pytorch-gcn-tox21-hypertune.ipynb failed CI

Link to the notebook:
https://github.com/aws/amazon-sagemaker-examples/blob/master/sagemaker-python-sdk/dgl_gcn_tox21/pytorch-gcn-tox21-hypertune.ipynb

Error:

---------------------------------------------------------------------------
Exception encountered at "In [7]":
---------------------------------------------------------------------------
ClientError Traceback (most recent call last)
in
----> 1 tuner.fit(inputs={'training-code': code_location})

/opt/conda/lib/python3.6/site-packages/sagemaker/tuner.py in fit(self, inputs, job_name, include_cls_metadata, estimator_kwarg

[...]

ClientError: An error occurred (ValidationException) when calling the CreateHyperParameterTuningJob operation: Cannot find the requested image: 521695447989.dkr.ecr.us-west-2.amazonaws.com/sagemaker-dgl-pytorch-gcn-tox21:latest with tag: latest and digest: null. Please check if your ECR image exists and role arn:aws:iam::521695447989:role/ProdBuildSystemStack-ReleaseBuildRoleFB326D49-QK8LUA2UI1IC has proper pull permissions for SageMaker: ecr:BatchCheckLayerAvailability, ecr:BatchGetImage, ecr:GetDownloadUrlForLayer

Start with sagemaker-python-sdk/dgl_gcn_tox21/pytorch-gcn-tox21-hypertune.ipynb and inspect cell In [7], where tuner.fit fails. Investigate why the SageMaker hyperparameter tuning job cannot find the referenced ECR image, then rerun the notebook's CI check; done means the notebook completes without the ValidationException.

Written by the indexing model from the issue text.