Merge branch 'main' into dl/fx/openvino_quantizer

Author: svekars

index.rst

@@ -3,13 +3,11 @@ Welcome to PyTorch Tutorials
 
 **What's new in PyTorch tutorials?**
 
-* `Dynamic Compilation Control with torch.compiler.set_stance <https://pytorch.org/tutorials/recipes/torch_compiler_set_stance_tutorial.html>`__
-* `Accelerating PyTorch Transformers by replacing nn.Transformer with Nested Tensors and torch.compile() <https://pytorch.org/tutorials/intermediate/transformer_building_blocks.html>`__
-* `Understanding the torch.export Flow and Solutions to Common Challenges <https://pytorch.org/tutorials/recipes/torch_export_challenges_solutions.html>`__
-* Updated `torch.export Tutorial <https://pytorch.org/tutorials/intermediate/torch_export_tutorial.html#constraints-dynamic-shapes>`__ with automatic dynamic shapes ``Dim.AUTO``
-* Updated `torch.export AOTInductor Tutorial for Python runtime <https://pytorch.org/tutorials/recipes/torch_export_aoti_python.html>`__
-* Updated `Using User-Defined Triton Kernels with torch.compile <https://pytorch.org/tutorials/recipes/torch_compile_user_defined_triton_kernel_tutorial.html#composability>`__ with new ``torch.library.triton_op``
-* Updated `Compile Time Caching in torch.compile <https://pytorch.org/tutorials/recipes/torch_compile_caching_tutorial.html>`__ with new ``Mega-Cache``
+* `Utilizing Torch Function modes with torch.compile <https://pytorch.org/tutorials/recipes/torch_compile_torch_function_modes.html>`__
+* `Context Parallel Tutorial <https://pytorch.org/tutorials/prototype/context_parallel.html>`__
+* `PyTorch 2 Export Quantization with Intel GPU Backend through Inductor <https://pytorch.org/tutorials/prototype/pt2e_quant_xpu_inductor.html>`__
+* `(beta) Explicit horizontal fusion with foreach_map and torch.compile <https://pytorch.org/tutorials/recipes/foreach_map.html>`__
+* Updated `Inductor Windows CPU Tutorial <https://pytorch.org/tutorials/prototype/inductor_windows.html>`__
 
 .. raw:: html
 

prototype_source/inductor_windows.rst

@@ -44,18 +44,21 @@ Next, let's configure our environment.
     
     "C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Auxiliary/Build/vcvars64.bat"
 #. Create and activate a virtual environment: ::
+
 #. Install `PyTorch 2.5 <https://pytorch.org/get-started/locally/>`_ or later for CPU Usage. Install PyTorch 2.7 or later refer to `Getting Started on Intel GPU <https://pytorch.org/docs/main/notes/get_start_xpu.html>`_ for XPU usage.
+
 #. Here is an example of how to use TorchInductor on Windows:
-.. code-block:: python
-
-    import torch
-    device="cpu" # or "xpu" for XPU
-    def foo(x, y):
-        a = torch.sin(x)
-        b = torch.cos(x)
-        return a + b
-    opt_foo1 = torch.compile(foo)
-    print(opt_foo1(torch.randn(10, 10).to(device), torch.randn(10, 10).to(device)))
+
+    .. code-block:: python
+
+        import torch
+        device="cpu" # or "xpu" for XPU
+        def foo(x, y):
+            a = torch.sin(x)
+            b = torch.cos(x)
+            return a + b
+        opt_foo1 = torch.compile(foo)
+        print(opt_foo1(torch.randn(10, 10).to(device), torch.randn(10, 10).to(device)))
 
 #. Below is the output of the above example::
 

recipes_source/foreach_map.py

@@ -1,6 +1,6 @@
 """
-(beta) Explicit horizontal fusion with foreach_map and torch.compile
-============================================================
+Explicit horizontal fusion with foreach_map and torch.compile
+===============================================================
 
 **Author:** `Michael Lazos <https://github.com/mlazos>`_
 """
@@ -13,11 +13,17 @@
 #  allows conversion of any pointwise op in ``torch`` to a horiztonally fused foreach
 #  variant. In this tutorial, we will demonstrate how to implement the Adam optimizer
 #  with ``foreach_map`` to generate a fully fused kernel.  
-# 
 #
 # .. note::
 #
-#    This tutorial requires PyTorch 2.7.0 or later.
+#    This recipe describes a prototype feature. Prototype features are typically
+#    at an early stage for feedback and testing and are subject to change.
+#
+# Prerequisites
+# -------------
+#
+# * PyTorch v2.7.0 or later
+#
 
 #####################################################################
 # Model Setup