nni.git * Run the MNIST example. diff --git a/docs/en_US/Tutorial/InstallationWin.rst b/docs/en_US/Tutorial/InstallationWin.rst index 1935bcd213..383ec09c2c 100644 --- a/docs/en_US/Tutorial/InstallationWin.rst +++ b/docs/en_US/Tutorial/InstallationWin.rst @@ -40,7 +40,7 @@ If you want to contribute to NNI, refer to `setup development environment What's NEW!