Merge branch 'main' into add-issue-template-2

Author: Svetlana Karslioglu

.jenkins/validate_tutorials_built.py

@@ -14,7 +14,6 @@
     "profiler",
     "saving_loading_models",
     "introyt/captumyt",
-    "introyt/trainingyt",
     "examples_nn/polynomial_module",
     "examples_nn/dynamic_net",
     "examples_nn/polynomial_optim",

beginner_source/introyt/captumyt.py

@@ -106,13 +106,13 @@
 To install Captum in an Anaconda or pip virtual environment, use the
 appropriate command for your environment below:
 
-With ``conda``:
+With ``conda``::
 
-``conda install pytorch torchvision captum -c pytorch``
+    conda install pytorch torchvision captum -c pytorch
 
-With ``pip``:
+With ``pip``::
 
-``pip install torch torchvision captum``
+    pip install torch torchvision captum
 
 Restart this notebook in the environment you set up, and you’re ready to
 go!

beginner_source/introyt/tensorboardyt_tutorial.py

@@ -24,14 +24,14 @@
 To run this tutorial, you’ll need to install PyTorch, TorchVision,
 Matplotlib, and TensorBoard.
 
-With ``conda``:
+With ``conda``::
 
-``conda install pytorch torchvision -c pytorch``
-``conda install matplotlib tensorboard``
+    conda install pytorch torchvision -c pytorch
+    conda install matplotlib tensorboard
 
-With ``pip``:
+With ``pip``::
 
-``pip install torch torchvision matplotlib tensorboard``
+    pip install torch torchvision matplotlib tensorboard
 
 Once the dependencies are installed, restart this notebook in the Python
 environment where you installed them.

beginner_source/introyt/trainingyt.py

@@ -81,8 +81,8 @@
 validation_set = torchvision.datasets.FashionMNIST('./data', train=False, transform=transform, download=True)
 
 # Create data loaders for our datasets; shuffle for training, not for validation
-training_loader = torch.utils.data.DataLoader(training_set, batch_size=4, shuffle=True, num_workers=2)
-validation_loader = torch.utils.data.DataLoader(validation_set, batch_size=4, shuffle=False, num_workers=2)
+training_loader = torch.utils.data.DataLoader(training_set, batch_size=4, shuffle=True)
+validation_loader = torch.utils.data.DataLoader(validation_set, batch_size=4, shuffle=False)
 
 # Class labels
 classes = ('T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',

recipes_source/recipes/amp_recipe.py

@@ -11,15 +11,15 @@
 range of ``float32``.  Mixed precision tries to match each op to its appropriate datatype,
 which can reduce your network's runtime and memory footprint.
 
-Ordinarily, "automatic mixed precision training" uses `torch.autocast <https://pytorch.org/docs/stable/amp.html#torch.cuda.amp.autocast>`_ and
+Ordinarily, "automatic mixed precision training" uses `torch.autocast <https://pytorch.org/docs/stable/amp.html#torch.autocast>`_ and
 `torch.cuda.amp.GradScaler <https://pytorch.org/docs/stable/amp.html#torch.cuda.amp.GradScaler>`_ together.
 
 This recipe measures the performance of a simple network in default precision,
 then walks through adding ``autocast`` and ``GradScaler`` to run the same network in
 mixed precision with improved performance.
 
 You may download and run this recipe as a standalone Python script.
-The only requirements are Pytorch 1.6+ and a CUDA-capable GPU.
+The only requirements are PyTorch 1.6 or later and a CUDA-capable GPU.
 
 Mixed precision primarily benefits Tensor Core-enabled architectures (Volta, Turing, Ampere).
 This recipe should show significant (2-3X) speedup on those architectures.
@@ -105,7 +105,7 @@ def make_model(in_size, out_size, num_layers):
 ##########################################################
 # Adding autocast
 # ---------------
-# Instances of `torch.cuda.amp.autocast <https://pytorch.org/docs/stable/amp.html#autocasting>`_
+# Instances of `torch.autocast <https://pytorch.org/docs/stable/amp.html#autocasting>`_
 # serve as context managers that allow regions of your script to run in mixed precision.
 #
 # In these regions, CUDA ops run in a dtype chosen by autocast
@@ -310,7 +310,7 @@ def make_model(in_size, out_size, num_layers):
 # 1. Disable ``autocast`` or ``GradScaler`` individually (by passing ``enabled=False`` to their constructor) and see if infs/NaNs persist.
 # 2. If you suspect part of your network (e.g., a complicated loss function) overflows , run that forward region in ``float32``
 #    and see if infs/NaNs persist.
-#    `The autocast docstring <https://pytorch.org/docs/stable/amp.html#torch.cuda.amp.autocast>`_'s last code snippet
+#    `The autocast docstring <https://pytorch.org/docs/stable/amp.html#torch.autocast>`_'s last code snippet
 #    shows forcing a subregion to run in ``float32`` (by locally disabling autocast and casting the subregion's inputs).
 #
 # Type mismatch error (may manifest as CUDNN_STATUS_BAD_PARAM)