Find the latest features, API, examples and tutorials in our **official documentation**: [NNI Doc](https://nni.readthedocs.io/)
NNI automates feature engineering, neural architecture search, hyperparameter tuning, and model compression for deep learning. Find the latest features, API, examples and tutorials in our **[official documentation](https://nni.readthedocs.io/) ([简体中文版点这里](https://nni.readthedocs.io/zh/))**. Quick links:
***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)
**Neural Network Intelligence (NNI)** is a lightweight and powerful toolkit to help users **automate**<ahref="https://nni.readthedocs.io/en/stable/FeatureEngineering/Overview.html">Feature Engineering</a>, <ahref="https://nni.readthedocs.io/en/stable/NAS/Overview.html">Neural Architecture Search</a>, <ahref="https://nni.readthedocs.io/en/stable/Tuner/BuiltinTuner.html">Hyperparameter Tuning</a> and <ahref="https://nni.readthedocs.io/en/stable/Compression/Overview.html">Model Compression</a>.
NNI provides CommandLine Tool as well as an user friendly WebUI to manage training experiments. The tool manages automated machine learning (AutoML) experiments, **dispatches and runs** experiments' trial jobs generated by tuning algorithms to search the best neural architecture and/or hyper-parameters in **different training environments** like <ahref="https://nni.readthedocs.io/en/stable/TrainingService/LocalMode.html">Local Machine</a>, <ahref="https://nni.readthedocs.io/en/stable/TrainingService/RemoteMachineMode.html">Remote Servers</a>, <ahref="https://nni.readthedocs.io/en/stable/TrainingService/PaiMode.html">OpenPAI</a>, <ahref="https://nni.readthedocs.io/en/stable/TrainingService/KubeflowMode.html">Kubeflow</a>, <ahref="https://nni.readthedocs.io/en/stable/TrainingService/FrameworkControllerMode.html">FrameworkController on K8S (AKS etc.)</a>, <ahref="https://nni.readthedocs.io/en/stable/TrainingService/DLTSMode.html">DLWorkspace (aka. DLTS)</a>, <ahref="https://nni.readthedocs.io/en/stable/TrainingService/AMLMode.html">AML (Azure Machine Learning)</a>, <ahref="https://nni.readthedocs.io/en/stable/TrainingService/AdaptDLMode.html">AdaptDL (aka. ADL)</a> , other cloud options and even <ahref="https://nni.readthedocs.io/en/stable/TrainingService/HybridMode.html">Hybrid mode</a>.
With the extensible API, you can customize your own AutoML algorithms and training services. To make it easy for new users, NNI also provides a set of build-in state-of-the-art AutoML algorithms and out of box support for popular training platforms.
Within the following table, we summarized the current NNI capabilities, we are gradually adding new capabilities and we'd love to have your contribution.
* Those who want to **try different AutoML algorithms** in their training code/model.
* Those who want to run AutoML trial jobs **in different environments** to speed up search.
* Researchers and data scientists who want to easily **implement and experiment new AutoML algorithms**, may it be: hyperparameter tuning algorithm, neural architect search algorithm or model compression algorithm.
* ML Platform owners who want to **support AutoML in their platform**.
## **Installation**
### **Install**
NNI supports and is tested on Ubuntu >= 18.04, Windows 10 >= 21H2, and macOS >= 11.
Simply run the following `pip install` in an environment that has `python 64-bit >= 3.7`.
If you want to try latest code, please [install NNI](https://nni.readthedocs.io/en/stable/installation.html) from source code.
(TBD: figures and tables)
For detail system requirements of NNI, please refer to [here](https://nni.readthedocs.io/en/stable/Tutorial/InstallationLinux.html#system-requirements) for Linux & macOS, and [here](https://nni.readthedocs.io/en/stable/Tutorial/InstallationWin.html#system-requirements) for Windows.
## Installation
<details>
<summary> Installation FAQ </summary>
* If there is any privilege issue, add `--user` to install NNI in the user directory.
* Currently NNI on Windows supports local, remote and pai mode. Anaconda or Miniconda is highly recommended to install [NNI on Windows](https://nni.readthedocs.io/en/stable/Tutorial/InstallationWin.html).
* If there is any error like `Segmentation fault`, please refer to [FAQ](https://nni.readthedocs.io/en/stable/Tutorial/FAQ.html). For FAQ on Windows, please refer to [NNI on Windows](https://nni.readthedocs.io/en/stable/Tutorial/InstallationWin.html#faq).
</details>
### **Run your first example**
<details>
<summary> set up and run the example </summary>
* Download the examples via clone the source code.
See the [NNI installation guide](https://nni.readthedocs.io/en/stable/installation.html) to install from pip, or build from source.
*Wait for the message `INFO: Successfully started experiment!` in the command line. This message indicates that your experiment has been successfully started. You can explore the experiment using the `Web UI url`.
Wait for the message `INFO: Successfully started experiment!` in the command line. This message indicates that your experiment has been successfully started. You can explore the experiment using the `Web UI url` shown in the console.
* Open the `Web UI url` in your browser, you can view detailed information of the experiment and all the submitted trial jobs as shown below. [Here](https://nni.readthedocs.io/en/stable/Tutorial/WebUI.html) are more Web UI pages.
NNI has a monthly release cycle (major releases). Please let us know if you encounter a bug by [filling an issue](https://github.com/microsoft/nni/issues/new/choose).
