Merge branch 'main' into add-meta-tags

Author: svekars

beginner_source/onnx/intro_onnx.py

@@ -39,13 +39,14 @@
 
   - `ONNX <https://onnx.ai>`_ standard library
   - `ONNX Script <https://onnxscript.ai>`_ library that enables developers to author ONNX operators,
-    functions and models using a subset of Python in an expressive, and yet simple fashion.
+    functions and models using a subset of Python in an expressive, and yet simple fashion
+  - `ONNX Runtime <https://onnxruntime.ai>`_ accelerated machine learning library.
 
 They can be installed through `pip <https://pypi.org/project/pip/>`_:
 
 .. code-block:: bash
 
-  pip install --upgrade onnx onnxscript
+  pip install --upgrade onnx onnxscript onnxruntime
 
 To validate the installation, run the following commands:
 

prototype_source/gpu_quantization_torchao_tutorial.py

@@ -35,14 +35,12 @@
 #
 # Segment Anything Model checkpoint setup:
 #
-# 1. Go to the `segment-anything repo <checkpoint https://github.com/facebookresearch/segment-anything/tree/main#model-checkpoints>`_ and download the ``vit_h`` checkpoint. Alternatively, you can just use ``wget``: `wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth --directory-prefix=<path>
+# 1. Go to the `segment-anything repo checkpoint <https://github.com/facebookresearch/segment-anything/tree/main#model-checkpoints>`_ and download the ``vit_h`` checkpoint. Alternatively, you can use ``wget`` (for example, ``wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth --directory-prefix=<path>``).
 # 2. Pass in that directory by editing the code below to say:
 #
-# .. code-block::
-#
-# {sam_checkpoint_base_path}=<path>
+# .. code-block:: bash
 #
-# This was run on an A100-PG509-200 power limited to 330.00 W
+#   {sam_checkpoint_base_path}=<path>
 #
 
 import torch
@@ -297,7 +295,7 @@ def get_sam_model(only_one_block=False, batchsize=1):
 # -----------------
 # In this tutorial, we have learned about the quantization and optimization techniques
 # on the example of the segment anything model.
-
+#
 # In the end, we achieved a full-model apples to apples quantization speedup
 # of about 7.7% on batch size 16 (677.28ms to 729.65ms). We can push this a
 # bit further by increasing the batch size and optimizing other parts of