Add doc on iGPU to v0.22 doc

doc/sphinx/source/conf.py

@@ -73,11 +73,11 @@ author = 'Intel Corporation'
 # built documents.
 #
 # The short X.Y version.
-version = '0.21'
+version = '0.22'
 
 # The Documentation full version, including alpha/beta/rc tags. Some features
 # available in the latest code will not necessarily be documented first
-release = '0.21.0'
+release = '0.22-doc'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

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

@@ -20,15 +20,45 @@ something like:
 
 
 Find or display nGraph Version
--------------------------------
-
+==============================
 
+If you're working with the :doc:`../python_api/index`, the following command 
+may be useful:
 
 .. code-block:: console
 
    python3 -c "import ngraph as ng; print('nGraph version: ',ng.__version__)";
 
 
+Localize environment variables
+==============================
+
+Another generic configuration option is to activate ``NGRAPH_CPU_DEBUG_TRACER``, 
+a runtime environment variable and useful tool for data scientists. 
+
+To activate this tool, set the ``env`` var ``NGRAPH_CPU_DEBUG_TRACER=1``.
+It will dump ``trace_meta.log`` and ``trace_bin_data.log``. 
+
+To specify the names of logs with those flags:
+
+:: 
+
+  NGRAPH_TRACER_LOG = "meta.log"
+  NGRAPH_BIN_TRACER_LOG = "bin.log"
+
+The meta_log line contains::
+ 
+  kernel_name, serial_number_of_op, tensor_id, symbol_of_in_out, num_elements, shape, binary_data_offset, mean_of_tensor, variance_of_tensor
+
+A line example from unit-test might look like::
+
+  K=Add S=0 TID=0_0 >> size=4 Shape{2, 2} bin_data_offset=8 mean=1.5 var=1.25
+
+The binary_log line contains::
+
+  tensor_id, binary data (tensor data)
+
+
 FMV
 ---
 

doc/sphinx/source/frameworks/onnx_integ.rst

@@ -10,7 +10,7 @@ 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>`_.
+as an ONNX backend using the add-on package `nGraph ONNX`_.
 
 
 .. note:: In order to support ONNX, nGraph must be built with the 
@@ -33,8 +33,7 @@ for nGraph, ONNX and NumPy:
 Importing an ONNX model
 -----------------------
 
-You can download models from the `ONNX Model Zoo <onnx_model_zoo_>`_.
-For example ResNet-50:
+You can download models from the `ONNX Model Zoo`_. For example, ResNet-50:
 
 ::
 
@@ -92,9 +91,9 @@ data:
 
 
 Find more information about nGraph and ONNX in the 
-`nGraph-ONNX <ngraph_onnx>`_ GitHub repository.
+`nGraph ONNX`_ GitHub repository.
 
 
-.. _ngraph_onnx: https://github.com/NervanaSystems/ngraph-onnx/
-.. _ngraph_onnx_building: https://github.com/NervanaSystems/ngraph-onnx/blob/master/BUILDING.md
-.. _onnx_model_zoo: https://github.com/onnx/models
+.. _ngraph ONNX: https://github.com/NervanaSystems/ngraph-onnx
+.. _ngraph ONNX building: https://github.com/NervanaSystems/ngraph-onnx/blob/master/BUILDING.md
+.. _ONNX model zoo: https://github.com/onnx/models

doc/sphinx/source/inspection/index.rst

@@ -6,9 +6,9 @@ Visualization Tools
 nGraph provides serialization and deserialization facilities, along with the 
 ability to create image formats or a PDF. 
 
-When visualization is enabled, a ``dot`` file gets generated, along with a 
-``png``. The default can be adjusted by setting the 
-``NGRAPH_VISUALIZE_TREE_OUTPUT_FORMAT`` flag to another format, like PDF. 
+When visualization is enabled, ``svg`` files for your graph get generated. The 
+default can be adjusted by setting the ``NGRAPH_VISUALIZE_TRACING_FORMAT`` 
+flag to another format, like PNG or PDF. 
 
 .. note:: Large graphs are usually not legible with formats like PDF.
 

doc/sphinx/source/project/doc-contributor-README.rst

@@ -18,9 +18,36 @@
 Contributing to documentation
 =============================
 
-.. important:: Read this for changes affecting **anything** in ``ngraph/doc``
+.. note:: Tips for contributors who are new to the highly-dynamic 
+   environment of documentation in AI software:
 
