Merge branch 'main' into fix/ppo-mujoco-colab

Author: abhitorch81

docathon-leaderboard.md

@@ -60,7 +60,7 @@ were not merged or issues that have been closed without a merged PR.
 | kiszk | 20 | https://github.com/pytorch/pytorch/pull/128337, https://github.com/pytorch/pytorch/pull/128123, https://github.com/pytorch/pytorch/pull/128022, https://github.com/pytorch/pytorch/pull/128312 |
 | loganthomas | 19 | https://github.com/pytorch/pytorch/pull/128676, https://github.com/pytorch/pytorch/pull/128192, https://github.com/pytorch/pytorch/pull/128189, https://github.com/pytorch/tutorials/pull/2922, https://github.com/pytorch/tutorials/pull/2910, https://github.com/pytorch/xla/pull/7195 |
 | ignaciobartol | 17 | https://github.com/pytorch/pytorch/pull/128741, https://github.com/pytorch/pytorch/pull/128135, https://github.com/pytorch/pytorch/pull/127938, https://github.com/pytorch/tutorials/pull/2936 |
-| arunppsg | 17 | https://github.com/pytorch/pytorch/pull/128391, https://github.com/pytorch/pytorch/pull/128021, https://github.com/pytorch/pytorch/pull/128018, https://github.com/pytorch-labs/torchfix/pull/59 |
+| arunppsg | 17 | https://github.com/pytorch/pytorch/pull/128391, https://github.com/pytorch/pytorch/pull/128021, https://github.com/pytorch/pytorch/pull/128018, https://github.com/meta-pytorch/torchfix/pull/59 |
 | alperenunlu | 17 | https://github.com/pytorch/tutorials/pull/2934, https://github.com/pytorch/tutorials/pull/2909, https://github.com/pytorch/pytorch/pull/104043 |
 | anandptl84 | 10 | https://github.com/pytorch/pytorch/pull/128196, https://github.com/pytorch/pytorch/pull/128098 |
 | GdoongMathew | 10 | https://github.com/pytorch/pytorch/pull/128136, https://github.com/pytorch/pytorch/pull/128051 |

index.rst

@@ -700,14 +700,14 @@ Welcome to PyTorch Tutorials
    :header: Building an ExecuTorch iOS Demo App
    :card_description: Explore how to set up the ExecuTorch iOS Demo App, which uses the MobileNet v3 model to process live camera images leveraging three different backends: XNNPACK, Core ML, and Metal Performance Shaders (MPS).
    :image: _static/img/ExecuTorch-Logo-cropped.svg
-   :link: https://github.com/pytorch-labs/executorch-examples/tree/main/mv3/apple/ExecuTorchDemo
+   :link: https://github.com/meta-pytorch/executorch-examples/tree/main/mv3/apple/ExecuTorchDemo
    :tags: Edge
 
 .. customcarditem::
    :header: Building an ExecuTorch Android Demo App
    :card_description: Learn how to set up the ExecuTorch Android Demo App for image segmentation tasks using the DeepLab v3 model and XNNPACK FP32 backend.
    :image: _static/img/ExecuTorch-Logo-cropped.svg
-   :link: https://github.com/pytorch-labs/executorch-examples/tree/main/dl3/android/DeepLabV3Demo#executorch-android-demo-app
+   :link: https://github.com/meta-pytorch/executorch-examples/tree/main/dl3/android/DeepLabV3Demo#executorch-android-demo-app
    :tags: Edge
 
 .. customcarditem::

intermediate_source/transformer_building_blocks.py

@@ -62,7 +62,7 @@
 
 If you are only interested in performant attention score modifications, please
 check out the `FlexAttention blog <https://pytorch.org/blog/flexattention/>`_ that
-contains a `gym of masks <https://github.com/pytorch-labs/attention-gym>`_.
+contains a `gym of masks <https://github.com/meta-pytorch/attention-gym>`_.
 
 """
 
@@ -71,7 +71,7 @@
 # ===============================
 # First, we will briefly introduce the four technologies mentioned in the introduction
 #
-# * `torch.nested <https://pytorch.org/tutorials/prototype/nestedtensor.html>`_
+# * `torch.nested <https://pytorch.org/tutorials/unstable/nestedtensor.html>`_
 #
 # Nested tensors generalize the shape of regular dense tensors, allowing for
 # representation of ragged-sized data with the same tensor UX. In the context of
@@ -157,7 +157,7 @@
 #   skipped, performance and memory usage improve.
 #
 # We'll demonstrate the above by building upon the ``MultiheadAttention`` layer in the
-# `Nested Tensor tutorial <https://pytorch.org/tutorials/prototype/nestedtensor.html>`_
+# `Nested Tensor tutorial <https://pytorch.org/tutorials/unstable/nestedtensor.html>`_
 # and comparing it to the ``nn.MultiheadAttention`` layer.
 
 import torch
@@ -675,7 +675,7 @@ def benchmark(func, *args, **kwargs):
 # of the ``MultiheadAttention`` layer that allows for arbitrary modifications
 # to the attention score. The example below takes the ``alibi_mod``
 # that implements `ALiBi <https://arxiv.org/abs/2108.12409>`_ from
-# `attention gym <https://github.com/pytorch-labs/attention-gym>`_ and uses it
+# `attention gym <https://github.com/meta-pytorch/attention-gym>`_ and uses it
 # with nested input tensors.
 
 from torch.nn.attention.flex_attention import flex_attention
@@ -892,8 +892,8 @@ def forward(self, x):
 # etc. Further, there are several good examples of using various performant building blocks to
 # implement various transformer architectures. Some examples include
 #
-# * `gpt-fast <https://github.com/pytorch-labs/gpt-fast>`_
-# * `segment-anything-fast <https://github.com/pytorch-labs/segment-anything-fast>`_
+# * `gpt-fast <https://github.com/meta-pytorch/gpt-fast>`_
+# * `segment-anything-fast <https://github.com/meta-pytorch/segment-anything-fast>`_
 # * `lucidrains implementation of NaViT with nested tensors <https://github.com/lucidrains/vit-pytorch/blob/73199ab486e0fad9eced2e3350a11681db08b61b/vit_pytorch/na_vit_nested_tensor.py>`_
 # * `torchtune's implementation of VisionTransformer <https://github.com/pytorch/torchtune/blob/a8a64ec6a99a6ea2be4fdaf0cd5797b03a2567cf/torchtune/modules/vision_transformer.py#L16>`_
 

unstable_source/gpu_quantization_torchao_tutorial.py

@@ -7,7 +7,7 @@
 In this tutorial, we will walk you through the quantization and optimization
 of the popular `segment anything model <https://github.com/facebookresearch/segment-anything>`_. These
 steps will mimic some of those taken to develop the
-`segment-anything-fast <https://github.com/pytorch-labs/segment-anything-fast/blob/main/segment_anything_fast/modeling/image_encoder.py#L15>`_
+`segment-anything-fast <https://github.com/meta-pytorch/segment-anything-fast/blob/main/segment_anything_fast/modeling/image_encoder.py#L15>`_
 repo. This step-by-step guide demonstrates how you can
 apply these techniques to speed up your own models, especially those
 that use transformers. To that end, we will focus on widely applicable