Merge branch 'main' into ir

Author: Svetlana Karslioglu

en-wordlist.txt

@@ -52,6 +52,7 @@ CPUs
 CPython
 CUDA
 Caffe
+callable's
 Captum
 Captum's
 CartPole
@@ -72,6 +73,8 @@ DQN
 DataPipe
 DataPipes
 DataLoaders
+Decompositions
+decompositions
 DeepMind
 DeiT
 DenseNet
@@ -160,6 +163,8 @@ ONNX Runtime
 ONNX Script
 OpenAI
 OpenMP
+Opset
+opset
 Ornstein
 OU
 PIL

intermediate_source/_torch_export_nightly_tutorial.py

@@ -33,7 +33,7 @@
 #
 # ``torch.export`` extracts single-graph representations from PyTorch programs
 # by tracing the target function, given example inputs.
-# ``torch.export.export()`` is the main entrypoint for ``torch.export``.
+# ``torch.export.export()`` is the main entry point for ``torch.export``.
 #
 # In this tutorial, ``torch.export`` and ``torch.export.export()`` are practically synonymous,
 # though ``torch.export`` generally refers to the PyTorch 2.X export process, and ``torch.export.export()``
@@ -546,9 +546,9 @@ def forward(self, x):
 # `here <https://github.com/pytorch/pytorch/blob/b460c3089367f3fadd40aa2cb3808ee370aa61e1/torch/_decomp/decompositions.py>`__.
 # If you would like to use some of these existing decomposition functions,
 # you can pass in a list of operators you would like to decompose to the
-# :func:`get_decompositions <https://github.com/pytorch/pytorch/blob/b460c3089367f3fadd40aa2cb3808ee370aa61e1/torch/_decomp/__init__.py#L191>`__
-# function, which will return a decomposition table using the pre-implemented
-# decompositions.
+# `get_decompositions <https://github.com/pytorch/pytorch/blob/b460c3089367f3fadd40aa2cb3808ee370aa61e1/torch/_decomp/__init__.py#L191>`__
+# function, which will return a decomposition table using existing
+# decomposition implementations.
 
 class M(torch.nn.Module):
     def __init__(self):

intermediate_source/torch_export_nightly_tutorial.rst

@@ -30,7 +30,7 @@ Basic Usage
 
 ``torch.export`` extracts single-graph representations from PyTorch programs
 by tracing the target function, given example inputs.
-``torch.export.export()`` is the main entrypoint for ``torch.export``.
+``torch.export.export()`` is the main entry point for ``torch.export``.
 
 In this tutorial, ``torch.export`` and ``torch.export.export()`` are practically synonymous,
 though ``torch.export`` generally refers to the PyTorch 2.X export process, and ``torch.export.export()``
@@ -770,9 +770,9 @@ Most ATen operators already have decompositions, which are located
 `here <https://github.com/pytorch/pytorch/blob/b460c3089367f3fadd40aa2cb3808ee370aa61e1/torch/_decomp/decompositions.py>`__.
 If you would like to use some of these existing decomposition functions,
 you can pass in a list of operators you would like to decompose to the
-:func:`get_decompositions <https://github.com/pytorch/pytorch/blob/b460c3089367f3fadd40aa2cb3808ee370aa61e1/torch/_decomp/__init__.py#L191>`__
-function, which will return a decomposition table using the pre-implemented
-decompositions.
+`get_decompositions <https://github.com/pytorch/pytorch/blob/b460c3089367f3fadd40aa2cb3808ee370aa61e1/torch/_decomp/__init__.py#L191>`__
+function, which will return a decomposition table using existing
+decomposition implementations.
 
 .. code-block:: python
 

prototype_source/prototype_index.rst

@@ -97,7 +97,7 @@ Prototype features are not available as part of binary distributions like PyPI o
    :card_description: Prune BERT to be 2:4 sparse and accelerate for inference.
    :image: _static/img/thumbnails/cropped/generic-pytorch-logo.png
    :link: prototype/semi_structured_sparse.html
-   :tags: Model-Optimziation
+   :tags: Model-Optimiziation
 
 .. Mobile