training_pipeline_pytorch_mnist.ipynb failed CI

Link to the notebook:
https://github.com/aws/amazon-sagemaker-examples/blob/master/step-functions-data-science-sdk/training_pipeline_pytorch_mnist/training_pipeline_pytorch_mnist.ipynb

Error:

---------------------------------------------------------------------------
Exception encountered at "In [3]":
---------------------------------------------------------------------------
HTTPError Traceback (most recent call last)
in
3 datasets.MNIST('data', download=True, transform=transforms.Compose([
4 transforms.ToTensor(),
----> 5 transforms.Normalize((0.1307,), (0.3081,))
6 ]))

/opt/conda/lib/python3.6/site-packages/torchvision/datasets/mnist.py in __init__(self, root, train, transform, target_transform, download)
77
78 if download:
---> 79 self.download()
80
81 if not self._check_exists():

/opt/conda/lib/python3.6/site-packages/torchvision/datasets/mnist.py in download(self)
144 for url, md5 in self.resources:
145 filename = url.rpartition('/')[2]
--> 146 down

[...]

HTTPError: HTTP Error 503: Service Unavailable

Start with step-functions-data-science-sdk/training_pipeline_pytorch_mnist/training_pipeline_pytorch_mnist.ipynb and inspect cell In [3], where the PyTorch MNIST dataset download fails with HTTP 503. Re-run the notebook or its CI check to confirm the failure and determine a reliable resolution; done means the notebook completes successfully in CI.

Written by the indexing model from the issue text.