Merge branch 'master' into distr-tutorial-landing

Author: Svetlana Karslioglu

.jenkins/validate_tutorials_built.py

@@ -25,7 +25,6 @@
     "forward_ad_usage",
     "parametrizations",
     "reinforcement_q_learning",
-    "text_to_speech_with_torchaudio",
     "mnist_train_nas",  # used by ax_multiobjective_nas_tutorial.py
     "fx_conv_bn_fuser",
     "super_resolution_with_onnxruntime",

beginner_source/basics/optimization_tutorial.py

@@ -28,7 +28,7 @@
 from torch import nn
 from torch.utils.data import DataLoader
 from torchvision import datasets
-from torchvision.transforms import ToTensor, Lambda
+from torchvision.transforms import ToTensor
 
 training_data = datasets.FashionMNIST(
     root="data",

beginner_source/ddp_series_multigpu.rst

@@ -123,6 +123,18 @@ Distributing input data
    +   sampler=DistributedSampler(train_dataset),
    )
 
+-  Calling the ``set_epoch()`` method on the ``DistributedSampler`` at the beginning of each epoch is necessary to make shuffling work 
+   properly across multiple epochs. Otherwise, the same ordering will be used in each epoch.
+
+.. code:: diff
+
+   def _run_epoch(self, epoch):
+       b_sz = len(next(iter(self.train_data))[0])
+   +   self.train_data.sampler.set_epoch(epoch)
+       for source, targets in self.train_data:
+         ...
+         self._run_batch(source, targets)
+
 
 Saving model checkpoints
 ~~~~~~~~~~~~~~~~~~~~~~~~

beginner_source/introyt/autogradyt_tutorial.py

@@ -40,7 +40,7 @@
 
 ###########################################################################
 # A machine learning model is a *function*, with inputs and outputs. For
-# this discussion, we’ll treat the inputs a as an *i*-dimensional vector
+# this discussion, we’ll treat the inputs as an *i*-dimensional vector
 # :math:`\vec{x}`, with elements :math:`x_{i}`. We can then express the
 # model, *M*, as a vector-valued function of the input: :math:`\vec{y} =
 # \vec{M}(\vec{x})`. (We treat the value of M’s output as
@@ -226,7 +226,7 @@
 # of which should be :math:`2 * cos(a)`. Looking at the graph above,
 # that’s just what we see.
 # 
-# Be aware than only *leaf nodes* of the computation have their gradients
+# Be aware that only *leaf nodes* of the computation have their gradients
 # computed. If you tried, for example, ``print(c.grad)`` you’d get back
 # ``None``. In this simple example, only the input is a leaf node, so only
 # it has gradients computed.

conf.py

@@ -89,6 +89,7 @@
     'gallery_dirs': ['beginner', 'intermediate', 'advanced', 'recipes', 'prototype'],
     'filename_pattern': '.py',
     'ignore_pattern': re.compile(f"({'|'.join(NOT_RUN)}).py$"),
+    'promote_jupyter_magic': True,
     'backreferences_dir': None
 }
 
@@ -140,9 +141,9 @@
 # built documents.
 #
 # The short X.Y version.
-version = torch.__version__
+version = str(torch.__version__)
 # The full version, including alpha/beta/rc tags.
-release = torch.__version__
+release = str(torch.__version__)
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

intermediate_source/text_to_speech_with_torchaudio.py

@@ -7,9 +7,6 @@
 
 """
 
-# %matplotlib inline
-
-
 ######################################################################
 # Overview
 # --------
@@ -52,8 +49,11 @@
 # encoding.
 # 
 
-# When running this example in notebook, install DeepPhonemizer
-# !pip3 install deep_phonemizer
+# %%
+#  .. code-block:: bash
+#
+#      %%bash
+#      pip3 install deep_phonemizer
 
 import torch
 import torchaudio

requirements.txt

@@ -27,6 +27,7 @@ pytorch-lightning
 torchx
 ax-platform
 nbformat>=4.2.0
+deep_phonemizer==0.0.17
 
 # PyTorch Theme
 -e git+https://github.com/pytorch/pytorch_sphinx_theme.git#egg=pytorch_sphinx_theme