We appreciate all contributions. If you are planning to contribute any bug-fixes, please do so without further discussions.
If you plan to contribute new features, new tuners, new training services, etc. please first open an issue or reuse an exisiting issue, and discuss the feature with us. We will discuss with you on the issue timely or set up conference calls if needed.
To learn more about making a contribution to NNI, please refer to our [How-to contribution page](https://nni.readthedocs.io/en/stable/contribution.html).
For more usages, please see [NNI tutorials](https://nni.readthedocs.io/en/stable/tutorials.html).
We appreciate all contributions and thank all the contributors!
If you want to contribute to NNI, be sure to review the [contribution guidelines](https://nni.readthedocs.io/en/stable/notes/contributing.html), which includes instructions of submitting feedbacks, best coding practices, and code of conduct.
We use [GitHub issues](https://github.com/microsoft/nni/issues) to track tracking requests and bugs.
Please use [NNI Discussion](https://github.com/microsoft/nni/discussions) for general questions and new ideas.
For questions of specific use cases, please go to [Stack Overflow](https://stackoverflow.com/questions/tagged/nni).
## **Feedback**
*[File an issue](https://github.com/microsoft/nni/issues/new/choose) on GitHub.
* Open or participate in a [discussion](https://github.com/microsoft/nni/discussions).
* Discuss on the NNI [Gitter](https://gitter.im/Microsoft/nni?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge) in NNI.
Participating discussions via the following IM groups is also welcomed.
Join IM discussion groups:
|Gitter||WeChat|
|----|----|----|
|| OR ||
Over the past few years, NNI has received thousands of feedbacks on GitHub issues, and pull requests from hundreds of contributors.
We appreciate all contributions from community to make NNI thrive.
* **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>`_
<a href="TrainingService/AdaptDLMode.html">AdaptDL (aka. ADL)</a>, other cloud options and even <a href="TrainingService/HybridMode.html">Hybrid mode</a>.
</p>
<!-- Who should consider using NNI -->
<div>
<h2 class="title">Who should consider using NNI</h2>
<ul>
<li>Those who want to <b>try different AutoML algorithms</b> in their training code/model.</li>
<li>Those who want to run AutoML trial jobs <b>in different environments</b> to speed up search.</li>
<li class="rowHeight">Researchers and data scientists who want to easily <b>implement and experiement new AutoML
algorithms</b>
, may it be: hyperparameter tuning algorithm,
neural architect search algorithm or model compression algorithm.
</li>
<li>ML Platform owners who want to <b>support AutoML in their platform</b></li>
<li><a href="https://github.com/microsoft/nni/issues/new/choose">File an issue</a> on GitHub.</li>
<li>Open or participate in a <a href="https://github.com/microsoft/nni/discussions">discussion</a>.</li>
<li>Discuss on the <a href="https://gitter.im/Microsoft/nni?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge">NNI Gitter</a> in NNI.</li>
</ul>
<div>
<div class="rowHeight">Join IM discussion groups:</div>
Targeting at openness and advancing state-of-art technology,
<a href="https://www.microsoft.com/en-us/research/group/systems-and-networking-research-group-asia/">Microsoft Research (MSR)</a>
had also released few
other open source projects.</p>
<ul id="relatedProject">
<li class="rowHeight">
<a href="https://github.com/Microsoft/pai">OpenPAI</a> : an open source platform that provides complete AI model
training and resource management
capabilities, it is easy to extend and supports on-premise,
cloud and hybrid environments in various scale.
</li>
<li class="rowHeight">
<a href="https://github.com/Microsoft/frameworkcontroller">FrameworkController</a> : an open source
general-purpose Kubernetes Pod Controller that orchestrate
all kinds of applications on Kubernetes by a single controller.
</li>
<li class="rowHeight">
<a href="https://github.com/Microsoft/MMdnn">MMdnn</a> : A comprehensive, cross-framework solution to convert,
visualize and diagnose deep neural network
models. The "MM" in MMdnn stands for model management
and "dnn" is an acronym for deep neural network.
</li>
<li class="rowHeight">
<a href="https://github.com/Microsoft/SPTAG">SPTAG</a> : Space Partition Tree And Graph (SPTAG) is an open
source library
for large scale vector approximate nearest neighbor search scenario.
</li>
<li class="rowHeight">
<a href="https://github.com/Microsoft/SPTAG">nn-Meter</a> : An accurate inference latency predictor for DNN models on diverse edge devices.
</li>
</ul>
<p>We encourage researchers and students leverage these projects to accelerate the AI development and research.</p>
</div>
<!-- License -->
<div>
<h2 class="title">License</h2>
<p>The entire codebase is under <a href="https://github.com/microsoft/nni/blob/master/LICENSE">MIT license</a></p>
</div>
</div>
<h2>Get Support and Contribute Back</h2>
NNI is maintained on the `NNI GitHub repository <https://github.com/microsoft/nni>`_. We collect feedbacks and new proposals/ideas on GitHub. You can:
* Open a `GitHub issue <https://github.com/microsoft/nni/issues>`_ for bugs and feature requests.
* Open a `pull request <https://github.com/microsoft/nni/pulls>`_ to contribute code (make sure to read the `contribution guide </contribution>` before doing this).
* Participate in `NNI Discussion <https://github.com/microsoft/nni/discussions>`_ for general questions and new ideas.
