Merge branch 'main' into redirect-tts

Author: Svetlana Karslioglu

beginner_source/basics/autogradqs_tutorial.py

@@ -130,9 +130,7 @@
 
 ######################################################################
 # There are reasons you might want to disable gradient tracking:
-#   - To mark some parameters in your neural network as **frozen parameters**. This is
-#     a very common scenario for
-#     `finetuning a pretrained network <https://pytorch.org/tutorials/beginner/finetuning_torchvision_models_tutorial.html>`__
+#   - To mark some parameters in your neural network as **frozen parameters**.
 #   - To **speed up computations** when you are only doing forward pass, because computations on tensors that do
 #     not track gradients would be more efficient.
 

beginner_source/blitz/autograd_tutorial.py

@@ -276,9 +276,6 @@
 # It is useful to "freeze" part of your model if you know in advance that you won't need the gradients of those parameters
 # (this offers some performance benefits by reducing autograd computations).
 #
-# Another common usecase where exclusion from the DAG is important is for
-# `finetuning a pretrained network <https://pytorch.org/tutorials/beginner/finetuning_torchvision_models_tutorial.html>`__
-#
 # In finetuning, we freeze most of the model and typically only modify the classifier layers to make predictions on new labels.
 # Let's walk through a small example to demonstrate this. As before, we load a pretrained resnet18 model, and freeze all the parameters.
 

beginner_source/ddp_series_multigpu.rst

@@ -177,8 +177,8 @@ Running the distributed training job
    +  ddp_setup(rank, world_size)
       dataset, model, optimizer = load_train_objs()
       train_data = prepare_dataloader(dataset, batch_size=32)
-   -  trainer = Trainer(model, dataset, optimizer, device, save_every)
-   +  trainer = Trainer(model, dataset, optimizer, rank, save_every)
+   -  trainer = Trainer(model, train_data, optimizer, device, save_every)
+   +  trainer = Trainer(model, train_data, optimizer, rank, save_every)
       trainer.train(total_epochs)
    +  destroy_process_group()