-For updates to the nGraph Library ``/doc`` repo, please submit a PR with 
+   * A good place to start is "document something you figured out how to 
+     get working". Content changes and additions should be targeted at 
+     something more specific than "developers". If you don't understand 
+     how varied and wide the audience is, you'll inadvertently break or block 
+     things.
+
+   * There are experts who work on all parts of the stack; try asking how 
+     documentation changes ought to be made in their respective sections.
+
+   * Start with something small. It is okay to add a "patch" to fix a typo 
+     or suggest a word change; larger changes to files or structure require 
+     research and testing first, as well as some logic for why you think 
+     something needs changed.
+
+   * Most documentation should wrap at about ``80``. We do our best to help 
+     authors source-link and maintain their own code and contributions; 
+     overwriting something already documented doesn't always improve it.
+
+   * Be careful editing files with links already present in them; deleting 
+     links to papers, citations, or sources is discouraged.
+
+   * Please do not submit Jupyter* notebook code to the nGraph Library
+     or core repos; best practice is to maintain any project-specific 
+     examples, tests, or walk-throughs in a separate repository and to link 
+     back to the stable ``op`` or Ops that you use in your project.
+
+For updates within the nGraph Library ``/doc`` repo, please submit a PR with 
 any changes or ideas you'd like integrated. This helps us maintain trackability 
 with respect to changes made, additions, deletions, and feature requests.
 
@@ -31,12 +58,11 @@ files to another format with a tool like ``pypandoc`` and share a link
 to your efforts on our `wiki`_. 
 
 Another option is to fork the `ngraph repo`_, essentially snapshotting it at 
-that point in time, and to build a Jupyter\* notebook or other set of docs around 
+that point in time, and to build Jupyter\* notebook or other set of docs around 
 it for a specific use case. Add a note on our wiki to show us what you 
 did; new and novel applications may have their projects highlighted on an 
 upcoming `ngraph.ai`_ release.   
 
-
 .. note:: Please do not submit Jupyter* notebook code to the nGraph Library 
    or core repos; best practice is to maintain any project-specific examples, 
    tests, or walk-throughs in a separate repository.
@@ -45,9 +71,11 @@ upcoming `ngraph.ai`_ release.
 Documenting source code examples 
 --------------------------------
 
-When **verbosely** documenting functionality of specific sections of code -- whether 
-they are entire code blocks within a file, or code strings that are **outside** 
-the nGraph Library's `documentation repo`_, here is an example of best practice: 
+When **verbosely** documenting functionality of specific sections of code -- 
+whether they are entire code blocks within a file, or code strings that are 
+**outside** the nGraph Library's `documentation repo`_, here is an example 
+of best practice: 
+
 
 Say a file has some interesting functionality that could benefit from more 
 explanation about one or more of the pieces in context. To keep the "in context" 
@@ -73,8 +101,7 @@ the folder where the ``Makefile`` is that generates the documentation you're
 writing. 
 
 See the **note** at the bottom of this page for more detail about how 
-this works in the current |version| version of Intel nGraph library 
-documentation. 
+this works in the current |version| version of nGraph Library documentation.
 
 
 Adding captions to code blocks 

python/BUILDING.md

@@ -2,8 +2,8 @@
 
 ## Building nGraph Python Wheels
 
-If you want to try a newer version of nGraph's Python API than is available from
-PyPI, you can build your own latest version from the source code. This 
+If you want to try a newer version of nGraph's Python API than is available 
+from PyPI, you can build your own latest version from the source code. This 
 process is very similar to what is outlined in our [ngraph_build] instructions 
 with two important differences:
 

python/README.md

@@ -18,8 +18,8 @@ boost compared to native implementations.
 
 nGraph can be used directly with the [Python API][api_python] described here, or 
 with the [C++ API][api_cpp] described in the [core documentation]. Alternatively, 
-its performance benefits can be realized through a frontend such as 
-[TensorFlow][frontend_tf], [MXNet][frontend_mxnet], and [ONNX][frontend_onnx].
+its performance benefits can be realized through frontends such as 
+[TensorFlow][frontend_tf], [PaddlePaddle][paddle_paddle] and [ONNX][frontend_onnx].
 You can also create your own custom framework to integrate directly with the 
 [nGraph Ops] for highly-targeted graph execution.
 
@@ -77,7 +77,7 @@ print('Result = ', result)
 
 [up to 45X]: https://ai.intel.com/ngraph-compiler-stack-beta-release/
 [frontend_onnx]: https://pypi.org/project/ngraph-onnx/
-[frontend_mxnet]: https://pypi.org/project/ngraph-mxnet/ 
+[paddle_paddle]: https://ngraph.nervanasys.com/docs/latest/frameworks/paddle_integ.html 
 [frontend_tf]: https://pypi.org/project/ngraph-tensorflow-bridge/
 [ngraph_github]: https://github.com/NervanaSystems/ngraph "nGraph on GitHub"
 [ngraph_building]: https://github.com/NervanaSystems/ngraph/blob/master/python/BUILDING.md "Building nGraph"