ngraph.ai theme (#2654)

* [ONNX] Add documentation

* Update documentation contributor's instructions

* Doc theme to match ngraph.ai

* Minor formatting fixes and PR feedback

* ToC fixes

* ToC fixes

* Add changes

* Be consistent with BUILDDIR

* Be consistent with substitution

* Update Makefile

doc/sphinx/Makefile

+#
 # Robust Makefile for Sphinx documentation
 #
 
 # You can set these variables from the command line.
 SPHINXOPTS    =
 SPHINXBUILD   = sphinx-build
-SPHINXPROJ    = IntelnGraphlibrary
+SPHINXPROJ    = nGraphCompilerStack
 SOURCEDIR     = source
 BUILDDIR      = build
+ALLSPHINXOPTS = ${SOURCEDIR}
 
 # Put it first so that "make" without argument is like "make help".
 help:

doc/sphinx/source/backend-support/index.rst

@@ -25,6 +25,15 @@ framework developer customize targeted solutions. Experimental APIs to support
 current and future nGraph Backends are also available; see, for example, the 
 section on the :ref:`plaidml_backend`.
 
+.. csv-table::
+   :header: "Backend", "Current nGraph support", "Future nGraph support"
+   :widths: 35, 10, 10
+
+   Intel® Architecture Processors (CPUs), Yes, Yes
+   Intel® Nervana™ Neural Network Processor™ (NNPs), Yes, Yes
+   NVIDIA\* CUDA (GPUs), Yes, Some 
+   AMD\* GPUs, Yes, Some
+
 
 .. _hybrid_transformer:
 

doc/sphinx/source/buildlb.rst

 .. buildlb.rst:
 
-###########################
-nGraph Library for backends 
-###########################
+###############
+Build and Test 
+###############
 
 This section details how to build the C++ version of the nGraph Library, which 
 is targeted toward developers working on kernel-specific operations, 

doc/sphinx/source/conf.py

@@ -73,11 +73,11 @@ author = 'Intel Corporation'
 # built documents.
 #
 # The short X.Y version.
-version = '0.15'
+version = '0.16'
 
 # The Documentation full version, including alpha/beta/rc tags. Some features
 # available in the latest code will not necessarily be documented first
-release = '0.15.0'
+release = '0.16.1'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.
@@ -171,7 +171,7 @@ latex_elements = {
 # (source start file, target name, title,
 #  author, documentclass [howto, manual, or own class]).
 latex_documents = [
-    (master_doc, 'IntelnGraphlibrary.tex', 'Intel nGraph Library',
+    (master_doc, 'nGraphCompilerStack.tex', 'nGraph Compiler Stack Documentation',
      'Intel Corporation', 'manual'),
 ]
 
@@ -181,11 +181,10 @@ latex_documents = [
 # One entry per manual page. List of tuples
 # (source start file, name, description, authors, manual section).
 man_pages = [
-    (master_doc, 'intelngraphlibrary', 'Intel nGraph Library',
+    (master_doc, 'ngraphcompiler', 'nGraph Compiler stack',
      [author], 1)
 ]
 
-
 # -- Options for Texinfo output -------------------------------------------
 
 # Grouping the document tree into Texinfo files. List of tuples

doc/sphinx/source/frameworks/generic-configs.rst

@@ -8,18 +8,6 @@ This section details some of the *configuration options* and some of the
 your system already has a version of nGraph installed with one of our supported
 backends. 
 
-.. csv-table::
-   :header: "Backend", "Current nGraph support", "Future nGraph support"
-   :widths: 35, 10, 10
-
-   Intel® Architecture Processors (CPUs), Yes, Yes
-   Intel® Nervana™ Neural Network Processor™ (NNPs), Yes, Yes
-   NVIDIA\* CUDA (GPUs), Yes, Some 
-   :abbr:`Field Programmable Gate Arrays (FPGA)` (FPGAs), Coming soon, Yes
-   `Movidius`_, Not yet, Yes
-   Other, Not yet, Ask
-
-
 Regardless of the framework, after the :doc:`../buildlb` step, a good place 
 to start usually involves making the libraries available to the framework. On 
 Linux\* systems built on Intel® Architecture, that command tends to looks 

doc/sphinx/source/frameworks/onnx_integ.rst

@@ -19,17 +19,26 @@
 ONNX Support
 ============
 
-nGraph is able to import and execute ONNX models. Models are converted to nGraph's internal representation and converted to ``Function`` objects, which can be compiled and executed on one of nGraph's backends.
 
