03b_tensorflow_preprocessing.ipynb failed CI

Link to the notebook:
https://github.com/aws/amazon-sagemaker-examples/blob/master/prep_data/image_data_guide/03b_tensorflow_preprocessing.ipynb

Error:

---------------------------------------------------------------------------
Exception encountered at "In [4]":
---------------------------------------------------------------------------
NotFoundError Traceback (most recent call last)
in
----> 1 image_folder = tfds.ImageFolder('./data_structured')

/usr/local/lib/python3.7/site-packages/tensorflow_datasets/core/folder_dataset/image_folder.py in __init__(self, root_dir, shape, dtype)
95 # Extract the splits, examples, labels
96 root_dir = os.path.expanduser(root_dir)
---> 97 self._split_examples, labels = _get_split_label_images(root_dir)
98
99 # Update DatasetInfo labels

/usr/local/lib/python3.7/site-packages/tensorflow_datasets/core/folder_dataset/image_folder.py in _get_split_label_images(root_dir)
203 split_examples = collections.defaultdict(list)
204 labels = set()
--> 205

[...]

NotFoundError: Could not find directory ./data_structured

Open prep_data/image_data_guide/03b_tensorflow_preprocessing.ipynb and inspect cell In [4], where TensorFlow Datasets cannot find ./data_structured. Trace the notebook's data preparation steps and determine what must be present for that path; done means the notebook no longer raises this error in CI.

Written by the indexing model from the issue text.