[ONNX 3] Add ONNX registry tutorial

Co-authored-by: Thiago Crepaldi <thiagofc@microsoft.com>

Author: 2 people

_static/img/onnx/custom_addandround_function.png

@@ -2,11 +2,11 @@
 (optional) Exporting a Model from PyTorch to ONNX and Running it using ONNX Runtime
 ===================================================================================
 
-.. Note::
+.. note::
     As of PyTorch 2.1, there are two versions of ONNX Exporter.
 
-    * ``torch.onnx.dynamo_export`is the newest (still in beta) exporter based on the TorchDynamo technology released with PyTorch 2.0
-    * ``torch.onnx.export`` is based on TorchScript backend and has been available since PyTorch 1.2.0
+    * ``torch.onnx.dynamo_export`is the newest (still in beta) exporter based on the TorchDynamo technology released with PyTorch 2.0.
+    * ``torch.onnx.export`` is based on TorchScript backend and has been available since PyTorch 1.2.0.
 
 In this tutorial, we describe how to convert a model defined
 in PyTorch into the ONNX format using the TorchScript ``torch.onnx.export` ONNX exporter.

_static/img/onnx/custom_addandround_model.png

@@ -8,3 +8,7 @@ ONNX
 2. export_simple_model_to_onnx_tutorial.py
     Export a PyTorch model to ONNX
     https://pytorch.org/tutorials/beginner/onnx/export_simple_model_to_onnx_tutorial.html
+
+3. onnx_registry_tutorial.py
+    Introduction to ONNX
+    https://pytorch.org/tutorials/beginner/onnx/onnx_registry_tutorial.html

_static/img/onnx/custom_aten_add_function.png

@@ -1,7 +1,8 @@
 # -*- coding: utf-8 -*-
 """
 `Introduction to ONNX <intro_onnx.html>`_ ||
-**Export a PyTorch model to ONNX**
+**Export a PyTorch model to ONNX** ||
+`Introduction to ONNX Registry <onnx_registry_tutorial.html>`_
 
 Export a PyTorch model to ONNX
 ==============================
@@ -104,7 +105,7 @@ def forward(self, x):
 export_output.save("my_image_classifier.onnx")
 
 ######################################################################
-# The ONNX file can be loaded back into memory and checked if it is well formed with the following code:
+# You can load the ONNX file back into memory and check if it is well formed with the following code:
 
 import onnx
 onnx_model = onnx.load("my_image_classifier.onnx")

_static/img/onnx/custom_aten_add_model.png

@@ -1,6 +1,7 @@
 """
 **Introduction to ONNX** ||
-`Export a PyTorch model to ONNX <export_simple_model_to_onnx_tutorial.html>`_
+`Export a PyTorch model to ONNX <export_simple_model_to_onnx_tutorial.html>`_ ||
+`Introduction to ONNX Registry <onnx_registry_tutorial.html>`_
 
 Introduction to ONNX
 ====================
@@ -32,13 +33,21 @@
 Dependencies
 ------------
 
+PyTorch 2.1.0 or newer is required.
+
 The ONNX exporter depends on extra Python packages:
 
   - `ONNX <https://onnx.ai>`_
   - `ONNX Script <https://onnxscript.ai>`_
 
 They can be installed through `pip <https://pypi.org/project/pip/>`_:
 
+.. note::
+   This tutorial leverages `onnxscript <https://github.com/microsoft/onnxscript#readme>`__
+   to create custom ONNX operators. onnxscript is a Python library that allows users to
+   create custom ONNX operators in Python. It is a prerequisite learning material for
+   this tutorial. Please make sure you have read the onnxscript tutorial before proceeding.
+
 .. code-block:: bash
 
   pip install --upgrade onnx onnxscript