-You can use nGraph's Python API to run an ONNX model and nGraph can be used as an ONNX backend using the add-on package `nGraph-ONNX <ngraph_onnx>`_.
+nGraph is able to import and execute ONNX models. Models are converted to 
+nGraph's internal representation and converted to ``Function`` objects, which 
+can be compiled and executed on one of nGraph's backends.
 
-.. note:: In order to support ONNX, nGraph must be built with the ``NGRAPH_ONNX_IMPORT_ENABLE`` flag. See `Building nGraph-ONNX <ngraph_onnx_building>`_ for more information. All nGraph packages published on PyPI are built with ONNX support.
+You can use nGraph's Python API to run an ONNX model and nGraph can be used 
+as an ONNX backend using the add-on package `nGraph-ONNX <ngraph_onnx>`_.
+
+
+.. note:: In order to support ONNX, nGraph must be built with the 
+   ``NGRAPH_ONNX_IMPORT_ENABLE`` flag. See `Building nGraph-ONNX 
+   <ngraph_onnx_building>`_ for more information. All nGraph packages 
+   published on PyPI are built with ONNX support.
 
 
 Installation
 ------------
 
-In order to prepare your environment to use nGraph and ONNX, install the Python packages for nGraph, ONNX and NumPy:
+To prepare your environment to use nGraph and ONNX, install the Python packages
+for nGraph, ONNX and NumPy:
 
 ::
 
@@ -48,7 +57,8 @@ For example ResNet-50:
     $ tar -xzvf resnet50.tar.gz
 
 
-Use the following Python commands to convert the downloaded model to an nGraph ``Function``:
+Use the following Python commands to convert the downloaded model to an nGraph 
+``Function``:
 
 .. code-block:: python
 
@@ -65,14 +75,16 @@ Use the following Python commands to convert the downloaded model to an nGraph `
     <Function: 'resnet50' ([1, 1000])>
 
 
-This creates an nGraph ``Function`` object, which can be used to execute a computation on a chosen backend.
-
+This creates an nGraph ``Function`` object, which can be used to execute a 
+computation on a chosen backend.
 
 Running a computation
 ---------------------
 
-You can now create an nGraph ``Runtime`` backend and use it to compile your ``Function`` to a backend-specific ``Computation`` object.
-Finally, you can execute your model by calling the created ``Computation`` object with input data.
+You can now create an nGraph ``Runtime`` backend and use it to compile your 
+``Function`` to a backend-specific ``Computation`` object. Finally, you can 
+execute your model by calling the created ``Computation`` object with input 
+data:
 
 .. code-block:: python
 
@@ -94,7 +106,8 @@ Finally, you can execute your model by calling the created ``Computation`` objec
              ...
 
 
-You can find more information about nGraph and ONNX in the `nGraph-ONNX <ngraph_onnx>`_ GitHub repository.
+Find more information about nGraph and ONNX in the 
+`nGraph-ONNX <ngraph_onnx>`_ GitHub repository.
 
 
 .. _ngraph_onnx: https://github.com/NervanaSystems/ngraph-onnx/

doc/sphinx/source/index.rst

@@ -22,15 +22,24 @@ nGraph Compiler stack
 ######################
 
 
-.. toctree::
-   :maxdepth: 1 
+`nGraph`_ is an open-source graph compiler for :abbr:`Artificial Neural Networks (ANNs)`. 
+The nGraph Compiler stack provides an inherently efficient graph-based compilation 
+infrastructure designed to be compatible with the many of the upcoming 
+:abbr:`Application-Specific Integrated Circuits (ASICs)`, like the Intel® Nervana™ 
+Neural Network Processor (Intel® Nervana™ NNP), while also unlocking a massive 
+performance boost on any existing hardware targets in your neural network: both GPUs 
+and CPUs. Using its flexible infrastructure, you will find it becomes much easier 
+to create Deep Learning (DL) models that can adhere to the "write once, run anywhere" 
+mantra that enables your AI solutions to easily go from concept to production to scale.
 
