chore: update submodules (#187)

Co-authored-by: ydcjeff <[email protected]>

src/tutorials/intermediate/01-cifar10-distributed.md

@@ -31,7 +31,7 @@ The type of distributed training we will use is called data parallelism in which
 >
 > -- <cite>[Distributed Deep Learning 101: Introduction](https://towardsdatascience.com/distributed-deep-learning-101-introduction-ebfc1bcd59d9)</cite>
 
-PyTorch provides a [torch.nn.parallel.DistributedDataParallel](https://pytorch.org/docs/stable/generated/torch.nn.parallel.DistributedDataParallel.html) API for this task however the implementation that supports different backends + configurations is tedious. In this example, we will see how to can enable data distributed training which is adaptable to various backends in just a few lines of code alongwith:
+PyTorch provides a [torch.nn.parallel.DistributedDataParallel](https://pytorch.org/docs/stable/generated/torch.nn.parallel.DistributedDataParallel.html) API for this task however the implementation that supports different backends + configurations is tedious. In this example, we will see how to enable data distributed training which is adaptable to various backends in just a few lines of code alongwith:
 * Computing training and validation metrics
 * Setup logging (and connecting with ClearML)
 * Saving the best model weights
@@ -229,7 +229,7 @@ def get_model(config):
 
 ### Optimizer
 
-Then we can setup the optimizer using hyperameters from `config` and pass it through [`auto_optim()`](https://pytorch.org/ignite/generated/ignite.distributed.auto.auto_optim.html#ignite.distributed.auto.auto_optim).
+Then we can setup the optimizer using hyperparameters from `config` and pass it through [`auto_optim()`](https://pytorch.org/ignite/generated/ignite.distributed.auto.auto_optim.html#ignite.distributed.auto.auto_optim).
 
 
 ```python
@@ -651,7 +651,7 @@ torchrun --nproc_per_node=2 main.py run --backend="nccl"
 
 
 
-### Run with internal spawining (`torch.multiprocessing.spawn`)
+### Run with internal spawning (`torch.multiprocessing.spawn`)
 
 ```
 python -u main.py run --backend="nccl" --nproc_per_node=2

src/tutorials/intermediate/02-Machine_Translation_using_PyTorch_Ignite.md

@@ -40,7 +40,7 @@ This notebook uses Models, Dataset and Tokenizers from Huggingface, hence they c
 ## Common Configuration
 We maintain a config dictionary which can be extended or changed to store parameters required during training. We can refer back to this code when we will use these parameters later.
 
-In this example we are using ``t5-small``, which has 60M parameters. The way t5 models work is they taske an input with the a task-specific prefix. This prefix (like "Translate English to German") will let our model know which task it needs to perform. For more details refer to the original paper [here](https://arxiv.org/abs/1910.10683).
+In this example we are using ``t5-small``, which has 60M parameters. The way t5 models work is they take an input with a task-specific prefix. This prefix (like "Translate English to German") will let our model know which task it needs to perform. For more details refer to the original paper [here](https://arxiv.org/abs/1910.10683).
 
 
 Here we train on less number of iterations per step and on a limited dataset, this can be modified using the ``train_dataset_length`` and ``epoch_length`` config.
@@ -254,7 +254,7 @@ The forward pass is wrapped in the autocast context manager for mixed precision
 Gradient accumulation is implemented as batch size of 1 would lead to noisy updates otherwise. Check the ``accumulation_steps`` variable in config to define the number of steps to accumulate the gradient. 
 
 #### Trainer Handlers
-Handlers can be defined and attached directly to the trainer engine. Here we also make use of a special function : `setup_common_training_handlers` which has a lot of the commonly used, useful handlers (like `save_every_iters`, `clear_cuda_cache` etc) already defined. To know more about this function, refer to the docs [here](https://pytorch.org/ignite/contrib/engines.html#ignite.contrib.engines.common.setup_common_training_handlers). 
+Handlers can be defined and attached directly to the trainer engine. Here we also make use of a special function : `setup_common_training_handlers` which has a lot of the commonly used, useful handlers (like `save_every_iters`, `clear_cuda_cache`, etc) already defined. To know more about this function, refer to the docs [here](https://pytorch.org/ignite/contrib/engines.html#ignite.contrib.engines.common.setup_common_training_handlers). 
 
 
 ```python