-   project/introduction.rst
+Frameworks using nGraph to execute workloads have shown `up to 45X`_ performance 
+boost compared to native implementations. For a high-level overview, see the 
+:doc:`project/introduction`.
 
 
 .. toctree::
    :maxdepth: 1
-   :caption: Framework Support
+   :caption: Connecting Frameworks
    
    frameworks/index.rst
    frameworks/validated/list.rst
@@ -41,16 +50,16 @@ nGraph Compiler stack
    :maxdepth: 1
    :caption: nGraph Core
 
+   buildlb.rst
    core/overview.rst
    core/fusion/index.rst
    nGraph Core Ops <ops/index.rst>
    core/constructing-graphs/index.rst
    core/passes/passes.rst
-   buildlb.rst
    
 .. toctree::
    :maxdepth: 1
-   :caption: Python API
+   :caption: nGraph Python API
 
    python_api/index.rst
 
@@ -65,7 +74,7 @@ nGraph Compiler stack
 
 .. toctree::
    :maxdepth: 1
-   :caption: Distributed training
+   :caption: Distributed Training
 
    distr/index.rst
 
@@ -83,13 +92,14 @@ nGraph Compiler stack
    :maxdepth: 1
    :caption: Tutorials
 
-   tutorials/index.rst
+   nGraph.ai Tutorials <https://www.ngraph.ai/tutorials>   
 
 
 .. toctree::
    :maxdepth: 1
    :caption: Project Metadata
 
+   project/introduction.rst
    project/release-notes.rst
    project/contribution-guide.rst
    project/governance.rst
@@ -102,3 +112,9 @@ Indices and tables
 
    * :ref:`search`
    * :ref:`genindex`
+
+
+
+
+.. _nGraph: https://www.ngraph.ai
+.. _up to 45X: https://ai.intel.com/ngraph-compiler-stack-beta-release/
\ No newline at end of file

doc/sphinx/source/project/introduction.rst

@@ -118,14 +118,6 @@ expand your network's hardware. Each integration is unique to the framework
 and its set of deep learning operators, its view on memory layout, its 
 feature set, etc.
 
-.. _figure-B:
-
-.. figure:: ../graphics/intro_kernel_to_fw_accent.png
-   :width: 555px
-   :alt: 
-
-   Each of these connections represents significant work for what will 
-   ultimately be a brittle setup that is enormously expensive to maintain.
 
 nGraph solves this problem with nGraph bridges. A bridge takes a computational 
 graph and reconstructs it in the nGraph IR with a few primitive nGraph 
@@ -146,12 +138,16 @@ of each parameter for each operation. In the past, the number of required
 kernels was limited, but as the AI research and industry rapidly develops, the 
 final product of required kernels is increasing exponentially. 
 
-.. _figure-C:
+.. _figure-B:
 
 .. figure:: ../graphics/intro_kernel_explosion.png
    :width: 555px
    :alt: 
 
+   Each of these connections represents significant work for what will 
+   ultimately be a brittle setup that is enormously expensive to maintain.
+
+
 
 PlaidML addresses the kernel explosion problem in a manner that lifts a heavy 
 burden off kernel developers. It automatically lowers networks from nGraph 

doc/sphinx/source/tutorials/index.rst

 .. tutorials/index:
 
+
+.. This will hold the organization of the tutorials we put on ngraph.ai  
+
+   .. it will need to be organized in a way that is navigable for the many kinds of frameworks and backends we support in the "Compiler stack".  It will need to be workable with a sitemap structure. The initial example is for the latest nGraph-TensorFlow bridge.  
+
+
+:orphan:
+
 ##########
 Tutorials
 ##########